JP2003266352A - Robot device and control method therefor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a robot device provided with a command management means capable of dividing and integrating resources, preventing disturbance of sequence due to twice hitting of a cancel command, and complementing parameter and command name when handling a metacommand. <P>SOLUTION: A command manager corresponds to a plurality of clients from, for example, a situation depending action hierarchy (SBL) of a high-order layer and a client from a reflex action part (Reflexive SBL). Moreover, the command manager corresponds to a plurality of agents of a low-order layer, Agent 1, Agent 2, and Agent 3. A plurality of these agents in the low-order layer understand commands from the clients in accordance with scenes and execute processing based on independent discrimination for resources related to devices required for the operation of the robot device such as a body trunk unit, a head part unit, an arm part unit, a leg part unit, etc. Each agent, Agent 1, Agent 2, and Agent 3, executes processing for a plurality of resources, respectively. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a robot apparatus and a control method of the robot apparatus which realize autonomous communication and realize realistic communication, and more particularly,
The present invention relates to an autonomous robot apparatus having a function of recognizing external information such as images and sounds and reflecting its own behavior on the information and a control method of the robot apparatus.

[0002]

2. Description of the Related Art A mechanical device that makes a motion similar to a human motion by using an electrical or magnetic action is called a "robot". The origin of the robot is the Slavic word "ROBO".
It is said that it is derived from "TA (slave machine)." In Japan, robots began to be popular since the end of the 1960s, but most of them are automation of production work in factories.
It was an industrial robot such as a manipulator and a transfer robot for the purpose of unmanned operation.

Recently, a pet type robot imitating the body mechanism of a four-legged animal such as dogs, cats, and bears and its movements, or the body mechanism of an animal that performs two-legged upright walking, such as a human or a monkey, has been used. "Humanoid" that imitates movement
Alternatively, research and development on the structure of legged mobile robots such as "humanoid robots" and their stable walking control have progressed, and expectations for their practical application are increasing. These legged mobile robots are more unstable than crawler robots, making posture control and walking control difficult, but they are superior in that they can realize flexible walking and running operations such as climbing stairs and climbing over obstacles. There is.

Stationary type robots, such as arm type robots, which are implanted in a specific place and used,
Work only in fixed and local work spaces such as assembly and selection of parts. On the other hand, a mobile robot has a non-limitative work space, and can freely move on a predetermined route or on a non-route to perform a predetermined or arbitrary human work on behalf of a human or a dog. Alternatively, various services that replace other life forms can be provided.

One of the uses of legged mobile robots is to perform various difficult tasks on behalf of industrial activities and production activities. For example, maintenance work in nuclear power plants, thermal power plants, and petrochemical plants, parts transportation / assembly work in manufacturing plants, cleaning in high-rise buildings, agency of dangerous work / difficult work such as rescue at fire sites and the like. .

Further, as another application of the legged mobile robot, it is more closely related to life, that is, "symbiosis" or "entertainment" with humans, rather than the above-mentioned work support.
There are uses. This kind of robot faithfully reproduces the motion mechanism of humans, dogs (pets), bears and other relatively intelligent legged walking animals and the rich emotional expression using the limbs. In addition to simply faithfully executing the pre-entered motion pattern, it also dynamically responds to the words and attitudes received from the user (or other robot) (such as "praise", "scrib", "hit", etc.). It is also required to realize corresponding and lively response expressions.

In the conventional toy machine, the relationship between the user operation and the response motion is fixed, and the motion of the toy cannot be changed according to the user's preference. As a result, the user eventually gets tired of the toy that repeats only the same operation.

On the other hand, an intelligent robot that autonomously operates generally has a function of recognizing external information and reflecting its own behavior on it. That is, the robot realizes autonomous thinking and motion control by changing the emotion model or the instinct model based on input information such as voice, image, and tactile sense from the external environment to determine the motion. That is, by providing the emotion model and the instinct model, the robot can realize realistic communication with humans at a higher intellectual level.

In order for a robot to perform an autonomous operation according to a change in environment, conventionally, an action is described by a combination of simple action descriptions such as receiving an action for one observation result and taking an action. It was By mapping the actions to these inputs, it is possible to express complex actions that are not unique by introducing functions such as randomness, internal state (emotion / instinct), learning, and growth.

However, in such a simple action mapping method, if the player loses sight of the ball even for a moment during the action of chasing and kicking the ball, the information regarding the ball is not retained, and the action of searching for the ball from the beginning is performed. Therefore, the use of recognition results is inconsistent.

Further, the robot needs to solve a correspondence problem of which recognition result corresponds to which previously observed result when a plurality of balls are observed. Also,
In order to use more complex recognition results such as face recognition, skin color recognition, voice recognition, etc., which skin color area corresponds to which person in the face, and which person this voice is Need to solve.

[0012]

By the way, in the above-mentioned autonomous robot apparatus, resource management has not been carried out so much.

For example, since one resource is assigned to one agent, it is not necessary to divide and integrate the resources. Also, the cancel command is transparent to the agent, and there is a possibility of disturbing the sequence such as hitting twice. Also, when dealing with metacommands, we did not complete parameters and command names. In addition, the agent did not have a route through which quick feedback information could be sent. Also, only one client is allowed, and command control is not performed for multiple clients. At the same time, there is no command response control for multiple clients.

The present invention has been made in view of the above-mentioned circumstances, and it divides and integrates resources, prevents the sequence from being disturbed due to double hitting of a cancel command, and handles meta commands. An object of the present invention is to provide a robot apparatus and a method for controlling the robot apparatus, which are capable of complementing parameters and command names.

Further, the present invention has a route through which fast feedback information etc. on the agent side can flow.
Further, another object of the present invention is to provide a robot apparatus and a robot apparatus control method capable of controlling commands to a plurality of clients and simultaneously supporting a command response control to a plurality of clients.

[0016]

In order to solve the above problems, a robot apparatus according to the present invention is an autonomous robot that acts autonomously based on a recognition result of an internal state based on an emotional model or an external environment. In the apparatus, a client unit that is an action generation unit that generates the autonomous action, a command management unit that manages a command related to the action generated by the client unit, and a command that is managed by the command management unit are received. The command management means divides the resource into basic resources at the time of input and then outputs the resource at the time of output when a plurality of resources are specified in the command from the client means. To unite.

As a result, it is not necessary to limit the number of basic resources possessed by the agent. In addition, command resources can now be properly managed even by freely setting them.

In order to solve the above-mentioned problems, the robot apparatus according to the present invention is an autonomous robot apparatus which acts autonomously based on a recognition result for an internal state or an external environment based on an emotion model, Client means that is a behavior generation unit that generates a behavior, command management means that manages a command related to the behavior that the client means generates, and a resource that receives the command managed by the command management means and causes the behavior And a command managing means, the command managing means sends a cancel command for the command from the client means, and further receives a cancel command from the corresponding resource for the cancel command from the client means for the same resource. Cancel command of When the reply came that, at the same time to make a reply, sent to the appropriate client.

As a result, there is no cancellation that disturbs the useless status, and it is possible to operate with only necessary and sufficient commands.

In order to solve the above-mentioned problems, a robot apparatus according to the present invention is an autonomous robot apparatus which acts autonomously based on a recognition result of an internal state based on an emotion model or an external environment, Client means that is a behavior generation unit that generates a behavior, command management means that manages a command related to the behavior that the client means generates, and a resource that receives the command managed by the command management means and causes the behavior When the command management means receives from the client means a meta-command regarding parameter change of a command already sent, the command management means converts the command into a corresponding command and causes the agent means to send.

As a result, it becomes possible to take out necessary information using the metacommand and send it to the agent. In addition, by supplementing the response from the agent and creating a correct response, the response from the metacommand can now be received correctly.

In order to solve the above-mentioned problems, a robot apparatus according to the present invention is an autonomous robot apparatus which acts autonomously based on a recognition result of an internal state based on an emotion model or an external environment, Client means that is a behavior generation unit that generates a behavior, command management means that manages a command related to the behavior that the client means generates, and a resource that receives the command managed by the command management means and causes the behavior When the information about the change of the state due to the driving of the resource is sent from the agent means, the command managing means sends the information to the client designated by the agent means.

As a result, by receiving the information from the agent, the Conifguratio of the upper client
n It has become possible to reduce dependence and make more appropriate actions by effectively using information from agents.

