JP6793905B2 - Robot behavior simulation device - Google Patents

Description

The present invention relates to a robot behavior simulation device, and more particularly to, for example, a robot behavior simulation device operating in a human coexistence environment.

An example of a conventional simulation device of this type is disclosed in Patent Document 1. Patent Document 1 discloses a robot simulator that can set an off-limits space according to the position of a human body.

Further, Non-Patent Document 1-3 also discloses a conventional simulation device of this type. In these simulation devices of Non-Patent Documents 1-3, a person in the simulator is operated and moved by a keyboard.

Japanese Patent Application Laid-Open No. 2009-90400 [B25J 19/06, 13/08 G05B 19/4061, 19/18] G. Echeverria and N. Lassabe and A. Degroote and S. Lemaignan, Modular OpenRobots Simulation Engine: MORSE, Proceedings of the IEEE ICRA, 2011. G. Echeverria, et al., Simulating Complex Robotic Scenarios with MORSE, SIMPAR2012, pp. 197-208, 2012. https://www.openrobots.org/wiki/morse/

In both the robot simulators of Patent Document 1 and Non-Patent Document 1-3, human beings are treated only as "human-shaped objects" and human behavior is not considered, so they are like communication robots. It was insufficient as a simulation device for robots operating in a human coexistence environment.

Therefore, the main object of the present invention is to provide a novel robot behavior simulation device.

Another object of the present invention is to provide a robot behavior simulation device that considers human behavior.

The present invention has adopted the following configuration in order to solve the above problems. The reference numerals and supplementary explanations in parentheses indicate the correspondence with the embodiments described later in order to help the understanding of the present invention, and do not limit the present invention in any way.

The first invention is a robot behavior simulation device that simulates behavior in a virtual simulation space of a robot that operates in an environment coexisting with humans, such as a communication robot, according to a robot behavior determination program, in the virtual simulation space. A human agent position presentation unit that presents position data indicating the position of a human agent, a robot position presentation unit that presents position data indicating the position of a robot in the virtual simulation space, and an action pattern that a person takes when the robot enters the field of view. A human agent generation unit and a robot position that generate one or more human agents in the virtual simulation space according to a preset anti-robot behavior pattern database and human agent position data presented by the human agent position presentation unit. The current robot position in the virtual simulation space according to the robot position data presented by the presentation unit , and the position of the person agent in the virtual simulation space according to the position data presented by the person agent position presentation unit. based on the data of the action pattern indicated by the pair robot behavior pattern database, the robot at the next step in the human agent action determining unit that determines the behavior of the person agent in the next step, and before Symbol virtual simulation space This is a robot behavior simulation device including a robot behavior simulation unit that determines the behavior of the robot according to the robot behavior determination program .

In the first invention, the simulation device (10: a reference code exemplifying a corresponding portion in the embodiment; the same applies hereinafter) is a virtual simulation space ( 10: a virtual simulation space of a robot such as a communication robot that operates in an environment coexisting with humans). The behavior in 200) is simulated according to the robot action determination program . The simulation device (10) includes a human agent position presentation unit (20) that presents position data indicating the position of the human agent in the virtual simulation space, and a robot position presentation unit (20) that presents position data indicating the position of the robot in the virtual simulation space. 20), and an anti-robot behavior pattern database (16) in which the behavior patterns that a person takes when the robot enters the field of view are preset. A human agent generation unit (18, S3) is included, and this human agent generation unit generates one or more human agents in the virtual simulation space (200) according to the position data of the human agent presented by the human agent position presentation unit. Generate. The human agent action determination unit (18, S5) follows the current robot position in the virtual simulation space according to the robot position data presented by the robot position presentation unit , and the position data presented by the human agent position presentation unit. Based on the position of the human agent in the virtual simulation space and the behavior pattern data shown by the robot behavior pattern database, the behavior of the human agent in the next step is determined. The robot behavior simulation unit (22, S7) determines the robot behavior in the next step in the virtual simulation space according to the robot behavior determination program (24). Then, the robot action determination program (24) is modified according to the simulation result by the simulation device (10), if necessary.

According to the first invention, the behavior of the robot can be simulated by performing the simulation so as to reproduce the behavior of the people around the robot in the virtual simulation space, so that the robot does not actually move in the real space. It is possible to efficiently develop a robot action decision program.

The second invention is subordinate to the first invention, and further includes a person pattern database in which patterns of people such as adults, children, men, and women for a person agent to be input to the virtual simulation space are registered in advance, and the person agent. The generation unit is a robot behavior simulation device that generates a human agent with a human pattern given from a human pattern database.

In the second invention, the person pattern database (14) presets the shape of a person, that is, the shape (pattern) of a person such as an adult, a child, a man, or a woman. A human agent is generated in the virtual simulation space according to the human pattern given from this database (14).

According to the second invention, various patterns of human agents can be introduced into the virtual simulation space.

The third invention is a robot behavior simulation device that is subordinate to the first or second invention and further includes a physics engine unit that realizes the behavior of each agent according to a physical law.

In the third invention, the physics engine unit (26) realizes the behavior of each agent according to a physical law or a physical force. For example, if there is no interference with other obstacles, structures or other agents, each agent will behave as intended, and if interference occurs, each agent will follow the laws of physics (law of mechanics). The actual movement of is processed.

