JP2019079096A - State improvement device, state improvement method, and control program - Google Patents

Abstract

To enable a stimulus according to an operator's self-confidence degree of state estimation.SOLUTION: A state improvement device comprises: a state estimation unit 202 for estimating a driver's state; a self-confidence degree calculation unit 203 for calculating a self-confidence degree of the driver's state estimated by the state estimation unit 202; and a stimulus control unit 208 for giving an improvement stimulus for improving the driver's state estimated by the state estimation unit 202. The stimulus control unit 208 provides a stimulus in which a discomfort to be given to the driver becomes smaller among the improvement stimuli according to the lower self-confidence degree calculated by the self-confidence degree calculation unit 203, while providing a stimulus in which the effect of improvement of the driver's state becomes greater among the improvement stimuli according to the higher self-confidence degree calculated by the self-confidence degree calculation unit 203.SELECTED DRAWING: Figure 2

Description

  The present disclosure relates to a state improvement device, a state improvement method, and a control program for improving the state of a worker.

  Patent Document 1 discloses a technique for estimating the condition of a worker and providing the worker with a stimulus for optimizing the ability required to carry out the work. For example, Patent Document 1 estimates the driver's state from face images, biological information, etc., and outputs music or aroma to optimize sleepiness, concentration, and stress necessary for driving. Technology for releasing air conditioning and generating air conditioning wind is disclosed.

JP, 2016-88497, A

  However, it is difficult to estimate the worker's state with an accuracy of 100% due to individual differences in the correlation between the information used to estimate the state and the actual state. If the worker's state estimation result is not correct, even though the improvement effect on the worker's state is low, the worker may be highly offensive to perform a stimulus and may give the worker high discomfort. is there.

  One object of the present disclosure is to provide a state improvement device, a state improvement method, and a control program that enable stimulation in accordance with the worker's degree of confidence in state estimation.

  The above object is achieved by a combination of the features of the independent claims, and the subclaims define further advantageous embodiments of the disclosure. The reference numerals in parentheses described in the claims indicate the correspondence with specific means described in the embodiments described later as one aspect, and do not limit the technical scope of the present disclosure. .

  In order to achieve the above object, a first state improvement device of the present disclosure includes a target determination unit (204) that determines a target value of a worker's state, and a state estimation unit (202) that estimates a worker's state And a confidence calculation unit (203) that calculates the confidence of the worker's condition estimated by the condition estimation unit, and a stimulation control unit that performs improvement stimulus for improving the worker's condition estimated by the condition estimation unit If the degree of confidence calculated by the degree of confidence calculation unit is less than the threshold value, the stimulus control unit determines the target value determined by the target determination unit and the state of the worker estimated by the state estimation unit. Improve stimulation regardless of whether there is a gap between

  In order to achieve the above object, the first state improvement method of the present disclosure determines a target value of the state of the worker, estimates the state of the worker, and calculates the confidence of the state of the worker to be estimated. If the degree of confidence to be calculated is less than the threshold value, to improve the state of the worker to be estimated regardless of whether there is a divergence between the target value to be determined and the state of the worker to be estimated. Make an improvement stimulus.

  In order to achieve the above object, a first state improvement program of the present disclosure includes a computer, a target determination unit (204) that determines a target value of a state of an operator, and a state estimation unit that estimates the state of the operator (202), the confidence calculation unit (203) that calculates the confidence of the worker's state estimated by the condition estimation unit, and the target determination if the confidence calculated by the confidence calculation unit is less than the threshold value The improvement stimulus for improving the condition of the worker estimated by the condition estimation unit is performed regardless of whether there is a divergence between the target value determined by the unit and the condition of the worker estimated by the condition estimation unit It functions as a stimulation control unit (208) that

  According to this, the degree of confidence of the state of the worker to be estimated is calculated, and if the degree of confidence to be calculated is less than the threshold value, between the target value of the state of the worker to be determined and the state of the worker to be estimated Regardless of whether or not there is a gap between the Therefore, the stimulation according to the worker's state estimation confidence is made possible.

  In order to achieve the above object, a second state improvement apparatus of the present disclosure calculates a state estimation unit (202) that estimates the state of the worker, and a degree of confidence of the state of the worker estimated by the state estimation unit The degree of confidence calculation unit (203) and a stimulation control unit (208) for performing improvement stimulus for improving the state of the worker estimated by the state estimation unit, the stimulation control unit is calculated by the confidence degree calculation unit If the degree of confidence is greater than or equal to the threshold, the improvement stimulus is performed with a stronger stimulus than when the degree of confidence is less than the threshold.

  In order to achieve the above object, the second state improvement method of the present disclosure estimates the state of the worker, calculates the degree of confidence of the state of the worker to be estimated, and the degree of confidence calculated is greater than or equal to the threshold In this case, a stimulus that is stronger than that in the case where the degree of confidence is less than the threshold causes an improvement stimulus for improving the condition of the worker to be estimated.

  In order to achieve the above object, a second state improvement program of the present disclosure includes a computer, a state estimation unit (202) that estimates a state of an operator, and a degree of confidence of the state of the operator estimated by the state estimation unit When the confidence degree calculated by the confidence degree calculation unit (203) to calculate the degree and the confidence degree calculation unit is equal to or higher than the threshold, the stimulus is estimated by the state estimation unit with a stronger stimulus than when the confidence degree is less than the threshold Function as a stimulation control unit (208) that performs improvement stimulation to improve the condition of the worker.

  According to this, the degree of confidence of the state of the worker to be estimated is calculated, and when the degree of confidence calculated is equal to or higher than the threshold, the improvement stimulus is performed with a stronger stimulus than in the case where the degree of confidence is less than the threshold It will Therefore, stimulation according to the degree of confidence of the state estimation of the worker is made possible, and stronger stimulation is performed to improve when it is considered that the degree of confidence is higher than the threshold and the improvement effect of the state of the worker is higher. It is possible to enhance the effect.

  In order to achieve the above object, the third state improvement apparatus of the present disclosure calculates a state estimation unit (202) that estimates the state of the worker, and a degree of confidence of the state of the worker estimated by the state estimation unit The degree of confidence calculation unit (203) and a stimulation control unit (208) for performing improvement stimulus for improving the state of the worker estimated by the state estimation unit, the stimulation control unit is calculated by the confidence degree calculation unit Change the stimulus to improve depending on your level of confidence.

  In order to achieve the above object, the third state improvement method of the present disclosure estimates the state of the worker, calculates the degree of confidence of the state of the worker to be estimated, and estimates according to the degree of confidence to be calculated. Change the improvement stimulus given to improve the worker's condition.

  In order to achieve the above object, a third state improvement program of the present disclosure includes a computer, a state estimation unit (202) that estimates a state of an operator, and a degree of confidence of the state of the operator estimated by the state estimation unit The degree of confidence calculation unit (203) that calculates the, and the stimulation control unit that changes the improvement stimulus that is performed to improve the state of the worker estimated by the state estimation unit according to the degree of confidence calculated by the degree of confidence calculation unit Function as (208).

  According to this, the degree of confidence of the state of the worker to be estimated is calculated, and the improvement stimulus to be performed to improve the state of the worker to be estimated is changed according to the degree of confidence to be calculated. Therefore, it is possible to enable stimulation according to the degree of confidence of the worker's state estimation, and to perform stimulation with a higher improvement effect depending on the degree of confidence or less discomfort given to the worker? It will be possible to change.