In order to solve the above-mentioned problems, a robot apparatus according to the present invention is an autonomous robot apparatus which acts autonomously based on a recognition result of an internal state based on an emotion model and an external environment, A plurality of client units that are action generation units that generate various actions, a command management unit that manages commands related to the actions generated by the plurality of client units, and a command that is managed by the command management unit The command management means arbitrates a plurality of commands sent from the plurality of client means according to the priority added by the plurality of client means. And sends it to the agent means.

As a result, by supporting a plurality of clients, it becomes possible to combine the layer in which the upper client thinks deeply with the reflective layer, and to give the action selecting section a hierarchical structure. Also, since the criteria at that time can be changed dynamically, it is possible to give priority to commands that match the situation more.

[0026]

BEST MODE FOR CARRYING OUT THE INVENTION A bipedal type robot apparatus shown as an example of the configuration of the present invention will be described in detail below with reference to the drawings. This humanoid robot device is a practical robot that supports human activities in various situations in the living environment and other daily life, and can act according to internal conditions (anger, sadness, joy, enjoyment, etc.), It is an entertainment robot that can express the basic actions that humans perform. That is, the behavior control can be autonomously performed based on the recognition result of the external stimulus such as the voice or the image.

As shown in FIG. 1, in the robot apparatus 1, a head unit 3 is connected to a predetermined position of a torso unit 2, two left and right arm units 4R / 4L, and two left and right legs. The units 5R / 5L are connected to each other (however, each of R and L is a suffix indicating each of right and left. The same applies hereinafter).

FIG. 2 schematically shows the joint degree of freedom structure of the robot apparatus 1. The neck joint supporting the head unit 3 includes a neck joint yaw axis 101 and a neck joint pitch axis 10
It has two degrees of freedom, namely 2 and the neck joint roll shaft 103.

Each arm unit 4R / 4L constituting the upper limb has a shoulder joint pitch axis 107, a shoulder joint roll axis 108, an upper arm yaw axis 109, and an elbow joint pitch axis 1.
10, forearm yaw axis 111, wrist joint pitch axis 112
And a wrist joint roll wheel 113 and a hand portion 114. The hand portion 114 is actually a multi-joint / multi-degree-of-freedom structure including a plurality of fingers. However, since the motion of the hand portion 114 has little contribution or influence to the posture control and the walking control of the robot apparatus 1, it is assumed that the degree of freedom is zero in this specification. Therefore, each arm has seven degrees of freedom.

Further, the trunk unit 2 has three degrees of freedom: a trunk pitch axis 104, a trunk roll axis 105, and a trunk yaw axis 106.

Each leg unit 5R / 5L constituting the lower limb has a hip joint yaw shaft 115, a hip joint pitch shaft 116, a hip joint roll shaft 117, and a knee joint pitch shaft 1.
18, an ankle joint pitch shaft 119, an ankle joint roll shaft 120, and a foot portion 121. In this specification, the intersection of the hip joint pitch axis 116 and the hip joint roll axis 117 defines the hip joint position of the robot apparatus 1. The foot 121 of the human body is actually a structure including a multi-joint, multi-degree-of-freedom foot, but the foot of the robot apparatus 1 has zero degrees of freedom. Therefore, each leg has 6 degrees of freedom.

In summary, the robot apparatus 1 as a whole has 3 + 7 × 2 + 3 + 6 × 2 = 32 degrees of freedom. However, the robot device 1 for entertainment is not necessarily limited to 32 degrees of freedom. It goes without saying that the degree of freedom, that is, the number of joints, can be appropriately increased or decreased in accordance with design / production constraint conditions and required specifications.

Each degree of freedom of the robot apparatus 1 as described above is actually implemented by using an actuator.
It is preferable that the actuator be small and lightweight in view of demands such as eliminating extra bulges in appearance and approximating the shape of a natural human body, and performing posture control for an unstable structure such as bipedal walking. .

FIG. 3 schematically shows the control system configuration of the robot apparatus 1. As shown in the figure, the robot device 1 includes a trunk unit 2, which represents human limbs.
The head unit 3, the arm units 4R / 4L, the leg units 5R / 5L, and the control unit 10 that performs adaptive control for realizing cooperative operation between the units.

The overall operation of the robot apparatus 1 is controlled by the control unit 10. Control unit 10
Is a CPU (Central Processing Unit) or DRA
An interface (not shown) for exchanging data and commands with the main control unit 11 including main circuit components (not shown) such as M and flash ROM, and the power supply circuit and each component of the robot apparatus 1. And the peripheral circuit 12 including.

In implementing the present invention, the installation place of the control unit 10 is not particularly limited. Although it is mounted on the trunk unit 2 in FIG. 3, it may be mounted on the head unit 3. Alternatively, the control unit 10 may be provided outside the robot apparatus 1 to communicate with the body of the robot apparatus 1 in a wired or wireless manner.

The degree of freedom of each joint in the robot apparatus 1 shown in FIG. 2 is realized by an actuator corresponding to each joint. That is, the head unit 3 includes a neck joint yaw axis 101, a neck joint pitch axis 102, and a neck joint roll axis 10.
3, the neck joint yaw axis actuator A ₂ representing each of
A neck joint pitch axis actuator A ₃ and a neck joint roll axis actuator A ₄ are arranged.

Further, the head unit 3 is provided with a CCD (Charge Coupled Device) camera for picking up an image of an external situation, and a distance sensor for measuring a distance to an object located in front of the head unit 3. A microphone for collecting external sound, a speaker for outputting voice, a touch sensor for detecting the pressure received by the physical action of the user such as "stroking" or "striking" are provided. ing.

The trunk unit 2 has a trunk pitch axis actuator A ₅ , a trunk roll axis actuator A ₆ , which expresses the trunk pitch axis 104, trunk roll axis 105, and trunk yaw axis 106, respectively. A trunk yaw axis actuator A ₇ is provided. Also, in the trunk unit 2,
The robot apparatus 1 is provided with a battery as a power source. This battery is composed of a chargeable / dischargeable battery.

The arm unit 4R / 4L is used for the upper arm unit.
Knit 4₁R / 4₁L and elbow joint unit 4_TwoR / 4_Two
L and forearm unit 4_ThreeR / 4_ThreeIt is subdivided into L,
Shoulder joint pitch axis 107, shoulder joint roll axis 108, upper arm yo
-Axis 109, elbow joint pitch axis 110, forearm yaw axis 11
1, wrist joint pitch axis 112, wrist joint roll axis 113
Of each shoulder joint pitch axis actuator A₈,shoulder
Joint roll axis actuator A₉, Upper arm yaw actu
Eta A₁₀, Elbow joint pitch axis actuator A ₁₁,
Elbow joint roll axis actuator A₁₂, Wrist joint pitch
Axis actuator A_Thirteen, Wrist joint roll axis actuator
Data A₁₄Has been deployed.

The leg unit 5R / 5L includes a thigh unit 5 ₁ R / 5 ₁ L and a knee unit 5 ₂ R / 5 ₂ L.
The hip joint yaw axis 115, the hip joint pitch axis 116, the hip joint roll axis 117, the knee joint pitch axis 118, the ankle joint pitch axis 119, and the ankle joint roll axis are subdivided into the tibial unit 5 ₃ R / 5 ₃ L. A hip joint yaw axis actuator A ₁₆ , a hip joint pitch axis actuator A ₁₇ , a hip joint roll axis actuator A ₁₈ , a knee joint pitch axis actuator A ₁₉ , an ankle joint pitch axis actuator A ₂₀ , and an ankle joint roll axis actuator A ₂₁ that represent each of 120. Has been deployed. The actuators A ₂ , A ₃ ... Used for the respective joints are more preferably composed of small AC servo actuators of the type that are directly connected to the gears and the servo control system is integrated into one chip and mounted in the motor unit. can do.

For each mechanical unit such as the trunk unit 2, the head unit 3, the arm units 4R / 4L, the leg units 5R / 5L, etc., the sub-control units 20, 21, 22R / of the actuator drive control unit are provided. 22L and 23R / 23L are provided. Furthermore, a ground contact confirmation sensor 30R for detecting whether or not the sole of each leg unit 5R / 5L has landed
In addition to wearing / 30L, the trunk unit 2 is equipped with a posture sensor 31 for measuring the posture.

The ground confirmation sensor 30R / 30L is composed of, for example, a proximity sensor or a micro switch installed on the sole of the foot. Further, the posture sensor 31 is composed of, for example, a combination of an acceleration sensor and a gyro sensor.

