CN119223269A - An autonomous navigation method for unmanned shotcrete robot - Google Patents

Abstract

The present application relates to the technical field of intelligent operation robots in coal mines, and in particular to an autonomous navigation method for unmanned shotcrete robots, wherein the hardware includes a laser radar, an inertial measurement unit (IMU), and an industrial computer, and the software consists of a map construction module, a path planning module, and a host computer, wherein: the laser radar obtains point cloud data; the IMU obtains robot posture information and calculates inter-frame odometers; the industrial computer receives and parses sensor data, constructs an environmental map, plans the optimal path, and transmits control instructions; the map construction module constructs an environmental map, obtains positioning information and sends it to the path planning module, and calculates odometer data at the same time; the path planning module generates path information and speed instructions. The present invention has two advantages: first, there is no need to scan the environmental map in advance during the robot deployment process; second, the mapping module provides odometer information to the path planning module while constructing the environmental map, reducing the dependence on hardware.

Description

Autonomous navigation method for unmanned guniting robot

Technical Field

The application relates to the technical field of intelligent operation robots for coal mines, in particular to an autonomous navigation method for an unmanned guniting robot.

Background

Underground roadways are important channels for coal mining and coal transportation, and guniting support is an important measure for guaranteeing roadway stability and safety. However, the traditional manual spraying has the defects of low operation efficiency, unstable operation quality, high safety risk and the like. Therefore, along with the continuous progress of technology and the continuous improvement of the intelligent level, the automatic completion of the guniting operation by a robot instead of manual guniting is a necessary development trend.

In order to realize complete unmanned operation, an autonomous navigation technology is indispensable to the guniting robot. However, most of the conventional navigation technologies are based on the mode of building a map and then navigating, that is, the robot must scan the environment map in advance for storage and then can be used by the path planning module, which brings a certain difficulty and inconvenience to the on-site deployment of the guniting robot.

Disclosure of Invention

In order to solve the technical problems, the invention provides an autonomous navigation method for an unmanned guniting robot, which realizes the navigation and the map building at the same time. According to the method, an environment map is not required to be established in advance, incremental map establishment is continuously carried out in the advancing process of the robot, firstly, the pose of the robot in the map is obtained, then the next target point position is calculated according to the current pose information and the single guniting distance and is sent to a path planning module, the path planning module generates an advancing path and simultaneously sends a speed control instruction to the robot so as to realize the movement of the robot, the map is continuously expanded in the advancing process of the robot so as to facilitate the next path planning, the guniting operation is started after the robot moves to a designated target point, and after the guniting operation is completed, the robot can repeat the processes until the whole guniting task is completed.

In a first aspect, the present application provides an autonomous navigation method for an unmanned guniting robot, which adopts the following technical scheme:

The autonomous navigation method for the unmanned guniting robot comprises a hardware part and a software part, wherein the hardware part comprises a laser radar, an IMU (inertial measurement unit), an industrial personal computer and a hydraulic controller, the software part comprises a map building module, a path planning module and an upper computer, the software part runs on the industrial personal computer, and the path planning module comprises a global planner and a local planner, wherein:

Step S1, the laser radar is used for acquiring point cloud data;

the IMU is used for acquiring the gesture of the robot and calculating the inter-frame odometer information through pre-integration;

Step S2, the industrial personal computer is used as a control center and is responsible for receiving and analyzing the point cloud data, the gestures and the inter-frame odometer information, constructing a real-time environment map, acquiring the robot positioning and planning the optimal path, simultaneously communicating with the upper computer, acquiring the pose of the next target point location, completing the receiving and transmitting of control instructions,

The map construction module is used for constructing a real-time environment map by utilizing the point cloud data, the gestures and the inter-frame odometer information, sending the environment map and the positioning information to the path planning module, and simultaneously calculating the odometer information for the local planner in a multi-sensor fusion mode;

The path planning module is used for generating path information and a speed instruction for controlling the movement of the robot based on the pose of the next target point position, the environment map and the obstacle data fed back by the laser radar, wherein the path information comprises a global path and a local path;

and step S3, the hydraulic controller is used for receiving the speed control instruction sent by the path planning module and converting the speed control instruction into a control signal of the proportional control valve so as to realize the movement of the guniting robot.