FIG. 1 is a diagram showing an example of a schematic configuration of a support system 1. It is a figure which shows an example of a rough structure of HCU20. It is a figure which shows an example of the driver state estimated from biometric information, and its confidence degree. It is a figure which shows an example of distribution of the confidence degree of the driver state in the broken line of FIG. It is a figure for showing an example of a 1st table. It is a figure for showing an example of a 2nd table. It is a figure which shows an example of the relationship between the improvement effect for every combination of the kind of improvement stimulus, and a driver state, and the discomfort degree given to a driver. It is a figure which shows an example of the kind of improvement stimulus for every driver state in a 1st table. It is a figure which shows an example of the kind of improvement stimulus for every driver state in a 2nd table. It is a flow chart which shows an example of the flow of the improvement stimulus related processing in HCU20.

  DETAILED DESCRIPTION Embodiments of the disclosure will be described with reference to the drawings. Note that, for convenience of explanation, among the plurality of embodiments, portions having the same functions as the portions shown in the figures used in the description so far are given the same reference numerals and may not be described. is there. The description in the other embodiments can be referred to for parts denoted by the same reference numerals.

(Embodiment 1)
<Schematic Configuration of Support System 1>
Hereinafter, the present embodiment will be described using the drawings. The support system 1 shown in FIG. 1 is used in a car (hereinafter simply referred to as a vehicle), and an HMI (Human Machine Interface) system 2, an ADAS (Advanced Driver Assistance Systems) locator 3, a periphery monitoring sensor 4, an air conditioning system 5 , A vehicle state sensor 6, a vehicle control ECU 7, and an automatic driving ECU 8. The HMI system 2, the ADAS locator 3, the periphery monitoring sensor 4, the air conditioning system 5, the vehicle state sensor 6, the vehicle control ECU 7, and the autonomous driving ECU 8 are connected to, for example, an in-vehicle LAN. The vehicle on which the support system 1 is mounted is hereinafter referred to as a vehicle.

  The ADAS locator 3 includes a GNSS (Global Navigation Satellite System) receiver, an inertial sensor, and a map database (hereinafter, DB) storing map data. The GNSS receiver receives positioning signals from multiple satellites. The inertial sensor includes, for example, a gyro sensor and an acceleration sensor. The map DB is a non-volatile memory, and stores map data such as link data, node data, and road shapes. The map data may be configured to include a three-dimensional map including a road shape and a point group of feature points of a structure.

  The ADAS locator 3 sequentially locates the vehicle position of the vehicle equipped with the ADAS locator 3 by combining the positioning signal received by the GNSS receiver and the measurement result of the inertial sensor. In addition, it is good also as a structure using the travel distance etc. which were calculated | required from the pulse signal sequentially output from the wheel speed sensor mounted in the own vehicle for positioning of a vehicle position. Then, the determined vehicle position is output to the in-vehicle LAN. When using a three-dimensional map consisting of a road shape and a point cloud of feature points of a structure as map data, the ADAS locator 3 does not use a GNSS receiver, but this three-dimensional map and the features of the road shape and structure The vehicle position of the vehicle may be specified using detection results of the surrounding area monitoring sensor 4 such as LIDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging) that detects a point cloud of points. The map data may be acquired from a server outside the vehicle via, for example, the in-vehicle communication module.

  The surrounding area monitoring sensor 4 is an obstacle around the vehicle such as moving objects such as pedestrians, non-human animals, bicycles, motorcycles, and other vehicles, falling objects on the road, guardrails, curbs, and stationary objects such as trees. To detect. In addition, road markings such as travel division lines and stop lines around the vehicle are detected. The surrounding area monitoring sensor 4 is, for example, a surrounding area monitoring camera that captures a predetermined range around the vehicle, or a sensor such as a millimeter wave radar, a sonar, or LIDAR that transmits a survey wave to the predetermined range around the vehicle. The surrounding surveillance camera sequentially outputs the captured image to be sequentially captured to the autonomous driving ECU 8 as sensing information. A sensor that transmits a search wave such as a sonar, millimeter wave radar, LIDAR or the like sequentially outputs a scan result based on a reception signal obtained when a reflection wave reflected by an obstacle is received to the autonomous driving ECU 8 as sensing information.

  The air conditioning system 5 is an air conditioning system for a vehicle that acquires, from the HCU 20, air conditioning request information including air conditioning related setting values and the like set by the occupant of the vehicle from the HCU 20, and adjusts the temperature, cleanliness, air flow and the like in the vehicle compartment. . The air conditioning system 5 includes an air conditioning control ECU 50, an air conditioning unit 51, and an aroma unit 52.

  The air conditioner unit 51 generates warm air and cold air supplied to the vehicle compartment from the air outlet provided in the instrument panel or the like. The aroma unit 52 atomizes an aroma oil such as an essential oil containing an aroma component. The aroma component atomized by the aroma unit 52 is mixed with the air flow generated by the air conditioning unit 51 and supplied into the vehicle compartment. The air conditioning control ECU 50 is an electronic control unit that controls the operation of the air conditioning unit 51 and the aroma unit 52, and is connected to the air conditioning unit 51 and the aroma unit 52.

  The vehicle state sensor 6 is a sensor group for detecting various states such as the traveling state and the operation state of the own vehicle. The vehicle state sensor 6 includes a wheel speed sensor that detects the vehicle speed of the vehicle, a steering sensor that detects the steering angle of the vehicle's steering, an accelerator position sensor that detects the opening degree of the accelerator pedal of the vehicle, and a brake of the vehicle There are a brake depression force sensor that detects the depression amount of the pedal, a temperature sensor that detects the room temperature of the vehicle, and the like. The vehicle state sensor 6 outputs the detection result to the in-vehicle LAN. The detection result of the vehicle state sensor 6 may be output to the in-vehicle LAN via the ECU mounted on the own vehicle.

  The vehicle control ECU 7 is an electronic control unit that performs acceleration / deceleration control and / or steering control of the own vehicle. The vehicle control ECU 7 includes a steering ECU that performs steering control, a power unit control ECU that performs acceleration and deceleration control, and a brake ECU. The vehicle control ECU 7 acquires detection signals output from sensors such as an accelerator position sensor, a brake depression force sensor, a steering angle sensor, a vehicle speed sensor and the like mounted on the vehicle, and electronically controls the throttle, a brake actuator, EPS (Electric Power) Steering) outputs a control signal to each traveling control device such as a motor.

  The autonomous driving ECU 8 recognizes the surrounding environment of the own vehicle from the vehicle position and map data of the own vehicle acquired from the ADAS locator 3, the sensing information acquired from the surrounding area monitoring sensor 4, and the like. Further, the autonomous driving ECU 8 generates a travel plan for causing the vehicle to travel by automatic driving based on the recognized surrounding environment. For example, as a medium- and long-term travel plan, a recommended route for directing the vehicle to a destination is generated. Also, as a short-term travel plan, it is decided to execute steering for lane change, acceleration / deceleration for speed adjustment, and steering and braking for obstacle avoidance. Then, the automatic driving ECU 8 causes the vehicle control ECU 7 to automatically perform acceleration, braking, and / or steering of the vehicle in accordance with the generated travel plan, thereby performing automatic driving.