By the output of the ground contact confirmation sensor 30R / 30L, it is possible to determine whether each of the left and right legs is currently standing or swinging during an operation period such as walking or running. Further, the output of the posture sensor 31 can detect the inclination and posture of the trunk.

The main control unit 11 controls each sensor 30R / 30.
The control target can be dynamically corrected in response to the outputs of L and 31. More specifically, the sub control unit 20,
By performing adaptive control on each of 21, 22R / 22L and 23R / 23L, it is possible to realize a whole-body movement pattern in which the upper limbs, the trunk, and the lower limbs of the robot apparatus 1 are cooperatively driven.

The whole body motion of the robot apparatus 1 on the body is
Foot movement, ZMP (Zero Moment Point) trajectory, trunk movement
Movement, upper limb movement, waist height, etc.
Command to instruct the operation according to the setting contents
Sub control unit 20, 21, 22R / 22L, 23R / 23L
Transfer to. Then, each sub control unit 20, 21, ...
・ Interpret the received command from the main controller 11
Each actuator A _Two, A_Three... Drives for etc.
Output a control signal. "ZMP" here means walking
Of the point on the floor where the moment due to the floor reaction force inside becomes zero
The “ZMP orbit” is, for example, a robot
Means a locus of movement of the ZMP during the walking motion of the device 1.
It In addition, the concept of ZMP and ZMP
Regarding the points applied to the stability criterion, see Miomir Vukob.
ratovic "LEGGED LOCOMOTION ROBOTS" (Ichiro Kato
In "Walking Robot and Artificial Feet" (Nikkan Kogyo Shimbun))
Have been described.

As described above, in the robot apparatus 1, each of the sub-control units 20, 21, ... Interprets the received command from the main control unit 11 and each actuator A ₂ , A ₃
A drive control signal is output to ... to control the drive of each unit. As a result, the robot apparatus 1 makes a stable transition to the target posture and can walk in a stable posture.

Further, in the control unit 10 in the robot apparatus 1, in addition to the posture control as described above, various sensors such as an acceleration sensor, a touch sensor, a ground contact confirmation sensor, image information from a CCD camera, and voice from a microphone. Information is handled in a centralized manner. In the control unit 10, although not shown, various sensors such as an acceleration sensor, a gyro sensor, a touch sensor, a distance sensor, a microphone, a speaker, actuators, CCD cameras, and batteries are connected to the main control unit 11 via corresponding hubs. Has been done.

The main controller 11 sequentially takes in sensor data, image data, and audio data supplied from the above-mentioned sensors, and sequentially stores them in a predetermined position in the DRAM via the internal interface. Further, the main control unit 11 sequentially takes in the battery remaining amount data representing the battery remaining amount supplied from the battery, and stores this in the DRAM.
It is stored in a predetermined position inside. Each sensor data, image data, voice data, and battery remaining amount data stored in the DRAM are used when the main control unit 11 controls the operation of the robot apparatus 1.

The main controller 11 reads the control program and stores it in the DRAM at the initial stage when the power of the robot apparatus 1 is turned on. In addition, the main control unit 11 uses the sensor data, the image data, the audio data, and the battery remaining amount data that are sequentially stored in the DRAM from the main control unit 11 as described above, based on the self and surrounding conditions and the user. Judging whether or not there are instructions and how to work.

Further, the main control section 11 determines an action according to its own situation based on this determination result and the control program stored in the DRAM, and drives a necessary actuator based on the determination result. The robot apparatus 1 is caused to take actions such as so-called “gesture” and “hand gesture”.

In this way, the robot apparatus 1 can perform the action control according to the recognition result of the external stimulus and the change of the internal state based on the control program. FIG. 4 shows an action control system 50 adopted in the robot apparatus 1.
The basic architecture of is schematically shown.

Object-oriented programming can be incorporated into the illustrated behavior control system 50. In this case, each software is handled in module units called "objects" in which data (property) and a processing procedure (method) for the data are integrated. In addition, each object can pass properties and inherit methods by message communication and inter-object communication method using shared memory.

The behavior control system 50 uses the external environment (Envi
ronments), the visual recognition function unit 51,
The hearing recognition function unit 52 and the contact recognition function unit 53 are provided.

The visual recognition function section (Video) 51 is, for example, a charge coupled device (CCD).
Image recognition processing such as face recognition and color recognition and feature extraction are performed based on a captured image input through an image input device such as a camera. The visual recognition function unit 51 uses "MultiColorTracke
It consists of multiple objects such as "r", "FaceDetector", and "FaceIdentify".

The auditory recognition function unit (Audio) 52 performs voice recognition of voice data input through a voice input device such as a microphone to extract a feature or a word set (text).
Do recognition. The auditory recognition function unit 52 is composed of a plurality of objects such as "AudioRecog" and "AuthurDecoder" described later.

The contact recognition function unit (Tactile) 53 recognizes a sensor signal from a contact sensor built in, for example, the head of the machine body, and recognizes an external stimulus such as "stroked" or "struck". To do.

Internal status management unit (ISM: Internal Statu
s Manager) 54 has an instinct model and an emotional model,
The visual recognition function unit 51, the auditory recognition function unit 52,
External stimulus (E recognized by the touch recognition function unit 53
S: External Stimula) to manage internal states such as instinct and emotion of the robot apparatus 1.

The emotion model and the instinct model each have a recognition result and an action history as inputs, and manage emotion values and instinct values. The behavior model can refer to these emotional values and instinct values.

The short-term memory section (ShortTermMemory) 55 is
The visual recognition function unit 51, the auditory recognition function unit 52,
It is a functional module that holds a target or event recognized from the external environment by the contact recognition function unit 53 for a short period of time. For example, the input image from the camera is stored for a short period of about 15 seconds.

The long-term storage unit (LongTermMemory) 56 is used to hold information obtained by learning such as the name of an object for a super-period. The long-term storage unit 56, for example, associatively stores a change in internal state from an external stimulus in a certain action module.

The behavior control of the robot apparatus 1 is realized by the "reflexive behavior" realized by the reflexive behavior unit 59, the "situation dependent behavior" realized by the situation dependent behavior layer 58, and the "thinking behavior layer 57". It is roughly divided into "thinking behavior".

Deliberative Layer 57
Performs a relatively long-term action plan of the robot apparatus 1 based on the stored contents of the short-term storage unit 55 and the long-term storage unit 56.

The deliberative action is an action that is performed by inferring a plan or a plan for realizing it based on a given situation or an instruction from a human. Such inferences and plans are
Since the processing time and the calculation load are longer than the reaction time for maintaining the interaction, the robot apparatus 1 performs the reflexive behavior and the situation-dependent behavior while returning the reaction in real time,
Conduct deliberative actions by reasoning and planning.

Situation-dependent behavior hierarchy (SBL: Situated Beha)
The viorsLayer) 58 controls the action immediately corresponding to the current situation of the robot apparatus 1 based on the stored contents of the short-term storage unit 55 and the long-term storage unit 56 and the internal state managed by the internal state management unit 54. To do.

The state-dependent action hierarchy 58 prepares a state machine for each action, classifies the recognition result of the external information input by the sensor, depending on the action and the situation before that, and classifies the action as a machine. Expressed above. The situation-dependent action hierarchy 58 also realizes an action for keeping the internal state within a certain range (also called "homeostasis action"), and when the internal state exceeds the specified range, the internal state Activate the behavior so that it becomes easier for the behavior to return to within the range (actually, the behavior is selected in consideration of both the internal state and the external environment). Situation-dependent behavior has a slower reaction time than reflexive behavior (described later).

Reflection action layer 57 and situation-dependent action layer 58
Can be implemented as an application.

Reflexive action part (ConfigurationDependentAc
The tionsAndReactions or ReflexiveSBL) 59 is a functional module that realizes reflexive body movements according to the external stimuli recognized by the visual recognition function unit 51, the auditory recognition function unit 52, and the contact recognition function unit 53.

The reflex action is basically an action for directly receiving the recognition result of the external information input by the sensor, classifying the result, and directly determining the output action. For example, behaviors such as chasing a human face or nodding are preferably implemented as reflexive behaviors.