The present application may be further configured in a preferred embodiment, wherein the obstacle data includes static obstacle information and dynamic obstacle information, and the path planning module is configured to generate path information based on a pose of the next target point, the environment map, and the obstacle data fed back by the lidar, and the path planning module includes:

the global planner is used for generating the global path based on the pose of the next target point position and the static obstacle information;

the local planner is configured to generate the local path based on the global path and the dynamic obstacle information.

The application can be further configured in a preferred example that the environment map comprises the position and the outline of an obstacle and passable area information, and the path planning module is used for generating path information based on the pose of the next target point, the environment map and the obstacle data, and then comprises the steps of acquiring information for controlling the movement capacity of the robot, wherein the information of the movement capacity comprises the full-side displacement capacity and obstacle avoidance capacity;

Determining a desired speed based on the obstacle and passable area information, the full square displacement capability, and the obstacle avoidance capability, and generating a speed command based on the desired speed;

Correspondingly, the hydraulic controller is used for generating a control signal of the proportional control valve based on the speed command so as to drive the hydraulic motor to realize the movement of the robot.

The present application may be further configured in a preferred embodiment, wherein the functions of the path planning module further include:

Comparing whether the current pose is the same as the pose of the next target point in real time, wherein the current pose comprises the pose and a detection position;

if the target points are different, determining that the robot does not reach the next target point, and acquiring the residual arrival time;

Transmitting the residual arrival time and the residual distance information to operation side equipment;

if the target point positions are the same, determining that the robot reaches the next target point position, and sending a reaching instruction to a control system.

The present application may be further configured in a preferred embodiment, after the sending of the arrival command to the control system, further comprising:

And the control system generates a laser radar protective cover closing instruction and executes the guniting operation.

The present application may be further configured in a preferred embodiment, after the performing of the guniting operation, further comprising:

acquiring the actual guniting distance of the robot, and judging whether the actual guniting distance is the same as the preset guniting total distance;

If not, determining pose information of a next target point position according to the current pose information and the single guniting distance, and repeating guniting operation until the preset guniting total distance is completed.

In summary, the application has the following beneficial technical effects:

1. By acquiring the point cloud data and constructing the map while advancing according to the point cloud data, the environment map does not need to be established in advance, and the deployment of the guniting robot is facilitated.

2. The method has the advantages that the method does not depend on the encoder to construct the odometer information in the running process of the robot, reduces the dependence on hardware, and has wider application range.

3. By comparing the actual guniting distance with the preset guniting total distance, when the actual guniting distance is different from the preset guniting total distance, the robot continues to advance until the actual guniting distance is the same as the preset guniting total distance, so that the robot can autonomously complete the whole guniting flow, and the participation of staff is greatly reduced.

Drawings

FIG. 1 is a block diagram of a robot hardware according to an embodiment of the present application;

FIG. 2 is a flowchart of an autonomous navigation algorithm provided in an embodiment of the present application;

Fig. 3 is a flow chart of a guniting robot according to an embodiment of the application.

Detailed Description

The present application will be described in further detail with reference to fig. 1 to 3.

Modifications of the embodiments which do not creatively contribute to the application may be made by those skilled in the art after reading the present specification, but are protected by patent laws within the scope of the claims of the present application.