  Besides, it is assumed that the autonomous driving ECU 8 can switch the automation level of the autonomous driving (hereinafter, the automation level). As an example, the automation level may be five levels of automation levels 0 to 4 or the like.

  The automation level 0 is a stage where the driver operates all without automatizing the main control system such as the vehicle's brake, steering, throttle and motive power. In other words, it is a stage of non-operation of automatic driving which does not automatically control any of acceleration, braking, and steering. That is, it is the stage of manual operation. The automation level 1 is a stage of independently executing a function of one of the main control systems of the vehicle. In other words, it is the stage of automation of a specific function that automatically controls either acceleration, braking, or steering.

  The automation level 2 is a stage in which one of the main control systems of the vehicle is combined with an automated function. In other words, it is a stage of automation of multiple functions that automatically controls a plurality of acceleration, braking, and steering. The automation level 3 is a stage in which all the main control systems of the vehicle are automated, and the driver performs the driving operation only when the driver changes to the traffic condition to be driven. In other words, this is a semi-automatic operation stage in which all control of acceleration, braking and steering is automatically performed except in an emergency.

  The automation level 4 is a stage in which all the main control systems of the vehicle are automated, and it is not necessary for the driver to perform the driving operation at any time during traveling. In other words, even in an emergency, it is a fully automatic operation stage that automatically controls all of acceleration, braking, and steering. The automation level 1 to the automation level 4 can also be referred to as an operation phase of automatic driving that automatically controls at least one of acceleration, braking, and steering. The automation level is not limited to the example described above, and may be divided into five or more steps, or may be divided into four or less steps.

  Switching of the automation level in the autonomous driving ECU 8 is performed autonomously in accordance with the termination of the planned traveling of the autonomous driving section, the recognized traveling environment or the sensing failure in the surroundings monitoring sensor 4, etc. do it. In addition, it may be configured to be performed according to the input operation by the driver to the operation device 26.

  The HMI system 2 includes a human machine interface control unit (HCU) 20, a driver status monitor (DSM) 21, a biological sensor 22, a display device 23, an audio output device 24, a vibrator 25, and an operation device 26. The HMI system 2 receives an input operation from the driver, presents information to the driver, and monitors the state of the driver. This driver corresponds to a worker.

  The DSM 21 is configured of a near infrared light source and a near infrared camera, a control unit that controls these, and the like. The DSM 21 is disposed, for example, on the upper surface of the instrument panel in a posture in which the near infrared camera is directed to the driver's seat side of the vehicle. The DSM 21 captures the head of the driver, which has been irradiated with near-infrared light by the near-infrared light source, using a near-infrared camera. The image captured by the near infrared camera is subjected to image analysis by the control unit. The control unit detects, for example, the face direction and / or the gaze direction of the driver from the captured image. The DSM 21 extracts feature amounts of face regions such as the degree of eye opening of the driver and blinks from the captured image, and outputs the extracted feature amounts of the face region to the HCU 20.

  The biometric sensor 22 measures the biometric information of the driver, and sequentially outputs the measured biometric information to the HCU 20. The biometric sensor 22 may be provided on the vehicle such as provided on a steering wheel, a driver's seat or the like, or may be provided on a wearable device worn by a driver. When the biometric sensor 22 is provided in the wearable device worn by the driver, the HCU 20 may acquire the measurement result of the biometric sensor 22 through wireless communication, for example. Examples of biological information measured by the biological sensor 22 include electrocardiogram, respiration, electromyography, electroencephalogram, cerebral blood flow, perspiration, pulse wave, temperature, blood pressure, skin conductance and the like.

  Examples of the display device 23 include a combination meter, a CID (Center Information Display), a HUD (Head-Up Display), and an LED. The combination meter is disposed in front of the driver's seat. The CID is disposed above the center cluster in the vehicle cabin. The combination meter displays various images for information presentation on the display screen of the liquid crystal display based on the image data acquired from the HCU 20. The HUD projects the light of the image based on the image data acquired from the HCU 20 onto the projection area defined on the windshield. The light of the image reflected to the vehicle interior by the windshield is perceived by the driver sitting on the driver's seat. The driver can visually recognize the virtual image of the image projected by the HUD superimposed on the external scenery in front of the host vehicle. The LEDs are disposed at noticeable positions of a driver such as an instrument panel, and the light emission is controlled by the HCU 20.

  The audio output device 24 may be, for example, an audio speaker that outputs a sound, a buzzer that outputs a sound, or the like. The vibrator 25 is provided, for example, at a position where a driver of the vehicle contacts, such as a steering wheel, a seat of a driver's seat, etc., and gives the driver a stimulus by vibration. The vibration of the vibrator 25 is controlled by the HCU 20.

  The operating device 26 is a switch group operated by the driver. For example, as the operation device 26, there are a steering switch provided in a spoke of the steering of the own vehicle, a touch switch integrated with the display device 23 having a display, and the like.

  The HCU 20 includes a processor, a memory, an I / O, and a bus connecting these, and executes various processes by executing a control program stored in the memory. Execution of the control program by the processor corresponds to execution of a method corresponding to the control program. As used herein, memory is a non-transitory tangible storage medium that stores non-transitory computer readable programs and data. In addition, the non-transitional tangible storage medium is realized by a semiconductor memory or a magnetic disk. The HCU 20 corresponds to the state improvement device. Details of the process in the HCU 20 will be described later.

<Schematic Configuration of HCU 20>
Subsequently, a schematic configuration of the HCU 20 will be described with reference to FIG. The HCU 20 includes an information acquisition unit 201, a state estimation unit 202, a confidence degree calculation unit 203, a goal determination unit 204, a divergence determination unit 205, a correspondence storage unit 206, a stimulation determination unit 207, a stimulation control unit 208, and a learning unit 209. It is provided as a functional block. Note that part or all of the functions executed by the HCU 20 may be configured as hardware by one or more ICs or the like. Also, some or all of the functional blocks included in the HCU 20 may be realized by a combination of software execution by a processor and hardware components.

  The information acquisition unit 201 acquires various types of information input to the HCU 20. For example, the information acquisition unit 201 acquires a feature amount extracted by the DSM 21 or acquires biological information measured by the biological sensor 22. Further, the information acquisition unit 201 acquires the operation state detected by the vehicle state sensor 6, acquires the surrounding environment recognized by the automatic driving ECU 8, and acquires the current automation level. The current automation level may be acquired from the autonomous driving ECU 8. In addition, the information acquisition unit 201 acquires operation input information received by the operation device 26.

  The state estimation unit 202 estimates the state of the driver of the own vehicle (that is, the driver state). The state estimation unit 202 estimates the driver state from the feature amount of the face part acquired by the information acquisition unit 201, the biological information, the operation state, and the like, and the classifier. For example, in the case of estimation of the awakening degree, it is possible to use an eye opening degree as the feature amount of the face part, and an electroencephalogram, a pulse wave, a heartbeat or the like as the biological information. Further, as the operation state, it is obtained from the lateral vibration of the vehicle obtained from the position of the traveling division line in the surrounding environment recognized by the automatic driving ECU 8 and the steering angle sequentially detected by the steering angle sensor of the vehicle state sensor 6 It is possible to use the amount of variation of the steering operation that is performed. In addition, it does not matter as a structure using information other than what was mentioned here.