In the robot apparatus 1 according to this embodiment, the short-term storage unit 55 stores the results of a plurality of recognizers such as the visual recognition function unit 51, the auditory recognition function unit 52, and the touch recognition function unit 53 described above temporally. They are integrated so as to maintain spatial consistency, and the perception of each object in the external environment is provided as a short-term memory to the context-dependent behavior hierarchy (SBL) 58 and the like. Further, the recognition result of the external information inputted by the sensor is directly provided to the reflexive SBL 59.

Generally, if only the situation-dependent action hierarchy (SBL) 58 that determines an action in consideration of various conditions and the deliberation action hierarchy 57, the reaction of the robot becomes slow. Therefore, the behavior control system 50 of the robot apparatus 1 according to the present embodiment tries to determine the expression of the behavior in consideration of various conditions (internal state and external state), and therefore, the situation-dependent behavior hierarchy (SBL) 58. In addition to the consideration behavior layer 57, the reflexive action part (Reflexive SBL) 59 that determines the expression of an action under a certain single sensor condition is configured as a separate process.

Therefore, for example, the situation-dependent action hierarchy (S
A command can be transmitted asynchronously from the BL) 58 or the reflexive SBL 59. In the robot device 1 according to the present embodiment, since these asynchronous commands are arbitrated, the command manager (CM:
Command Manager) is used.

The command manager shown in FIG. 5 corresponds to a plurality of clients (Clients) from the upper layer, for example, the situation-dependent action hierarchy (SBL) 58, or clients from the reflexive action unit (Reflexive SBL) 59, for example. . The command manager also corresponds to a plurality of lower layer agents Agent1, Agent2, Agent3. These lower layer agents are the trunk unit 2, the head unit 3, and the arm unit 4R described above.
/ 4L, leg units 5R / 5L, and other resources related to the devices necessary for the robot device 1 to operate (Resourc
Under the command e), understand the command from the client according to the scene and execute the process based on the independent judgment. Each of the agents Agent1, Agent2, Agent3 executes processing for a plurality of resources.

FIG. 6 shows a flow of operations performed by each object that constitutes the behavior control system 50 shown in FIG.

The entire system operates by asynchronously communicating objects. Each object passes data (property) by message communication and inter-object communication method using shared memory, and inherits a program (method) that describes the behavior of things. The function of each object will be described below.

The color detection unit (MultiColorTracker) is an object for performing color detection, receives image data from an image input device such as a camera, extracts a color area based on a plurality of color models that it has in advance, and performs continuous extraction. Divided into regions. Position and size of each divided area,
Outputs information such as features, and stores it in the short-term memory (ShortTermM
send to emory).

The sound detection unit (AudioRecog) is an object that receives voice data from a voice input device such as a microphone and performs feature extraction and voice section detection. When the microphone is stereo, the sound source direction estimation in the horizontal direction can be performed. When it is determined that it is a voice section,
The feature amount and the sound source direction of the voice data of the section are sent to the short-term storage unit.

The contact detection unit is built in the head of the machine body or the like, detects an external stimulus such as "stroked" or "struck", and the detection result is sent to the short-term storage unit.

Although not shown, a face detection section (FaceDetector) is also provided as a detection section for this level. It is an object that detects a face area in an image frame, receives image data from an image input device such as a camera, and reduces and converts the image data into a nine-step scale image. A rectangular area corresponding to the face is searched from all the images. The overlapping candidate areas are reduced, and information such as the position, size, and feature amount regarding the area finally determined to be the face is output, and is sent to the object (FaceIdentify) that identifies the detected face image. This FaceIdentify generates the position and size information of the face image area and the person ID from the face image and sends it to the short-term storage unit.

Further, there are a ball identification section, a distance detection section, and the like. Details of the ball identification unit and the distance detection unit will be described later.

The short-term storage unit (ShortTermMemory) is an object that holds information about the external environment of the robot apparatus 1 for a relatively short period of time, and receives the feature amount and sound source direction of voice data from the sound detection unit. Also, information such as the position, size, and feature amount of each divided area is received from the color detection unit. Also, the detection result of the external stimulus is received from the contact detection unit. Then, these plural detection results are integrated so as to maintain temporal and spatial consistency, treated as meaningful symbol information, and passed to the situation-dependent action hierarchy.

SituatedBehaviorLayer
Is an object that determines the action (action depending on the situation) of the robot apparatus 1 based on the information from the short-term storage unit. Multiple actions can be evaluated and performed simultaneously. In addition, the action can be switched to put the aircraft in a sleep state, and another action can be activated.

The reflexive SBL determines the expression of a behavior under a single sensor condition among the color detecting unit, the sound detecting unit and the contact detecting unit.

Command Manager (CM: Command Ma
nager) arbitrates asynchronous commands from the situation-dependent action hierarchy (SBL) and the reflexive action unit (Reflexive SBL). That is, a plurality of resource arbitrations of each hardware of the robot apparatus 1 are performed through the processes of a plurality of agents for an asynchronous output command.

By the way, conventionally, in the autonomous robot apparatus, the resource management by the command manager has not been performed so much. This is because one resource was assigned to one agent. On the other hand, the robot apparatus 1 according to the present embodiment
Now, because resources can be divided and integrated, multiple resources can be assigned to one agent. The division and integration of this resource will be described later.

Further, conventionally, in the autonomous robot apparatus, only one client is permitted and command control is not performed for a plurality of clients.
At the same time, there is no command response control for multiple clients. On the other hand, in the robot apparatus 1, since the command manager arbitrates the command from the client side based on the priority,
It is now possible to have multiple clients. Correspondence to the plurality of clients will also be described later.

The control of resources and the like by the command manager of the robot apparatus 1 will be described in detail below.

First, the command conversion processing from the client to the agent in the command manager will be described with reference to FIG. Here, the process from the command manager receiving a command from the client to sending it to the agent will be described.

When the command is received from the client in step S1, the command manager operates in step S2.
Check to see if the command is a cancel command. If the command manager determines that the command is not a cancel command (NO), the command manager interprets the command (step S3). Then, in step S4, it is checked whether or not the interpreted command is a meta command that is a command that describes the data of the command. If it is not a meta command (NO), the command is a normal command, so the process proceeds to step S5 and the resource Manage. Regarding the management of this resource, a process of dividing and integrating the resource and sending it to the agent will be described later. And step S
At 6, the command is accumulated and transmitted to the queue of the output buffer provided in the command manager.

On the other hand, if it is determined that the command received in step S2 is a cancel command (YES),
The command manager interprets the cancel command in step S7.

Conventionally, the cancel command has been passed through to the agent by the command manager. As a result, for example, if a command for using a certain resource is sent twice while canceling for the same resource,
It could disrupt the sequence.

Therefore, the command manager of the robot apparatus 1 interprets the cancel command in step S7, and confirms whether the cancel command has already been issued and is in the process of execution, or waiting for a reply after hitting the cancel command. To do. This allows you to wait without wasting a lot of cancellations. Then, step S8
A cancel command is generated at, and the cancel command is transmitted at step S9. Details of the processing for the cancel command will be described later.

If it is determined in step S4 that the command is a meta command, the process proceeds to step S10 to perform meta command change processing.

When the meta command is handled, the conventional robot device does not complement the parameters and command names, but the robot device 1 of the present embodiment receives the meta command such as parameter change. Then, I converted it to the corresponding command and sent it to the agent. Also, when the response to the meta command is received from the agent, it is converted into the response of the corresponding meta command and sent to the client. The processing for this metacommand will also be described in detail later.

Next, with reference to FIGS. 8 and 9, a description will be given of the resource-based normal command reception and transmission processing following step S1, step S3, step S5, and step S6 in FIG. This is the case when the command received from the client is a command that uses multiple resources.

In the conventional robot apparatus, one resource is assigned to one agent. However, in the robot device 1 of the present embodiment, the command manager is provided with an internal buffer in addition to the output buffer so that the agent and the resource do not have to have a one-to-one correspondence.

In step S1 of FIG. 8, it is assumed that one command (UseResource: AB) for using two resources A and B is received from the client. Then, in the command interpretation process of step S3, the two resources of the one command (UseResource: AB) are divided into resources A and B, which are separately stored in the internal buffer. In the resource management process of step S5, the currently used resource and the divided resource A,
Compare with B. And if the buffers do not overlap,
In step S6, the commands are combined, and the combined commands are transmitted to the queues of the output buffers assigned to the agents with the accumulated timing.