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

In addition, the term "and/or" is merely an association relation describing the association object, and means that three kinds of relations may exist, for example, a and/or B, and that three kinds of cases where a exists alone, while a and B exist alone, exist alone. In this context, unless otherwise specified, the term "/" generally indicates that the associated object is an "or" relationship. The invention provides an autonomous navigation method of an unmanned guniting robot, which mainly comprises the following steps of firstly constructing a local environment map, obtaining the pose of the robot in the map, then calculating the pose of the next target point according to the current pose of the guniting robot and a single guniting distance, sending the pose of the next target point to a path planning module, generating a global path and a local path by the path planning module according to the pose of the next target point and barrier information around the robot, generating a speed control instruction so as to realize the movement of the robot, expanding the environment map continuously in the advancing process so as to provide a basis for the subsequent path planning, and continuously repeating the steps until the whole guniting task is completed. The method has the advantages that firstly, in the deployment process of the robot, a mode of navigation and map construction is adopted, an environment map does not need to be scanned in advance, and secondly, the map construction module can provide odometer information for the path planning module while constructing the environment map, so that dependence on hardware is reduced.

Further, the autonomous navigation method of the unmanned guniting robot is developed based on an ROS architecture, the autonomous navigation method is provided with a plurality of devices and modules, a hardware part comprises a laser radar, an IMU (inertial measurement unit), an industrial personal computer, a hydraulic controller and the like, a software part comprises a map building module, a path planning module, an upper computer and the like, the software part runs on the industrial personal computer, the path planning module comprises a global planner and a local planner, the devices and the modules can conduct data transmission in the form of topics, wherein the industrial personal computer is used as a control center of the robot, an environment map and odometer information are constructed by fusing information of the laser radar and the IMU, an optimal moving path and an optimal moving speed are generated, and a control instruction is sent to the hydraulic controller, so that movement of a chassis of the guniting robot is controlled.

In the working process of the guniting robot, a program is started and initialized, an industrial personal computer CAN establish communication connection with a laser radar through a network port to acquire data acquired by the laser radar, the industrial personal computer CAN read the data acquired by the IMU through a USB port, the industrial personal computer CAN also communicate with an upper computer in a wireless connection mode to realize remote monitoring instruction issuing and state feedback, the industrial personal computer CAN be in communication connection with a hydraulic controller through a CAN bus, and the hydraulic controller drives a hydraulic motor through a control proportional control valve, so that the movement of the guniting robot is realized.

And S1, a laser radar is used for scanning the surrounding environment by emitting laser beams and receiving reflected signals to obtain point cloud data, and a foundation is provided for constructing a roadway map, wherein the point cloud data comprises static obstacle information and dynamic obstacle information.

The IMU is used for feeding back the attitude information of the robot in real time and calculating the inter-frame odometer information through pre-integration operation; the method comprises the steps of acquiring the angular velocity and the acceleration of a robot, carrying out pre-integral operation on the angular velocity and the acceleration to acquire the attitude and the inter-frame odometer information, and effectively improving the positioning accuracy of the robot by using the IMU.

Step S2, the industrial personal computer can serve as a control center to receive and analyze point cloud data, postures and inter-frame odometer information and construct a real-time environment map, then acquire a robot to locate and plan an optimal path and communicate with an upper computer, and acquire the pose of a next target point to finish receiving and transmitting control instructions, as shown in FIG. 2, the flow chart of the autonomous navigation algorithm provided by the embodiment of the application mainly comprises two modules of map construction and path planning. The map construction module is used for realizing the construction of an environment map by acquiring information of sensors such as a laser radar, an IMU (inertial measurement unit) and the like, namely point cloud data, postures and interframe odometer information, sending a map (probability grid map) and positioning information of a robot in the map to the path planning module, generating odometer information of the robot for local path planning, expanding the probability grid map by the path planning module firstly, generating a global path by the global planner according to target point information and a global cost map (expanded static obstacle information), generating a local path and a robot speed control instruction by the local planner according to the global path and the local cost map (expanded dynamic obstacle information), and sending the local path and the robot speed control instruction to the hydraulic controller, so that the movement of the robot is realized, the local path can be attached to the global path as far as possible while avoiding obstacles, and the environment map can be expanded while the robot moves forwards.

Further, the path planning module is an important component of the autonomous navigation system of the robot and can generate path information based on the pose of the next target point, the environment map and the obstacle data, wherein the path planning module comprises a global planner and a local planner, and the path information comprises a global path and a local path.