  The classifier is, for example, a probabilistic classifier. In the following, the case of using a naive Bayesian classifier will be described as an example. The naive Bayesian classifier is a probabilistic classifier that outputs hypotheses with the highest posterior probability to the input, using a simple Bayesian probability model, and can be rephrased as a simple Bayesian classifier. The classifier used in the state estimation unit 202 can be reliably estimated in the past as a combination of data such as the feature amount of the face part acquired by the information acquisition unit 201, biological information, operation state, and the driver state as learning data. A classifier is constructed by giving combinations that are generally reliable. As a result, the data such as the feature amount of the face part, the biological information, the operation state and the like acquired by the information acquisition unit 201 is input, and the discriminator outputs the driver state having the highest temporal probability.

  In the following, the case where the state estimation unit 202 estimates the awakening degree as the driver state will be described as an example. In the present embodiment, the case where the awakening degree is divided into six sleepiness levels of 0 to 5 and estimated will be described as an example. The drowsiness levels divided into six levels are, in order from high to low awakeness, drowsiness level “0” which seems not to be completely sleepy (in other words, awake state), slightly sleepy drowsiness level “1”, sleepy The sleepiness level is “2”, the quite sleepy sleepiness level is “3”, the very sleepy sleepiness level is “4”, and the sleepy (in other words, low arousal state) sleepiness level is “5”.

  The degree of confidence calculation unit 203 calculates the degree of confidence of the driver state estimated by the state estimation unit 202. Here, the degree of confidence of the driver state will be described with reference to FIGS. 3 and 4. FIG. 3 is a diagram showing an example of the driver state estimated from the biological information and the degree of confidence thereof. FIG. 3 shows an example in the case of using the alpha wave content of brain waves as the biological information, and the vertical axis indicates the estimated driver state, and the horizontal axis indicates the alpha wave content. Further, black circles in FIG. 3 indicate learning data, solid lines indicate estimated values of the driver state, and shading indicate the degree of confidence.

  The darker part in FIG. 3 indicates that the probability density function has a higher value, that is, a part where the estimated function is likely to pass there, and has a higher degree of confidence. Also, the lighter color portion in FIG. 3 indicates that the probability density function has a lower value, that is, a portion where the estimated function is less likely to pass therethrough, and has a lower degree of confidence. Since the probability density is not a probability, it can take one or more values.

  FIG. 4 is a view showing an example of the distribution of the degree of confidence of the driver state in the broken line in FIG. In FIG. 4, the vertical axis represents the degree of confidence in estimation of the driver state, and the horizontal axis represents the driver state estimated. The solid line in FIG. 4 shows the distribution of confidence when the density of learning data is high and the confidence is high, and the dotted line shows the distribution of confidence when the density of learning data is low and the confidence is low. ing.

Since the discriminator is a probabilistic classifier, the value of the probability density function, that is, the value of the driver state with the highest degree of confidence is estimated by the state estimation unit 202 as the current driver state. Thus, the degree of confidence calculation unit 203 calculates the value of the probability density function of the driver state estimated by the state estimation unit 202 as the degree of confidence. Here, an example of calculation of the degree of confidence in the case of using a naive Bayesian classifier as a classifier will be described. When using a naive Bayesian classifier as the classifier, the value of the posterior probability density distribution of the following equation 1 may be calculated as the degree of confidence. In equation 1, θ is the driver state, p (θ) is the prior confidence as a prior probability, p (x | X ⁽ⁿ⁾ ) is the new confidence as the posterior probability density distribution, X ⁽ⁿ⁾ is for learning It shows a set of data.

Further, in the case of using a classifier using the bootstrap method, the confidence level may be calculated as follows, as an example. When a classifier using a bootstrap method is used, the step of estimating the driver state as a new sample is repeated B times, and the estimated values of B samples according to equation 2, the standard deviation according to equation 3, and the bias according to equation 4 Ask for Then, the probability density distribution is calculated as the degree of confidence by Equation 5.

  The target determination unit 204 determines a control target that is a target value of the driver state. The target determination unit 204 may determine, as a control target, a value closest to the current driver state estimated by the state estimation unit 202 among the required driver states which are currently the optimum driver state.

  As for the necessary driver state, a value set in advance may be determined according to the current automation level, the operation state, and / or the surrounding environment acquired by the information acquisition unit 201. As an example of the preset value, when the target driver state is the awakening degree, a lower sleepiness level may be set in advance as the automation level is lower. In addition, as the driving load estimated from the operation state and the surrounding environment becomes higher, a lower drowsiness level may be set in advance. The value set in advance may be configured to acquire one stored in advance in the non-volatile memory of the HCU 20, or may be acquired from a server outside the vehicle via the in-vehicle communication module.

  The required driver state may have a width such as, for example, drowsiness levels "0" to "1". For example, in the case where the required driver state is a sleepiness level “0” to “1” and the current driver state estimated by the state estimation unit 202 is a sleepiness level “3”, the goal determination unit 204 determines the sleepiness level “ "1" will be determined as the control target.

  The divergence determination unit 205 determines whether or not there is a divergence between the target value determined by the target determination unit 204 and the driver state estimated by the state estimation unit 202. For example, the divergence determination unit 205 calculates the difference between the current driver state value estimated by the state estimation unit 202 and the target value determined by the target determination unit 204. Then, when the difference is not zero, it is determined that there is a difference, and when the difference is zero, it may be determined that there is no difference. For example, when the difference is -1, it is determined that there is a divergence. When the degree of confidence calculated by the degree-of-confident calculating unit 203 is less than the threshold described later, the processing load in the HCU 20 is reduced by not performing the process itself in the divergence determining unit 205. Good.

  The correspondence relationship storage unit 206 stores the correspondence relationship in which the driver state and the target value, and the type of the stimulus for improving the driver state (hereinafter, improvement stimulus) are associated with the degree of confidence of the driver state estimation. doing. As an example, the correspondence relationship may be a data structure such as a table or a map. In the present embodiment, the correspondence storage unit 206 stores the correspondence (hereinafter, first table) used when the degree of confidence is higher and the correspondence (hereinafter, second table) used when the degree of confidence is lower. The following explanation is given assuming that

  Examples of the improvement stimulus include sound, voice, light, characters, pictures, aroma, vibration, air conditioning wind and the like. As the sound, there are music output from an audio speaker of the audio output device 24, an alarm sound output from a buzzer of the audio output device 24, and the like. As the voice, there is voice from the system side in the voice dialogue system which is outputted from the audio speaker of the voice output device 24 or the like. The light includes, for example, light emission of an LED of the display device 23. Examples of characters and images include characters and images displayed on a combination meter, CID, HUD and the like of the display device 23. The aroma includes an aroma having a waking effect generated from the aroma unit 52, and the like. As the vibration, vibration by the vibrator 25 can be mentioned. Examples of the air conditioning wind include cold air generated from the air conditioning unit 51 and the like. Ameliorating stimuli include not only stimuli that ameliorate a deteriorating driver condition, but also stimuli that ameliorate (i.e., improve) a driver condition that is unnecessarily good, such as relaxing a state that is too tense.

  Here, an example of the first table and the second table will be described with reference to FIGS. 5 and 6. FIG. 5 is a diagram for illustrating an example of the first table, and FIG. 6 is a diagram for illustrating an example of the second table. The alphabet of FIG. 5 and FIG. 6 has shown the kind of improvement stimulus.