As described above, in the robot apparatus 1, even when a command having a plurality of resources is received from the client, it is decomposed into basic resources at the time of input and included in the actual resource at the time of output, so that a partial resource can be obtained. It is possible to detect such conflicts, and it is possible to adjust the timing of issuing a command that is sent as one command and that spans multiple resources. That is, in the robot device 1, it is not necessary to limit the number of basic resources that the agent has. In addition, command resources can now be properly managed even by freely setting them.

Next, in FIG. 7, step S1, step S2, step S7, step S8 and step S9.
Regarding the processing for the cancel command that follows, FIG. 10
Will be described with reference to.

In this robot apparatus 1, even if the cancellation is hit twice in the upper layer such as the client 1, in order to prevent the cancellation that disturbs the useless status, the second cancellation is actually performed. Instead of sending a command, when a reply to the cancel command comes from the relevant resource, a reply is created at the same time and sent to the client.

When the command is received in step S21 in FIG. 10, the command to be canceled is retrieved from the command buffer in step S22. Also,
If it is searched in step S23 whether there is a command to be canceled in the internal buffer, the command is fetched there (POP). Furthermore, it is checked in step S24 whether or not there is any command to be canceled, and if it is determined that there is no command (NO), then step S2
The process proceeds to step 5 to create a reply to the canceled command group, further creates a reply to the cancel command in step S26, and sends the reply to the client in step S27.

On the other hand, if it is determined in step S24 that there are still commands to be canceled (YES),
In step S28, it is checked whether or not the resources of the canceled cancel command and the remaining cancel command are covered. If the resource is covered (YES), the process advances to step S29 to put in the reply waiting queue of the cancel command having the overlapping resource. In addition, the cancel command that has already been issued in step S28 and the resource are not covered (N
If it is determined to be O), the process proceeds to step S30 and a cancel command for the resource to be sent is generated and sent to the agent.

As described above, in the command manager, there is a possibility that the cancel command has already been entered when the cancel command comes from the client. Therefore, it is necessary to determine whether the cancel command is already being executed or whether the cancel command is entered and a reply is waited for. Check in S24 and S28. If the result of the check in step S28 is that the response to the previously entered cancel command is waiting, in step S29 a flag is put in the response queue for the cancel command with overlapping resources, and the cancel command from the resource is sent. When the response is received, it is assumed that the cancellation has been completed, and a command response is created in the command manager and the response is sent to the client.

In this way, the client side of the upper layer can judge that the correct response is returned to the cancel command at the correct timing. Therefore, in the command manager, the cancel command that disturbs the useless status is not passed through to the agent side, so that the robot apparatus 1 can function only with necessary and sufficient commands.

Next, with reference to FIG. 11, description will be given of the processing for the meta command following step S1, step S3, step S4 and step S10 in FIG. When the command coming from the client is, for example, "walk", the meta command relates to, for example, changing the parameter such as changing the speed. When a meta command arrives, the command manager changes only the value to be changed in step S41 if the parameter is changed, searches for what command it actually becomes (step S42), and newly updates in step S43. After creating a new command with the command name and member variable as a set and updating it using the buffer in step S44,
Transmission is performed after the queue is added to the output buffer (step S45).

Therefore, when the command manager of the robot apparatus 1 receives a meta command such as parameter change, it can actually convert the meta command into a corresponding command and send it to the agent. When a response to the metacommand is received from the agent, it can be converted into a response to the corresponding metacommand and sent to the client. In other words, it became possible to retrieve the necessary information using the metacommand and send it to the agent. In addition, by supplementing the response from the agent and creating a correct response, the response from the metacommand can now be received correctly.

Next, the process of transmitting the information from the agent side to the client in the command manager will be described. In the past, the only option was to inform the agent of the response to the command from the client. The agent did not have a route through which quick feedback information could be sent. Therefore, until now, there was no function to convey information from the agent side to the client side in the opposite direction.

Therefore, the command manager of the robot apparatus 1 can receive the information indicating the status change from the agent. Then, information such as the state change on the agent side can be sent to the client side. As a result, it is possible to reduce the configuration dependency of the upper client and make more appropriate action by effectively using the information from the agent.

FIG. 12 shows a processing procedure for transmitting information from the agent to the client. The information is received in step S51, the information is interpreted in step S52, and it is checked whether the information has a command. The flag is updated in step S53, and step S
The resource is updated at 54. The posture is updated in step S55. Next, the member variable is substituted in step S56. Thereafter, it is checked in step S57 whether or not the information is related to the specific command, and if it is determined that the information is related to the specific command (YES), step S58.
Notify the client that issued the command. It is not information related to the specific command in step S57 (NO)
If it is determined, all the clients are notified in step S59.

Here, the information voluntarily sent from the agent and received by the command manager is information known only to the agent. For example, warnings for limiter checks of joint angles and event information such as falls may be cited. The warning to the joint angle limiter check is, for example, information that the joint angle limiter is exceeded. This only needs to be reported to the client that issued the command. On the other hand, event information such as a fall must be notified to all clients.

The command manager, on the contrary, also has a mechanism for transmitting information from the client to the agent without blocking anything. When there is information that you want to send directly to the agent, add a special flag on the client side. As a result, the command manager passes the command with the special flag to the agent without controlling the resource.

As described above, the command manager prepares both paths for exchanging information other than the exchange of commands from the client side to the agent side and from the agent side to the client side.

Also, as described above, the command manager of the robot apparatus 1 can have a plurality of clients. If you have multiple clients, the context-dependent action hierarchy (client 1) that determines the main action and the reflex action part (client 2) that performs reflexive actions, commands from multiple clients will also compete for resources. Watch and issue commands efficiently so that they do not collide.

Therefore, the command manager manages the resources using the dynamic priority for the commands coming from a plurality of clients. The command manager can check the priority of commands from multiple clients while accumulating in a built-in buffer. That is, it receives commands from a plurality of clients, arbitrates the received commands based on the priority, and sends them to the agent.

Further, when the priority values for objects coming from the same object change every moment, it becomes as follows. For example, if a command from client 1 has already accumulated in the command manager buffer and a new command comes from client 1 and the priority value has changed,
Update what was originally there and decide the order in which the commands are issued with the new information.

Further, the command manager also controls the command response from the agent to the client, whether it is a response to a plurality of clients or a response to a specific client. Controls whether the response from the agent is returned to the specific client that received the command, or to all clients. If only the client that sent a response to a particular command needs it, return the response only to the specific client that issued the command. If it is better to respond to all clients, reply to all clients. return it. This is judged by the command manager by adding a flag on the agent side.

Next, the specific processing of the command manager will be described with reference to FIGS. 13 to 22. Figure 1
3 to 22 are sequence diagrams among the client, the command manager, and the agent. However, here, since the processing is related to resource management (RM), the client is described as an RM client, the command manager is described as an RM, and the agent is described as an RM agent.

First, a specific example of the normal process of the command manager for one command from one client will be described with reference to FIG. A command called Standing-byebye is issued from one client (RMClient2). Command manager (RM)
Controls the resource of Standing-byebye, and the agent (R) of the resource (Arm) specified by the command
Command to MAgent). The agent then responds that it has received the command Arm: Exec (Standing-byeb
ye): standing is returned to the command manager, then the command is executed, and the completion notification is sent: Free: Complete (Standing -b
yebye): Send standing to the command manager. The command manager receives the completion notification and notifies the client (RMClient2) that issued the command Free: Comp
Stream lete (Standing-byebye).

Next, a specific example of the normal process of the command manager for one command specifying a plurality of resources from one client will be described with reference to FIG. One client (RMClient1) issues the command Standing-byebye. Commanding Manager (RM) is Standing
-Controls the resource of byebye, and the resource specified in the command, here head unit Head and arm Arm
Command to the agent (RMAgent) of. At this time, the command manager divides the resource, puts it in the internal buffer, compares it with the resource currently being used, and divides the resource according to the comparison result, as described with reference to FIGS. 8 and 9. Are combined with each other, and the combined command is accumulated in each queue of the output buffer assigned to each agent and the timing is measured and transmitted to the agent.
Multiple resources are assigned to this agent. Then, the agent returns the response that it received the command HeadArm: Exec (Standing-byebye): standing to the command manager, then executes the command, and notifies completion: Free: Complete (Standing-byebye): standin
Send g to the command manager. The command manager receives the completion notification and notifies the client (RMClient1) that issued the command Free: Complete (Standing-
byebye).