The global planner is used for generating a global path based on the pose of the next target point and static obstacle data in an environment map, wherein the static obstacle data is point cloud data corresponding to an obstacle which does not move, specifically, the global planner traverses each grid in the environment map, and when the obstacle exists in the grid, the grid is not planned into the global path, so that an optimal path, namely the global path, is generated.

Further, the local planner is used for generating a local path based on the global path and the dynamic obstacle data, and in order to make the local path smoother, the local planner needs to acquire the real-time feedback speed of the guniting robot, and the information is calculated by the odometer information sent by the map construction module. The robot moves along the global path, and when the dynamic obstacle exists in the global path, the robot automatically avoids the obstacle if the dynamic obstacle can be avoided, and if the robot cannot avoid the obstacle, the robot stops advancing until the obstacle disappears and then continues advancing, namely the local path is attached to the global path as much as possible in the embodiment of the application, and when the obstacle is avoided, the robot temporarily deviates from the global path. While advancing, the laser radar can acquire new point cloud data in real time so as to expand the map.

Further, the path planning module can control the movement capability of the guniting robot, wherein the movement capability information comprises the capability of realizing the omnibearing movement and obstacle avoidance of the guniting robot, the expected speed is determined based on the obstacle information, the passable area information and the movement capability of the robot, namely, if the robot has an obstacle or needs to turn in the travelling process, the robot is controlled to decelerate so as to avoid the occurrence of the problems of rollover and the like of the robot, and the robot can travel according to the expected speed. Wherein the passable area information characterizes a grid in which no obstacle is present.

And step S3, the hydraulic controller drives the hydraulic motor by controlling the proportional control valve to realize autonomous running of the robot according to the planned path, wherein the industrial personal computer CAN communicate with the hydraulic controller through the CAN bus.

Meanwhile, the path planning module can also detect whether the robot reaches the next target point in real time, namely, the current pose of the robot is compared with the pose of the next target point, the current pose comprises the detection position and the pose of the current robot, and if the detection position of the robot is the same as the position of the next target point, and the pose of the robot is the same as the preset pose of the next target point, the robot is indicated to reach the next target point. Generating an arrival command, sending the arrival command to a control center, generating a laser radar protective cover closing command to close the protective cover of the laser radar, then performing guniting operation, and after the guniting operation is completed, regenerating a laser radar protective cover opening command.

If the detected position of the robot is different from the position of the next target point, or if the gesture of the robot is different from the preset gesture of the next target point, the robot is indicated to not reach the next target point, if the robot does not reach the next target point, the remaining arrival time is obtained, and the remaining arrival time and the remaining distance information are sent to the operation side device (for example, a handheld terminal), wherein the remaining arrival time is determined according to the remaining distance of the robot and the preset moving speed of the robot.

Further, after the guniting operation is completed, acquiring the actual guniting distance of the robot, and judging whether the actual guniting distance is the same as the preset total guniting distance;

if the two are the same, the robot is indicated to finish the guniting operation;

If the current pose information and the single guniting distance are different, determining the pose of the next target point position, and repeating the steps until the preset guniting total distance is reached. It can be appreciated that the autonomous navigation technology of the robot ensures that the robot can efficiently and accurately complete the guniting task in the whole guniting operation process, and reduces manual intervention.

Furthermore, the industrial personal computer can also store the map which the robot walks through, so that the later use is convenient.

As shown in fig. 3, a flow chart of a guniting robot according to an embodiment of the application is provided. The industrial personal computer can also acquire a single-shot guniting distance and a preset guniting total distance, wherein the single-shot guniting distance and the preset guniting total distance are input by a user at a front end interface when the robot is started, and it can be understood that the single-shot guniting distance and the guniting total distance are the basis of autonomous navigation and operation planning of the robot, and determine the coverage range of each guniting of the robot and the total working area to be covered. The pose of the next target point comprises the position of the next target point and the pose of the next target point, the position of the next target point is obtained according to the positioning of the robot and the single guniting distance, the process of determining the position of the next target point comprises the step of adding the current position and the single guniting distance to obtain the position of the next target point, and the pose of the next target point is given by people to ensure that the robot is parallel to two sides of a roadway.