  As shown in FIGS. 5 and 6, the improvement stimulus is associated with each combination of the driver state and the target value in any of the first table and the second table. Note that the improvement stimulus associated with one combination of the driver state and the target value may include a plurality of types of improvement stimuli. As one example, a plurality of types of improvement stimuli such as “cold air 5 ° C.” and “vibration 10 V” may be associated with one combination of the driver state and the target value.

  With regard to any of the first table and the second table, different improvement stimuli may be associated with each combination of the driver state and the target value. The different improvement stimuli referred to here may be different types of improvement stimuli or may be different types of improvement stimuli. In addition, it may be an improvement stimulus that differs in type and intensity. In any of the first table and the second table, it is preferable that a stronger improvement stimulus be associated as the degree of deviation between the driver state and the target value increases. The strength of the improvement stimulus is not limited to the intensity, but may be the length of stimulation time, the frequency of stimulation, the stimulation to a higher stimulation site of the improvement effect, the higher stimulation pattern of the improvement effect, etc. .

  Furthermore, in the first table and the second table, different improvement stimuli are associated even if the combination of the driver state and the target value is the same. The different improvement stimuli referred to here may be different types of improvement stimuli or may be improvement stimuli of different strengths. In addition, it may be an improvement stimulus that differs in type and intensity. As an example of different types, the difference between "cold air" and "vibration" can be mentioned. As an example when intensity differs, the difference of "cold air -5 ° C" and "cold air 5 ° C", etc. are mentioned. The difference in the strength of the improvement stimulus is not limited to the configuration in which the intensity differs, but may include configurations in which the stimulation time, the stimulation frequency, the stimulation site, the stimulation pattern, and the like differ.

  Further, in the first table and the second table, even if the combination of the driver state and the target value is the same, the improvement stimulus having a higher improvement effect is associated with the first table, and the second table Is associated with an improvement stimulus that causes less discomfort to the driver. In other words, it can be said that the first table is the effect standard and the second table is the discomfort standard.

  Here, an example of the improvement stimulus in the 1st table of an effect reference and the 2nd table of a discomfort reference | standard is demonstrated using FIGS. 7-9. FIG. 7 is a diagram showing an example of the relationship between the improvement effect for each combination of the type of improvement stimulus and the driver state and the degree of discomfort given to the driver. FIG. 8 is a diagram showing an example of the type of improvement stimulus for each driver state in the first table when the target value is “0”. FIG. 9 is a diagram showing an example of the type of improvement stimulus for each driver state in the second table when the target value is “0”. In FIG. 8 and FIG. 9, for the sake of convenience, the example in the case where the target value is “0” is shown for simplification of the description.

  In the example of FIG. 7, for the combination of the improvement stimulus "dialogue" and the sleepiness levels "0" to "5", the improvement effect "0.5" and the discomfort degree "0.3" are obtained, and the improvement stimulus " The combination of cold wind and drowsiness levels "1" to "2" has an improvement effect of "0.8" and a degree of discomfort of "0.5". In addition, for the combination of the improvement stimulus "music" and the drowsiness levels "0" to "1", the improvement effect "0.2" and the discomfort degree "0.2" are obtained, and the improvement stimulus "aroma" and drowsiness The combination with the levels "2" to "4" has the improvement effect "0.5" and the discomfort degree "0.5". And about the combination of improvement stimulus "alarm sound" and drowsiness levels "4"-"5", it is improvement effect "0.8" and discomfort degree "0.8".

  As shown in FIG. 8, in the first table of the effect criteria, the improvement stimulus is not associated with the combination of the sleepiness level “0” and the target value “0” in which there is no divergence between the driver state and the target value. On the other hand, for combinations in which there is a divergence between the driver state and the target value, higher improvement stimulations of the improvement effect are associated for each sleepiness level based on the relationship of FIG. 7. In the relationship of FIG. 7, the highest improvement effect is "0.8" of the improvement stimulus "cold wind" and "alarm sound". Therefore, the improvement stimulus "cold wind", " Corresponds to the alarm sound. Specifically, the improvement stimulus "cold wind" is associated with the drowsiness levels "1" to "2", and the improvement stimulus "alarm sound" is associated with the drowsiness levels "4" to "5". With regard to the sleepiness level "3", the improvement stimulus "dialogue" and "aroma" having the highest improvement effect among the corresponding improvement stimuli are associated.

  As shown in FIG. 9, in the second table of the discomfort level criteria, the improvement stimulus with a smaller degree of discomfort is associated with each drowsiness level based on the relationship of FIG. In the relationship of FIG. 7, the least in the degree of discomfort is “0.2” of the improvement stimulus “music”, so the improvement stimulus “music” is associated with the corresponding drowsiness levels “0” to “1”. . Subsequently, the next to the improvement stimulus "music" is "0.3" of the improvement stimulus "dialogue" which has the least discomfort degree, so among the corresponding drowsiness levels "0" to "5", the improvement stimulus " The improvement stimulus "dialogue" is associated with the sleepiness levels "2" to "5" to which the music is not associated.

  The above-described association may be performed in advance, for example, at the time of manufacture based on the relationship shown in FIG. Further, the relationship shown in FIG. 7 is stored in the electrically rewritable non-volatile memory of the HCU 20, and is updated while being sequentially updated according to learning by the learning unit 209 described later. The correspondence relationship stored in may be updated sequentially.

  Although the example in the case of changing the kind of improvement stimulus by 1st table and 2nd table was shown in the example demonstrated using FIGS. 7-9, it does not necessarily restrict to this. For example, when changing the strength of the improvement stimulus in the first table and the second table, or when changing the combination of the type and strength of the improvement stimulus, the first table of the effect criteria and the degree of discomfort from the same viewpoint It may be configured to be associated with the reference second table.

  Note that the stronger the improvement stimulus, the higher the improvement effect and the greater the discomfort given to the driver. Therefore, even if the combination of the driver state and the target value is the same, the first table is stronger than the second table. It can be reworded that the improvement stimulus is associated, and the second table is associated with the improvement stimulus weaker than the first table. For example, the improvement stimulus weaker than the first table is a kind of improvement stimulus having a weaker intensity, an improvement stimulus having the same type but a weak intensity, an improvement stimulus having a shorter stimulus time, an improvement stimulus having a lower stimulation frequency than the table 1 , A stimulus for improving the pattern starting from a weaker stimulus.

  The stimulus determination unit 207 calculates the driver state estimated by the state estimation unit 202, the target value determined by the target determination unit 204, and the confidence degree calculation unit 203 based on the correspondence relationship stored in the correspondence relationship storage unit 206. Determine the improvement stimulus corresponding to the combination of confidence level. Here, the case of using the first table and the second table described above as the correspondence stored in the correspondence storage unit 206 will be described as an example.