Next, the resource management process of the cancel command will be described with reference to FIG. A command called Standing-byebye is issued from one client (RMClient1), the command manager (RM) controls the resource, and passes the command to the agent of the resource designated by the command. Then, the agent returns a reply Arm: Exec (Standing-byebye): standing that the command is received to the command manager. Command N_STOP (Standing-byebye) to cancel standing bye from the same client (RMClient1) at the next timing
Is issued, the command manager sends the command to the agent of the corresponding resource Arm. Furthermore, when the client (RMClient1) issues the second cancel command Arm: N_STOP () here, the command manager does not issue the second cancel command to the agent based on the processing procedure shown in FIG. Free: Incomplete (Standing-byebye) NSto from agent
Arm p: Neutral as a client Arm: Incomplete (Standing-b
yebye) Response to the first cancel command after being sent as N_Stop When Free: Complete (N STOP): standing comes, two replies (first cancel command and second cancel command) to the client at the same time. Free: C
Create omplete (N STOP), Free: Complete (Arm: N_STOP) and send to client.

Next, a specific example of the process of transmitting the information from the agent side to the client will be described with reference to FIGS. 16 and 17. FIG. 16 shows the processing for sending the notification from the agent only to a specific client, and FIG. 17 shows the processing for sending the notification from the agent to all the clients. The command manager can determine whether the information from the agent side is sent only to a specific client or to all clients by giving an ID to the agent.

In FIG. 16, one client (RMCl
ient1) issues a command called Standing-byebye. Command manager (R
M) controls the resource of Standing-byebye, and passes the command to the agent (RMAgent) of the resource (Arm) specified by the command. The agent then responds that it has received the command Arm: Exec (Standin
g-byebye): Returns standing to the command manager and then tries to execute the command. At this time, the agent tries to notify the upper layer by issuing overlimitation information that the joint angle of the arm exceeds the limiter by the above command, and the overlimitation information Arm: Warning (Standing-byebye): standing:
Send OverLimitation to command manager. At this time, since this information needs to be notified only to the client that has issued the command, for example, the ID is not added and it is the default. Then, the command manager only sends the overlimitation information Arm: Warning (standing-byebye): standing: to the client that issued the command.
Notify Over Limitation. The agent then reports Free: Incomplete (Standi) that the standing by buy command was unsuccessful and is currently in neutral.
Send it to the command manager as ng-byebye): Falldown: Neutral. The command manager Fre
e: Incomplete (Standing-byebye): Notify only the client that issued the command as Falldown.

In FIG. 17, first, one client (RMClient1) stands by by Standin.
The command g-byebye is issued. The command manager (RM) controls the resource of Standing-byebye and passes the command to the agent (RMAgent) of the resource (Arm) specified by the command. Then,
Agent has responded to the command Arm: Ex
ec (Standing-byebye): Returns standing to the command manager and then executes the command. At this time, it is assumed that the user has fallen down if the arm is moved by the command. Then, the agent determines that the event of falling is event information that affects all clients, so it determines that the information should be notified to all clients, assigns an ID, and All: FallingDown (Standing-b
yebye): Send as standing to the command manager.
The command manager determines that the fall information All: FallingDown (Standing-byebye): standing is information to be passed to all clients by the flag, and not only the client (RMClient1) but also the client (RMClient1).
Pass it to 2) as well.

Next, a specific example of the process for the meta command will be described with reference to FIG. First of all, from one client (RMClient1) to Standing bye S
The command tanding-byebye is issued. The command manager (RM) controls the resources of Standing-byebye, and the resources specified by the command (Ar
Pass the command to the agent (RMAgent) of m). The agent then replies Ar that it has received the command.
m: Exec (Standing-byebye): standing is returned to the command manager and the command is executed. In the middle of the run,
From the same client, the metacommand ChangeParam (Standing-b
yebye) is issued, the command manager changes only the value that you want to change according to the meta command, searches for what command it actually becomes, and the new command Standing-b
yebye (2): Generate ChangeParam and send it to the agent. The agent uses the new command Standing-byebye
(2): Response to ChangeParam Arm: Exec (Standing-bye
bye (2)): returns standing to the command manager and executes the new command. And Arm: Complete (Sta
Notify the client via the command manager to perform nding-byebye (2): change: standing.
Here, the client sends the cancel command N_STOP (Sta
nding-byebye) is issued, the command manager notifies the agent of the corresponding resource Arm of the command. A description of the processing of the agent for this cancel command is omitted.

Next, the processing for the command for changing the resource will be described with reference to FIG. A command called Standing-byebye is issued from one client (RMClient). Command manager (R
M) controls the resource of Standing-byebye, and passes the command to the agent (RMAgent) of the resource (Arm) specified by the command. The agent then responds that it has received the command Arm: Exec (Standin
g-byebye): returns standing to the command manager and then executes that command. Standing Hello Standing changing resources from the same client here
-When a command "hello (change)" is sent, the command manager sends the command N_STOP (Standing-byeby)
Pass e) to the agent. In response, the agent
Free: Incomplete (Standing-byebye) NStop: Neutral is passed to the command manager, which indicates that the agent has received the command for the changed resource Arm: Incomplete (Standing-byebye) change
To the client. And the reply that the agent stopped standing bye Free: Complete (NS
When you send top): standing to the command manager, the command manager passes the command Standing-hello (change) to the corresponding resource. Then the agent responded that it received the command Arm: Exec (Standing-hello): s
Send tanding to the command manager and execute the command. When the execution is finished, the agent notifies the completion
Send Free: Complete (Standing-hello): standing to the command manager. The command manager receives the completion notification, and issues the command to the client (RMCl
flow Free: Complete (Standing-hello) to ient).

Next, the processing for the added command will be described with reference to FIG. The command Standing-byebye is issued from one client. The command manager controls the Standing-byebye resource and the resource (Arm) specified in the command
Command to the agent (RMAgent) of. The agent then replies Ar that it has received the command.
Execute m: Exec (Standing-byebye): standing after returning it to the command manager. Here, when an additional command called Standing-hello (add) is sent from the same client, the command manager stores this command in an internal buffer and notifies the completion notification from the agent Free: Complete (Standing-byebye): standin.
After g is sent, take it out of the buffer and pass the command Standing-hello to the agent. The agent responded that it received the command Arm: Exec (Standing-
Send hello): standing to the command manager and execute that command. When the command execution is complete, the agent sends a completion notification Free: Complete (Standing-hello):
Send standing to the command manager. The command manager receives the completion notification and notifies the client (RMClient) that issued the command to Free: Complete (Stan
ding-hello).

Next, the processing for the immediate command will be described with reference to FIG. This immediate command is a prompt response command that does not have to be executed if it cannot be executed immediately. The command Standing-byebye is issued from one client. Command Manager Standing
-Controls the resource of byebye and passes the command to the agent (RMAgent) of the resource (Arm) specified by the command. Then, the response that the agent received the command Arm: Exec (Standing-byebye): standing
To the command manager and then execute the command. Now from the same client Standing-hello
When the command (immidiate) is sent, the command manager notifies the completion notification from the agent Free: Compl.
ete (Standing-byebye): standing has not been sent, so the immediate command is sent to the agent and Free: Incomplete (Standing-hel is sent to the client.
lo): Returns RM. After that, the agent notifies the command manager of completion Free: Complete (Standing-byeby
e): standing is sent, and it is notified to the client (RMClient) that issued the command Free: Complete (Standing-bye
Even bye), the immediate command is not passed to the agent.

Next, arbitration processing based on priority for commands from a plurality of clients will be described with reference to FIG. First, the client 2 (RMCli
ent2) issues a Standing-byebye with a priority value of 50. The command manager searches for the Standing-byebye resource and passes the command to the agent (RMAgent) of the resource (Arm) specified by the command. Then, the agent returns a reply Arm: Exec (Standing-byebye): standing that the command is received to the command manager and then starts executing the command.
Here, from the client 1 (RMClient1) to the Standing-
Command Cr with higher priority value 100 than byebye
Suppose apHand is issued. Then, the command manager gives the agent St to give priority to the command CrapHand from the client 1, which has a high priority value.
anding-byebye stop command N_Stop (Standing_byebye)
Give out. And from the agent Free: Incomplete (St
anding-byebye) When NStop: Neutral is returned, Arm: Incomplete (Standing-byebye) chang is sent to the client.
After this, the agent sends Free: Complete (N STOP): st
After waiting for the notification of anding, the CrapHand with high priority is issued to the agent. The agent returns an anchor Arm: Exec (CrapHand): standing indicating that it has been received to the command manager, and upon completion of execution, returns a completion notification Free: Complete (CrapHand): standing to the command manager. The command manager sends the completion notification Free: Complete (CrapHand) to the corresponding client 1 (RM
Pass to Client1).