The foregoing is only a partial embodiment of the present application, and it should be noted that it will be apparent to those skilled in the art that modifications and adaptations can be made without departing from the principles of the present application, and such modifications and adaptations are intended to be comprehended within the scope of the present application.

Claims (6)

1. The autonomous navigation method for the unmanned guniting robot is characterized by comprising a hardware part and a software part, wherein the hardware part comprises a laser radar, an IMU (inertial measurement unit), an industrial personal computer and a hydraulic controller, the software part comprises a map building module, a path planning module and an upper computer, the software part runs on the industrial personal computer, and the path planning module comprises a global planner and a local planner, wherein:

Step S1, the laser radar is used for acquiring point cloud data;

the IMU is used for acquiring the gesture of the robot and calculating the inter-frame odometer information through pre-integration;

Step S2, the industrial personal computer is used as a control center and is responsible for receiving and analyzing the point cloud data, the gestures and the inter-frame odometer information, constructing a real-time environment map, acquiring the robot positioning and planning the optimal path, simultaneously communicating with the upper computer, acquiring the pose of the next target point location, completing the receiving and transmitting of control instructions,

The map construction module is used for constructing a real-time environment map by utilizing the point cloud data, the gestures and the inter-frame odometer information, sending the environment map and the positioning information to the path planning module, and simultaneously calculating the odometer information for the local planner in a multi-sensor fusion mode;

The path planning module is used for generating path information and a speed instruction for controlling the movement of the robot based on the pose of the next target point position, the environment map and the obstacle data fed back by the laser radar, wherein the path information comprises a global path and a local path;

and step S3, the hydraulic controller is used for receiving the speed control instruction sent by the path planning module and converting the speed control instruction into a control signal of the proportional control valve so as to realize the movement of the guniting robot.

2. An autonomous navigation method for an unmanned guniting robot according to claim 1, wherein the obstacle data comprises static obstacle information and dynamic obstacle information,

The path planning module is used for generating path information based on the pose of the next target point, the environment map and obstacle data fed back by the laser radar, and comprises the following steps:

the global planner is used for generating the global path based on the pose of the next target point position and the static obstacle information;

the local planner is configured to generate the local path based on the global path and the dynamic obstacle information.

3. The autonomous navigation method for an unmanned guniting robot according to claim 1, wherein the environment map comprises position and contour of an obstacle and passable area information, and the path planning module is used for generating path information based on the pose of the next target point, the environment map and the obstacle data, and further comprises information for controlling the movement capability of the robot, wherein the movement capability information comprises full-scale displacement capability and obstacle avoidance capability;

Determining a desired speed based on the obstacle and passable area information, the full square displacement capability, and the obstacle avoidance capability, and generating a speed command based on the desired speed;

Correspondingly, the hydraulic controller is used for generating a control signal of the proportional control valve based on the speed command so as to drive the hydraulic motor to realize the movement of the robot.

4. An autonomous navigation method for an unmanned guniting robot according to any of claims 1-3, wherein the path planning module functions further comprise:

Comparing whether the current pose is the same as the pose of the next target point in real time, wherein the current pose comprises the pose and a detection position;

if the target points are different, determining that the robot does not reach the next target point, and acquiring the residual arrival time;

Transmitting the residual arrival time and the residual distance information to operation side equipment;

if the target point positions are the same, determining that the robot reaches the next target point position, and sending a reaching instruction to a control system.

5. An autonomous navigation method for an unmanned guniting robot according to claim 4, further comprising, after sending an arrival command to a control system:

And the control system generates a laser radar protective cover closing instruction and executes the guniting operation.

6. An autonomous navigation method for an unmanned guniting robot according to claim 5, further comprising, after said performing a guniting operation:

acquiring the actual guniting distance of the robot, and judging whether the actual guniting distance is the same as the preset guniting total distance;

If not, determining pose information of the next target point according to the current pose information and the single guniting distance, and repeating guniting operation until the preset guniting total distance is completed.