  If the degree of confidence calculated by the degree of confidence calculation unit 203 is equal to or greater than the threshold, the stimulus determination unit 207 determines an improvement stimulus using the first table of the effect criteria, while if the degree of confidence is less than the threshold A second table of discomfort criteria is used to determine improvement stimuli. The threshold referred to here may be a value which can be arbitrarily set as long as it can be said that the estimation of the driver state is likely to be correct. After selecting whether to use the first table or the second table depending on the degree of confidence, the stimulus determination unit 207 selects the driver state estimated by the state estimation unit 202 and the target value determined by the target determination unit 204 in the selected table. It is determined as an improvement stimulus to carry out an improvement stimulus corresponding to the combination of

  A specific example is shown below using FIGS. 5 to 6. For example, when the degree of confidence is equal to or higher than the threshold value and the drowsiness level is estimated to be “4” and the target value is determined to be “0”, the improvement stimulus for implementing the improvement stimulus “U” in the first table of FIG. Decide. On the other hand, when the target value "3" is determined, the improvement stimulus "X" in the first table of FIG. 5 is determined as the improvement stimulus to be implemented. In addition, if the degree of confidence is less than the threshold, and the drowsiness level is estimated to be “4” and the target value is determined to be “0”, the improvement stimulus that implements the improvement stimulus “BC” in the second table of FIG. Decide.

  As in the case where the improvement stimulus "dialogue" and "aroma" are associated with the combination of the sleepiness level "3" and the target value "0" shown in FIG. If there are, it may be determined that all of them should be selected and implemented simultaneously, or it may be determined that any of them be selected and implemented alternately.

  The stimulation control unit 208 generates the improvement stimulus of the type determined by the stimulation determination unit 207 from the stimulation device that generates the improvement stimulus. As a stimulation device, there are a display device 23, an audio output device 24, a vibrator 25, an air conditioner unit 51, and an aroma unit 52. The stimulation control unit 208 may control the operation of the air conditioner unit 51 and the aroma unit 52 by outputting air conditioning request information to the air conditioning control ECU 50.

  The learning unit 209 is, among various information acquired by the information acquisition unit 201, information corresponding to the driver's reaction to the improvement stimulus performed by the stimulation control unit 208, and estimation of a new driver state by the state estimation unit 202. The driver response such as the result is used to learn the improvement effect and / or the degree of discomfort of the improvement stimulus. For example, the learning unit 209 may learn the improvement effect of the improvement stimulus depending on how much the estimation result of the new driver state in the state estimation unit 202 has improved as compared to before the improvement stimulus. Further, the learning unit 209 may learn the degree of discomfort of the driver with respect to the improvement stimulus based on the degree of discomfort of the driver indicated by, for example, the feature amount extracted by the DSM 21 among the various types of information acquired by the information acquiring unit 201. Then, the learning unit 209 corresponds to the correspondence relationship stored in the correspondence relationship storage unit 206 based on the result of learning the improvement effect and / or the degree of discomfort of the improvement stimulus for each driver, and the driver in the state estimation unit 202 The classifiers used to estimate the state are updated for each driver.

  For example, with regard to the correspondence stored in the correspondence storage unit 206, the correspondence shown in FIG. 7 is updated according to the improvement effect and the degree of discomfort by learning the relation shown in FIG. do it. For the improvement effect and the improvement stimulus that was learned to have a higher degree of discomfort than before, the value of the improvement effect and the degree of discomfort in FIG. 7 is increased while the improvement effect and the degree of discomfort are lower than before. The improvement effect and the degree of discomfort in FIG. 7 may be updated to a smaller value.

  In addition, for a discriminator used to estimate the driver state, for example, when the learned improvement effect is equal to or more than a predetermined value that can be arbitrarily set, data used to determine the improvement stimulus as the target of the learning May be updated by adding it as data for learning which has been able to reliably estimate the driver state.

<Improvement stimulus related processing in HCU 20>
Subsequently, an example of the flow of processing (hereinafter, improvement stimulus related processing) related to control for causing the HCU 20 to perform improvement stimulus will be described using the flowchart in FIG. 10. The process of the improvement stimulus related processing corresponds to the state improvement method. In the flowchart of FIG. 10, for example, the HCU 20 may be turned on and started when the ignition power of the vehicle is turned on. It is also possible to start as a condition that the setting for permitting generation of the improvement stimulus via the operation device 26 is on. In addition, it may be configured to start on condition that an operation input for requesting start of generation of improvement stimulus via the operation device 26 is received.

  First, in step S1, the information acquisition unit 201 detects the feature quantity extracted by the DSM 21, the biological information measured by the biological sensor 22, the operation state detected by the vehicle state sensor 6, the surrounding environment recognized by the automatic driving ECU 8, current automation It acquires various information input to the HCU 20, such as the level.

  In step S2, the state estimation unit 202 estimates the current driver state from the feature amount of the face part acquired by the information acquisition unit 201, the biological information, the operation state, and the like, and the discriminator. Further, in S2, the confidence degree calculation unit 203 calculates the confidence degree of the driver state estimated by the state estimation unit 202. Furthermore, in S2, the target determination unit 204 determines the necessary driver state in accordance with the current automation level, the operation state, and / or the surrounding environment acquired by the information acquisition unit 201.

  In step S3, the target determining unit 204 determines a value closest to the current driver state estimated in S2 among the required driver states determined in S2 as a target value. Further, in S3, the divergence determination unit 205 determines whether or not there is a divergence between the target value determined by the target determination unit 204 and the driver state estimated in S2.

  In step S4, when the degree of confidence calculated in S2 is equal to or more than the above-mentioned threshold (YES in S4), the process proceeds to step S5. On the other hand, if the degree of confidence is less than the threshold (NO in S4), the process proceeds to step S8. In step S5, when it is determined that the deviation between the target value determined by the target determination unit 204 and the driver state estimated in S2 is "0" and there is no divergence in S3 (YES in S5), the process proceeds to step S6. Move. On the other hand, when it is determined that there is a divergence (NO in S5), the process proceeds to step S7.

  In step S6, if it is the end timing of the improvement stimulus related process (YES in S6), the improvement stimulus related process is ended. On the other hand, if it is not the end timing of the improvement stimulus related process (NO in S6), the process returns to S1 and repeats the process. An example of the end timing of the improvement stimulus related process is when the setting for permitting generation of the improvement stimulus via the operation device 26 is turned off when the ignition power supply of the own vehicle is turned off.

  In step S7, based on the first table stored in the correspondence storage unit 206, the stimulation determination unit 207 implements an improvement stimulus corresponding to the combination of the driver state estimated in S2 and the target value determined in S3. Determined as an improvement stimulus. Then, the stimulation control unit 208 causes the improvement stimulation determined by the stimulation determination unit 207 to be implemented, and proceeds to step S9. On the other hand, in step S8, based on the second table stored in the correspondence relationship storage unit 206, the stimulus determination unit 207 determines the improvement stimulus corresponding to the combination of the driver state estimated in S2 and the target value determined in S3. Determined as the improvement stimulus to be implemented. Then, the stimulation control unit 208 causes the improvement stimulation determined by the stimulation determination unit 207 to be implemented, and proceeds to S9.

  In step S9, among the various types of information acquired by the information acquisition unit 201, the learning unit 209 corresponds to the information corresponding to the driver's reaction to the improvement stimulus performed by the stimulation control unit 208, and the new information from the state estimation unit 202. Get driver responses such as driver status estimation results. In step S10, the learning unit 209 learns the improvement effect and / or the degree of discomfort of the improvement stimulus using the driver response acquired in S9. Then, based on the learning result, the correspondence stored in the correspondence storage unit 206 and the classifier used for estimating the driver state in the state estimation unit 202 are updated according to each driver, and the process proceeds to S6. .