The specific processing example of the command manager has been described above. Next, a specific operation example of the robot device 1 will be described. In this specific example, as shown in FIG. 23, when the robot apparatus 1 is chasing the pink ball 90, a cliff is seen and the robot apparatus 1 stops. In FIG. 23, the robot apparatus 1 uses a PSD sensor 80 including a CCD sensor and the like to make a pink ball 90.
Looking at.

FIG. 24 shows the object structure. When the color detection unit detects the pink color and the ball identification unit identifies the ball, the action unit (for example, the client 1) playing with the ball goes to the ball and plays with the ball. Send a signal to the command manager. Further, the distance detection unit of the robot apparatus 1 continuously detects the distance to the ball and sends the distance detection signal to the emergency stop action unit (for example, the client 2).

The command manager receives the operation command signal as shown in FIGS. 8 and 9, interprets it, divides the resource, puts it in an internal buffer, and then compares it with the resource currently used. If the buffers do not overlap, the commands are combined and stored in the queue of the output buffer for timing, and sent to the agent of the corresponding resource. Therefore, the robot apparatus 1 approaches the ball 90 while detecting the distance to the ball 90.

The emergency stop action section issues an emergency stop command from the distance detection signal from the distance detection section and the running operation signal representing the command selected and executed by the command manager. As a result, the robot device 1 makes an emergency stop.

Next, each object will be described.
First, the color detector will be described. Pixel P (x, y) on the image
Have respective values of color information (y, u, v) as shown in FIG. Therefore, the center and the size of the area that matches the color space set to pink are detected.

Assuming that the color space of pixels is represented by 8 bits, as shown in FIG. 25, 0 ≦ y ≦ 255, 0 ≦ u ≦ 255, 0
Since it is expressed by ≦ v ≦ 255, the space designation method is shown in FIG.
, (U1, v1), (u2, v
It can be defined as a rectangle defined in 2). Here, Color [y] [0] = u1; Color [y] [1] = v1; Color [y] [2] = u2; Color [y] [3] = v2; Therefore, the color space color can be defined as Color [256] [4].

Next, the color detection algorithm will be described with reference to FIG. If the position of the pixel on the image is P (x0, x1) and the yuv data is Cu (x0, x1), Cv (x0, x1), Cy (x0, x1), after filtering through the color space, Image data (1 if it matches the set color space, 0 otherwise) Q (x0, x1) is Becomes

The color detection signal is obtained by adding the color space id to the above detection result.
Is added. Therefore, ColorDetectResult (id, centerX, centerY, sizeX, sizeY) = (PINK, 5,4,5,5).

Next, the ball identification section will be described.

From the color detection signal, the distance between the balls is judged from the size. As mentioned above, the ball is specified in color space as: If (ColorDetectResult.id == PINK) Here, assuming that the distance is proportional to the number of pixels, Distance = (k /
Pixel _total ) * sizeX * sizeY. However, Pixel
_total is the total number of pixels, and k is a distance multiplier around 1 pixel. In this example, Pixel _total = 9 * 7, sizeX = 5, siz
Set eY = 5, Distance = k / 63 * 25.

A calculation example is shown below. If you show the ball at 100 cm and it is 20 pixels, 100 = k / 63 *
20, k = 315 [cm / pixel]. So in this case Di
stance = 125 [cm].

Therefore, the ball identification signal objectDetect
(id, distance) becomes objectDetect (id, distance) = (BALL, 125 [cm]).

Next, the action section for playing with the ball is shown in FIG.
This will be described with reference to the flowchart of. In step S61, it is checked whether or not there is a ball, and in step S62, it is checked whether or not the distance is shorter than 10 cm. Further, in step S63, it is checked whether the distance is closer than 8 cm. Then, in step S61, when the ball is visible and the distance is longer than the contact distance (here, 10 cm) (NO in step S62), the ball is closer (step S66) and is closer than 8 cm (step S61).
If YES, go down to 64) (step S64), 8 cm
It is an action of a procedure of kicking by kicking (step S65) if the distance is up to 10 cm (NO in step S63).

Next, the distance detector will be described with reference to FIG.

The sensor result obtained by the PSD sensor is measured as distance.
_{Assuming psd} , the distance d _floor to the floor where the robot apparatus 1 is grounded is d _floor = distance _psd √2
Becomes The distance detection signal can be expressed as FloorDetect (d _floor ). Therefore, if there is a cliff 95 with a height of 20 cm,
The distance detection signal FloorDetect (d _floor ) becomes 70 cm.

Next, the emergency stop action section will be described with reference to FIG. In step S71, the distance detection signal FloorD
It is checked whether etect (d _floor ) exceeds 50 + thresould, and further it is checked in step S72 whether a command to move forward is being executed. The threshold thresould in step S71 depends on the performance of the robot apparatus on rough terrain, and is 5 cm here. Then, the distance detection signal from the distance detection unit exceeds 55 cm (step S7).
If YES in 1) and the in-execution operation signal representing the command selected and executed by the command manager is a command to walk forward (YES in step S72), an emergency stop command is issued (step S73). Emergency stop of the robot apparatus 1.

Next, the operation command signal will be described.

The operation command signal is (id, clientID, prior
ity, policy). In the above-mentioned specific example, it is transmitted as ("forward", PLAYBALL, 50, ADD). Where PL
AYBALL represents the action of playing with the ball, and ADD represents the policy of buffering if the commands are the same.

Then, the command manager selects and executes a command according to the following processing. First,
In step S81, it is checked whether they have the same clientID, and if they are the same, it is checked whether the new command is add or change. In case of different clientID, check which one has higher priority.

FIG. 31 shows a flowchart of command selection and execution. In step S81, the client ID is checked to see if the command is sent from the same client client. If it is determined that the same client has transmitted the data (YES), the process proceeds to step S82 and priori
Check ty to see if it has a higher priority than the previous command. If the priority of the new command is higher (YES), the process proceeds to step S83 to change the command, and then the in-execution operation command signal is transmitted to the emergency stop action part in step S84. Step S
If it is determined at 81 that the command is from a new client, the process proceeds to step S85 to determine whether the priority of the command from the new client is higher. If it is determined that the command from the new client has a higher priority, the process proceeds to step S83 to change the command. On the other hand, if it is determined in step S85 that the priority of the command from the new client is not high (NO), the process proceeds to step S86 to buffer the new command. Also, step S
If it is determined that the priority is not higher than the previous command in 82 (comparison of commands from the same client) (NO), the process proceeds to step S87 to check whether the selection policy is Add. Where A
If dd, proceed to step S86 to buffer a new command. If not Add, step S8
Go to step 8 and do not accept new commands.

In the concrete example of the action of finding and playing the pink ball described above with reference to FIG.
The forward command is executed by ("forward", PLAYBALL, 50, ADD) until is detected. When the cliff 95 is detected by the distance detection unit, a stop command is transmitted from the emergency stop action unit to the command manager at (“stop”, EMERGENCY, 100, ADD). Here, since the clientID is different,
From step S85 to step S83, the command is changed and the robot operation is stopped.

As described above, according to the embodiment of the present invention, in the robot device, resource division,
It is possible to perform integration, to prevent the sequence from being disturbed due to double hitting of a cancel command, and to supplement parameters and command names when handling metacommands.

Further, the agent side has a path through which quick feedback information and the like can flow, and command control for a plurality of clients is possible, and at the same time, command response control can be handled for a plurality of clients.

The present invention is not limited to the above-described embodiments, and it goes without saying that various modifications can be made without departing from the gist of the present invention. In the present embodiment, the bipedal legged mobile robot has been described, but the robot device is applicable as long as it operates according to the internal state, and the moving means is
The present invention is not limited to bipedal walking and even leg movement.

[0153]

The robot device and the control method of the robot device according to the present invention divides and integrates resources, prevents the sequence from being disturbed due to double hitting of a cancel command, and handles meta commands. , Parameter and command name can be completed. In addition, the agent has a route through which quick feedback information can be sent, and command control for a plurality of clients is possible. At the same time, command response control can be handled for a plurality of clients.

[Brief description of drawings]

FIG. 1 is an external perspective view of a robot device.

FIG. 2 is a schematic diagram showing joint degrees of freedom of a robot apparatus.

FIG. 3 is a diagram schematically showing a control system configuration of a robot apparatus.

FIG. 4 is a diagram showing a basic architecture of a behavior control system of a robot device.

FIG. 5 is a diagram showing a relationship between a command manager, a client, and an agent.

FIG. 6 is a diagram showing a flow of an operation performed by each object constituting the behavior control system shown in FIG.

FIG. 7 is a flowchart showing command conversion from a client to an agent by a command manager.

FIG. 8 is a flowchart showing division and combination of a plurality of resources by a command manager.

FIG. 9 is a diagram for explaining division and combination of the plurality of resources.

FIG. 10 is a flowchart showing processing of a cancel command by the command manager.

FIG. 11 is a flowchart showing processing of a meta command by the command manager.

FIG. 12 is a flowchart showing a process of notifying a client of information from an agent by a command manager.

FIG. 13 is a sequence diagram for a normal command.

FIG. 14 is a sequence diagram for a normal command that handles multiple resources.

FIG. 15 is a sequence diagram for a cancel command.

FIG. 16 is a sequence diagram for notification from an agent.

FIG. 17 is a sequence diagram when an agent robs a resource.

FIG. 18 is a sequence diagram for a meta command.

FIG. 19 is a sequence diagram of resource management for a change command.

FIG. 20 is a sequence diagram of resource management for an add command.

FIG. 21 is a sequence diagram of resource management for the immidiate command.

FIG. 22 is a sequence diagram for a plurality of clients.

FIG. 23 is a diagram showing a specific operation example in the robot apparatus.

FIG. 24 is a diagram showing an object configuration of the operation example.

FIG. 25 is a diagram showing a color space.

FIG. 26 is a diagram showing a color region on a plane.

FIG. 27 is a diagram for explaining a color detection algorithm.

FIG. 28 is a flowchart showing a processing procedure of an action unit playing with a ball.

FIG. 29 is a diagram for explaining a distance detection unit.

FIG. 30 is a flowchart showing a processing procedure in the distance detecting unit.

FIG. 31 is a flowchart showing a processing procedure of command selection / execution.

[Explanation of symbols]

1 robot device, 11 main control units, 50 behavior control system

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Shinichi Kariya             6-735 Kita-Shinagawa, Shinagawa-ku, Tokyo Soni             -Inside the corporation (72) Inventor Masahiro Fujita             6-735 Kita-Shinagawa, Shinagawa-ku, Tokyo Soni             -Inside the corporation F-term (reference) 3C007 AS36 CS08 LV14 MT08 MT14                       WA03 WA13 WB15 WB16 WB23                       WB25

Claims (17)

[Claims] 1. An autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, a client unit that is an action generation unit that generates the autonomous action, Command management means for managing a command relating to an action generated by the client means, and agent means for receiving a command managed by the command management means and driving a resource to cause the action, the command management A robot apparatus, wherein the means divides the resources into basic resources at the time of input when a plurality of resources are specified in the command from the client means, and merges the resources at the time of output. 2. The command management means stores the basic resource divided at the time of input in an internal buffer, compares it with the resource currently used, and merges it into the output buffer at the time of output if there is no duplication as a result of the comparison. The robot apparatus according to claim 1, wherein the robot apparatus stores resources and waits for an output timing. 3. An autonomous robot apparatus that autonomously acts based on a recognition result for an internal state or an external environment based on an emotion model, a client unit that is an action generating unit that generates the autonomous action, Command management means for managing a command related to an action generated by the client means, and agent means for receiving a command managed by the command management means and driving a resource for causing the action, the command management After the cancel command for the command from the client means is sent, the means further responds to the cancel command from the client means for the same resource when a reply to the previous cancel command from the corresponding resource comes at the same time. Make, applicable cry Robot apparatus characterized by sending the cement. 4. In an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, client means that is an action generating unit that generates the autonomous action, Command management means for managing a command related to an action generated by the client means, and agent means for receiving a command managed by the command management means and driving a resource for causing the action, the command management The means for receiving, when receiving from the client means a meta-command relating to a parameter change of a command already sent, actually converting the command into a corresponding command and sending it to the agent means. 5. The command management means, when receiving a response to a command from an agent that has sent a command converted from the meta command, converts the command response into a response to the meta command and The robot device according to claim 4, wherein the robot device is sent to a client means. 6. An autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, a client unit that is an action generation unit that generates the autonomous action, Command management means for managing a command related to an action generated by the client means, and agent means for receiving a command managed by the command management means and driving a resource for causing the action, the command management A robot apparatus characterized in that, when the information about the change in the state due to the driving of the resource is sent from the agent means, the means sends the information to the client designated by the agent means. 7. The command management means specifies information regarding the state change caused by the driving of the resource based on identification information added to the information regarding the state change caused by the driving of the resource by the agent means, a specific client means or a plurality of clients. 7. The robot device according to claim 6, wherein the robot device is supplied to the client means. 8. In an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, a plurality of client units that are action generating units that generate the autonomous action. And command management means for managing commands related to the actions generated by the plurality of client means, and agent means for receiving the commands managed by the command management means and driving the resources for causing the actions. The robot apparatus characterized in that the command management means arbitrates a plurality of commands sent from the plurality of client means according to the priority added by the plurality of client means and sends the command to the agent means. . 9. The robot apparatus according to claim 8, wherein the command management unit manages whether the reply from the agent unit is sent to all of the plurality of client units or to a specific client. . 10. In an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, client means that is an action generating unit that generates the autonomous action, Command management means for managing a command related to an action generated by the client means, and agent means for receiving a command managed by the command management means and driving a resource for causing the action, the command management A means for arbitrating a plurality of commands issued in time series from the client means based on a selection policy. 11. The robot apparatus according to claim 10, wherein the selection policy indicates a prompt response command, an addition command, and a change command. 12. A method for controlling an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, the action generating step of generating the autonomous action, The command management process includes a command management process that manages a command related to the action generated by the action generation process, and an agent process that receives a command managed by the command management process and drives a resource for causing the action. A means for controlling a robot apparatus, characterized in that when a plurality of resources are specified in the command from the action generation step, the means divides into basic resources at the time of input and merges the resources at the time of output. 13. A method of controlling an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, the action generating step of generating the autonomous action, The command management process includes a command management process that manages a command related to the action generated by the action generation process, and an agent process that receives a command managed by the command management process and drives a resource for causing the action. The process makes a response at the same time when a cancel command for the command from the action generation process is sent, and further for a cancel command for the same resource, when a response to the previous cancel command from the relevant resource comes, Robot device characterized by sending to a client Control method of. 14. A method of controlling an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, the action generating step of generating the autonomous action, The command management process includes a command management process that manages a command related to the action generated by the action generation process, and an agent process that receives a command managed by the command management process and drives a resource for causing the action. The method of controlling a robot apparatus, characterized in that, when the process receives a meta-command relating to parameter change of a command already sent from the action generation process, the meta-command is actually converted into a corresponding command and sent to the agent process. 15. A method of controlling an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, the action generating step of generating the autonomous action, The command management process includes a command management process that manages a command related to the action generated by the action generation process, and an agent process that receives a command managed by the command management process and drives a resource for causing the action. In the process, when the agent process sends information about a change in a state caused by driving a resource, the information is sent to a client of the action generation process designated by the agent process. 16. A control method of an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, comprising a plurality of action generating steps for generating the autonomous action. A command management step of managing commands related to the behaviors generated by the plurality of behavior generation steps, and an agent step of receiving a command managed by the command management step and driving a resource for causing the behavior. The robot characterized in that the command management step arbitrates a plurality of commands sent from the plurality of client means according to the priority added by the plurality of behavior generation steps and sends the commands to the agent step. Device control method. 17. A method of controlling an autonomous robot apparatus that autonomously acts based on a recognition result of an internal state or an external environment based on an emotion model, the action generating step of generating the autonomous action, The command management process includes a command management process that manages a command related to the action generated by the action generation process, and an agent process that receives a command managed by the command management process and drives a resource for causing the action. The step of arbitrating, based on a selection policy, a plurality of commands issued in time series from the action generating step, the robot apparatus control method.