  In the flowchart of FIG. 10, if it is not the end timing of the improvement stimulus related process in S6, the process returns to S1 to repeat the process, and the configuration is always executed when the end timing of the improvement stimulus related process is not S6. For example, it may be implemented once in several minutes. In addition, when it is set as the structure implemented once in several minutes, the various information acquisition of the information acquisition part 201 in S1 should just be taken as the structure which acquires the information for past several minutes collectively.

<Summary of Embodiment 1>
According to the configuration of the first embodiment, the confidence degree calculation unit 203 calculates the confidence degree of the driver state estimated by the state estimation unit 202, and when the confidence degree to be calculated is less than the threshold, the stimulus determination unit 207 determines. Regardless of whether or not there is a deviation between the target value of the driver state to be estimated and the driver state to be estimated, the improvement stimulus is to be performed. Therefore, the stimulation according to the confidence degree of the driver's state estimation is enabled. Also, since this improvement stimulus is a weaker stimulus than when the confidence level is higher than the threshold value, the improvement stimulus is performed, so it is considered that the confidence level is less than the threshold and the improvement effect of the driver condition is lower. If it does, it is possible to perform a weaker stimulation to reduce the discomfort given to the driver even if the state estimation result is off.

  Further, according to the configuration of the first embodiment, when the confidence degree calculation unit 203 calculates the confidence degree of the driver state estimated by the state estimation unit 202 and the confidence degree to be calculated is equal to or more than the threshold, the confidence degree is the threshold The stimulation that is stronger than when it is less than will cause improvement stimulation. Therefore, stimulation according to the degree of confidence of the driver's state estimation is enabled, and when the degree of confidence is equal to or higher than the threshold and the improvement effect of the driver state is considered to be higher, stronger stimulation is performed to enhance the improvement effect. It becomes possible.

  In addition, according to the configuration of the first embodiment, as the degree of confidence in the estimation of the state of the driver increases, the stimulation with a higher improvement effect is performed, while the degree of confidence decreases. The discomfort that it gives will cause less irritation. Therefore, while it is possible to allow an unpleasant improvement stimulus and prioritize the improvement effect when the degree of confidence is high, but when the degree of confidence is low, discomfort is given to the driver even when the state estimation result is out. It is possible to reduce and to enable improvement stimulation as needed.

  Even when the degree of confidence is low, it may be considered appropriate to use a stimulus with a higher improvement effect or a stronger improvement stimulus to improve the driver's condition, but the driver's discomfort is large. Therefore, in view of the fact that the setting is switched to a setting that does not permit the generation of the improvement stimulus, it is preferable to perform the stimulation with less discomfort for the driver when the degree of confidence is low.

Second Embodiment
In the first embodiment, the configuration in which the driver states are divided into six stages is shown, but the present invention is not necessarily limited to this. The driver state may be divided into two or more stages, or may be a continuous value. When the driver state is a continuous value, the deviation determination unit 205 can arbitrarily set the difference between the current driver state value estimated by the state estimation unit 202 and the target value determined by the target determination unit 204. While it may be determined not to deviate when it is less than a predetermined value, it may be determined to be divergent when it is more than this predetermined value.

(Embodiment 3)
In the above-mentioned embodiment, in addition to the degree of confidence in the estimation of the driver state, the stimulation determination unit 207 shows the configuration for determining the improvement stimulus corresponding to the combination of the current driver state and the target value. Absent. For example, instead of the current driver state, it may be configured to use the tendency of the driver state change from the past. In this case, the correspondence relationship stored in the correspondence relationship storage unit 206 is configured such that the difference between the current driver state and the past driver state is combined with the target value and associated with the improvement stimulus instead of the current driver state. And it is sufficient. Then, based on the correspondence relationship, instead of the current driver state, an improvement stimulus corresponding to a combination of the difference between the current driver state and the target value with respect to the past driver state may be determined.

  According to this, it is possible to determine the improvement stimulus according to the tendency of the driver state change from the past. Therefore, even if the current driver condition is the same, the improvement stimulus with high improvement effect and the improvement stimulus with small discomfort degree are different depending on the tendency of the driver condition change from the past. Depending on the tendency of change, it is possible to perform ameliorating stimulus with higher ameliorating effect and ameliorating stimulus with less discomfort.

(Embodiment 4)
In the above-mentioned embodiment, although the composition which classifies the correspondence stored in correspondence storage part 206 into two correspondence based on high and low degree of confidence was shown, it is not necessarily restricted to this. For example, the correspondence relationship stored in the correspondence relationship storage unit 206 according to the degree of confidence may be classified into three or more. In this case, the correspondence relationship stored in the correspondence relationship storage unit 206 is not only based on the two criteria of height of improvement effect and discomfort degree but also on other criteria such as duration of improvement effect, preference of driver, combination of these, etc. It may be configured as classified. When three or more criteria are used, the correspondence relationship may be a map or the like instead of a table. In addition, when using the preference of the driver as a reference, it is preferable to change the selection of a plurality of types of stimulation used as the improvement stimulus or change the ratio of using the plurality of types of stimulation according to the driver's preference. .

Embodiment 5
In the above embodiment, the HCU 20 includes the learning unit 209, and the correspondence stored in the correspondence storage unit 206 according to the learning result in the learning unit 209, and the identification used to estimate the driver state in the state estimation unit 202. Although the configuration has been shown in which the device is updated for each driver, it is not necessarily limited thereto. For example, the HCU 20 may not include the learning unit 209 and may not update the classifier and the correspondence according to the learning result.

Embodiment 6
In the above embodiment, the driver state is estimated using the classifier, and the value of the probability density function of the estimated driver state is calculated as the degree of confidence, but the present invention is not necessarily limited thereto. For example, the driver state may be estimated without using a classifier, and a value other than the value of the probability density function may be calculated as the degree of confidence. As an example, the driver state is estimated based on whether or not the driver state of a specific segment and the estimated condition are satisfied, and the degree of confidence is calculated according to the number of the driver state and the estimated condition of the specific segment It may be configured as

Seventh Embodiment
In the above-mentioned embodiment, although a waking degree is estimated as a driver state and a case where improvement stimulus for improving a waking degree is performed was mentioned as an example and explained, it does not necessarily restrict to this. For example, the driver states other than comfort and discomfort such as concentration, fatigue, tension, a look, and load may be estimated, and improvement stimuli may be performed to improve the driver states. In this case, a configuration that uses the classifier stored in the correspondence storage unit 206 for determining a driver state other than the awakening degree and the improvement stimulus for improving the driver state other than the awakening degree And it is sufficient.

  In addition, a plurality of types of driver states may be estimated individually, and an improvement stimulus for improving each driver state may be performed. In addition, the HCU 20 may be configured to comprehensively judge the estimation results of a plurality of types of driver states, and perform improvement stimulation only on the driver states requiring improvement.

(Embodiment 8)
Although the above-mentioned embodiment showed composition which HCU20 bears improvement stimulus related processing, it does not necessarily restrict to this. For example, the improvement stimulus related process may be performed by the HCU 20 and another ECU, or the improvement stimulus related process may be performed by another ECU.

(Embodiment 9)
Although the embodiment shows the configuration in which the support system 1 is used in a car, the present invention is not necessarily limited to this. The support system 1 can be used in various mobile bodies. For example, the support system 1 may be used in a vehicle other than a car such as a rail car or a motor bicycle, or in a mobile body other than a vehicle such as an aircraft or a ship. It is good also as composition used. Further, the support system 1 may be configured to be applied to state improvement of an indoor worker such as a house other than a mobile body or a facility.

  The present disclosure is not limited to the above-described embodiment, and various modifications can be made within the scope of the claims, and the present disclosure can be obtained by appropriately combining the technical means disclosed in different embodiments. Embodiments are also included in the technical scope of the present disclosure.

Reference Signs List 1 support system, 2 HMI system, 3 ADAS locator, 4 periphery monitoring sensor, 5 air conditioning control system, 6 vehicle state sensor, 7 vehicle control ECU, 8 autonomous driving ECU, 20 HCU (state improvement device), 21 DSM, 22 living body Sensor, 23 display device, 24 voice output device, 25 vibrator, 26 operation device, 50 air conditioning control ECU, 51 air conditioner unit, 52 aroma unit, 201 information acquisition unit, 202 state estimation unit, 203 confidence calculation unit, 204 target Decision unit, 205 Deviation determination unit, 206 Correspondence storage unit, 207 Stimulus determination unit, 208 Stimulation control unit

Claims (15)

A target determination unit (204) that determines a target value of the state of the worker;
A state estimation unit (202) for estimating the state of the worker;
A confidence degree calculation unit (203) for calculating the confidence degree of the state of the worker estimated by the state estimation unit;
And a stimulus control unit (208) for performing an improvement stimulus for improving the condition of the worker estimated by the condition estimation unit.
When the degree of confidence calculated by the degree-of-confident calculating unit is less than a threshold, the stimulus control unit determines the target value determined by the target determining unit and the state of the worker estimated by the state estimating unit. The state improvement device for performing the improvement stimulus regardless of whether or not there is a gap between the two.   When the degree of confidence calculated by the degree-of-confident calculating unit is less than a threshold, the stimulus control unit determines the target value determined by the target determining unit and the state of the worker estimated by the state estimating unit. The state according to claim 1, wherein the improvement stimulus is performed with a weaker stimulus than when the confidence degree calculated by the confidence degree calculation unit is equal to or more than a threshold value regardless of whether or not there is a divergence between Improvement device. A deviation determination unit (205) is provided that determines whether or not there is a divergence between the target value determined by the target determination unit and the state of the worker estimated by the state estimation unit.
When the stimulus control unit determines that there is a divergence by the divergence determination unit if the confidence degree calculated by the confidence calculation unit is equal to or greater than a threshold, the stimulation degree is less than the threshold. The state improvement device according to claim 1 or 2, wherein the improvement stimulus is performed with a stronger stimulus. A state estimation unit (202) for estimating the state of the worker;
A confidence degree calculation unit (203) for calculating the confidence degree of the state of the worker estimated by the state estimation unit;
And a stimulus control unit (208) for performing an improvement stimulus for improving the condition of the worker estimated by the condition estimation unit.
The stimulation control unit improves the state in which the improvement stimulation is performed with a stronger stimulus when the confidence degree calculated by the confidence degree calculation unit is equal to or higher than a threshold compared to when the confidence degree is less than the threshold. apparatus. A target determination unit (204) configured to determine a target value of the state of the worker;
When the degree of confidence calculated by the degree-of-confident calculating unit is less than a threshold, the stimulus control unit determines the target value determined by the target determining unit and the state of the worker estimated by the state estimating unit. The state improvement device according to claim 4, wherein the improvement stimulus is performed regardless of whether or not there is a divergence between the two. A state estimation unit (202) for estimating the state of the worker;
A confidence degree calculation unit (203) for calculating the confidence degree of the state of the worker estimated by the state estimation unit;
And a stimulus control unit (208) for performing an improvement stimulus for improving the condition of the worker estimated by the condition estimation unit.
The state improvement device, wherein the stimulation control unit changes the improvement stimulation according to the degree of confidence calculated by the degree-of-confident calculating unit. The state of the worker estimated by the state estimation unit is a state other than comfort and discomfort, and
The stimulation control unit performs stimulation with a higher effect of improving the state of the worker among the improvement stimulations, as the degree of confidence calculated by the confidence degree calculation unit increases. 7. The state improvement device according to claim 6, wherein, according to the decrease in the degree of confidence calculated by the degree-of-confident degree calculation unit, stimulation given to the worker with less discomfort among the improvement stimuli is performed. Used in vehicles,
The state of the worker estimated by the state estimation unit is the state of the driver of the vehicle, and
A target determination unit (204) for determining a target value of the state of the driver;
The state of the driver estimated by the state estimation unit based on the correspondence relationship in which the state of the driver, the target value, and the type of the improvement stimulus are associated with each other according to the degree of confidence, and the target A stimulus determination unit (207) for determining the improvement stimulus corresponding to a combination of the target value determined by the determination unit and the confidence degree calculated by the confidence degree calculation unit;
The state improvement device according to claim 6, wherein the stimulation control unit performs the improvement stimulation determined by the stimulation determination unit.   The state improvement apparatus according to any one of claims 1 to 8, wherein the degree of confidence calculation unit calculates the value of the probability density function of the state of the worker estimated by the state estimation unit as the degree of confidence. Determine the target value of the worker's condition,
Estimate the state of the worker,
Calculate the confidence of the operator's condition to be estimated,
If the degree of confidence to be calculated is less than a threshold, the state of the worker to be estimated is improved regardless of whether there is a divergence between the target value to be determined and the state of the worker to be estimated A state improvement method to perform an improvement stimulus to Estimate the worker's condition,
Calculate the confidence of the operator's condition to be estimated,
A state improvement method for performing an improvement stimulus for improving the state of the worker to be estimated with a stronger stimulus than when the confidence degree is less than the threshold when the confidence degree to be calculated is equal to or higher than the threshold . Estimate the worker's condition,
Calculate the confidence of the operator's condition to be estimated,
A state improvement method for changing an improvement stimulus to be performed to improve the state of the worker to be estimated according to the degree of confidence to be calculated. Computer,
A target determination unit (204) that determines a target value of the state of the worker;
A state estimation unit (202) for estimating the state of the worker;
A confidence degree calculation unit (203) for calculating the confidence degree of the state of the worker estimated by the state estimation unit;
When the degree of confidence calculated by the degree-of-confident calculating unit is less than a threshold, there is a divergence between the target value determined by the target determining unit and the state of the worker estimated by the state estimating unit. A control program for functioning as a stimulation control unit (208) that performs improvement stimulus for improving the state of the worker estimated by the state estimation unit regardless of whether or not it is. Computer,
A state estimation unit (202) for estimating the state of the worker;
A confidence degree calculation unit (203) for calculating the confidence degree of the state of the worker estimated by the state estimation unit;
When the degree of confidence calculated by the degree-of-confident calculating unit is equal to or higher than a threshold, the state of the worker estimated by the state estimating unit is improved with a stronger stimulus than when the confidence degree is less than the threshold. A control program to function as a stimulation control unit (208) that performs an improvement stimulus to Computer,
A state estimation unit (202) for estimating the state of the worker;
A confidence degree calculation unit (203) for calculating the confidence degree of the state of the worker estimated by the state estimation unit;
According to the degree of confidence calculated by the degree-of-confident degree calculation unit, it functions as a stimulation control unit (208) that changes an improvement stimulus to be performed to improve the state of the worker estimated by the state estimation unit. Control program.

Images