According to the third invention, consistent simulation results can be obtained.
The fourth invention is subordinate to the third invention, and further comprises a sensor output simulation unit that executes a simulation of sensors arranged in a virtual simulation space and saves sensor values according to the actions of each agent by the physics engine unit. It is a robot behavior simulation device.
In the fourth invention, the sensor output simulation unit (28) executes simulation of an image sensor or a distance sensor arranged in the virtual simulation space (200).

According to the present invention, the behavior of the robot can be simulated by simulating the behavior of the person around the robot, so that the robot does not actually move in the real space and coexists with the person in the real space. It is possible to efficiently develop an action decision program for an active robot.

The above-mentioned object, other object, feature and advantage of the present invention will become more apparent from the detailed description of the following examples made with reference to the drawings.

FIG. 1 is a block diagram showing a simulation apparatus according to an embodiment of the present invention. FIG. 2 is an illustrated diagram showing an example of a robot to which the present invention is applied. FIG. 3 is a block diagram showing an electrical configuration of the robot shown in FIG. FIG. 4 is an illustrated diagram showing an example of a human agent's behavior model for a robot. FIG. 5 is a schematic diagram showing another example of a human agent's behavior model for a robot. FIG. 6 is an illustrated diagram showing an example of the virtual simulation environment in the embodiment of FIG. FIG. 7 is an illustrated diagram showing an example of movement of a human agent. FIG. 8 is a flow chart showing the operation of the simulation device of the embodiment of FIG. FIG. 9 is a flow chart showing an example of an operation when measuring the performance of the robot in the simulation device of the embodiment of FIG.

With reference to FIG. 1, the simulation apparatus 10 of this embodiment is composed of hardware such as an ASIC (application specific integrated circuit) or a DSP (digital signal processor), but is like a program incorporated in a general-purpose computer. Software may be used. FIG. 1 is a functional block diagram of such a simulation device 10.

The robot targeted by the simulation device 10 is a robot 12 that operates in a coexistence environment with humans, such as the communication robots shown in FIGS. 2 and 3. Since such a robot 12 is affected by the behavior of a person existing in the environment, the simulation device 10 of this embodiment simulates the behavior of the human according to the behavior determination program of the robot. Also by simulating, we support the development of robot behavior decision programs for safe robot 12.

Here, prior to the description of the simulation device 10, the configuration of the robot 12 will be described with reference to FIGS. 2 and 3. The robot 12 includes a carriage 30, and two wheels 32 and one trailing wheel 34 for autonomously moving the robot 12 are provided on the lower surface of the carriage 30. The two wheels 32 are independently driven by a wheel motor 36 (see FIG. 3), and the carriage 30, that is, the robot 12 can be moved in any direction in the front-rear, left-right, and left-right directions.

A cylindrical sensor mounting panel 38 is provided on the trolley 30, and a large number of distance sensors 40 are mounted on the sensor mounting panel 38. These distance sensors 40 measure the distance to an object (such as a person or an obstacle) around the robot 12 by using, for example, infrared rays or ultrasonic waves.

A body 42 is provided upright on the sensor mounting panel 38. Further, the above-mentioned distance sensor 40 is further provided in the upper front center of the body 42 (position corresponding to the chest of a person), and measures the distance mainly to a person in front of the robot 12. Further, the body 42 is provided with a support column 44 extending from substantially the center of the upper end portion on the back surface side thereof, and an omnidirectional camera 46 is provided on the support column 44. The omnidirectional camera 46 captures the surroundings of the robot 12 and is distinguished from the eye camera 70 described later. As the omnidirectional camera 46, a camera using a solid-state image sensor such as a CCD or CMOS can be adopted.

Upper arms 50R and upper arms 50L are provided at the upper ends of both side surfaces of the body 42 (positions corresponding to human shoulders) by the shoulder joints 48R and shoulder joints 48L, respectively. Although not shown, the shoulder joint 48R and the shoulder joint 48L each have three orthogonal degrees of freedom. That is, the shoulder joint 48R can control the angle of the upper arm 50R around each of the three orthogonal axes. One axis (yaw axis) of the shoulder joint 48R is an axis parallel to the longitudinal direction (or axis) of the upper arm 50R, and the other two axes (pitch axis and roll axis) are orthogonal to the axis from different directions. It is the axis to do. Similarly, the shoulder joint 48L can control the angle of the upper arm 50L around each of the three orthogonal axes. One axis (yaw axis) of the shoulder joint 48L is an axis parallel to the longitudinal direction (or axis) of the upper arm 50L, and the other two axes (pitch axis and roll axis) are orthogonal to the axis from different directions. It is the axis to do.

Further, an elbow joint 52R and an elbow joint 52L are provided at the tips of the upper arm 50R and the upper arm 50L, respectively. Although not shown, the elbow joint 52R and the elbow joint 52L each have a degree of freedom of one axis, and the angles of the forearm 54R and the forearm 54L can be controlled around the axis of this axis (pitch axis).

A hand 56R and a hand 56L corresponding to a human hand are provided at the tips of the forearm 54R and the forearm 54L, respectively. These hands 56R and 56L are configured to be openable and closable, although detailed illustration is omitted, so that the robot 12 can grip or hold an object using the hands 56R and 56L. However, the shapes of the hands 56R and 56L are not limited to the shapes of the embodiments, and may have shapes and functions that closely resemble those of human hands.

Further, although not shown, the front surface of the trolley 30, the portion corresponding to the shoulder including the shoulder joint 48R and the shoulder joint 48L, the upper arm 50R, the upper arm 50L, the forearm 54R, the forearm 54L, the hand 56R and the hand 56L, respectively. , A contact sensor 58 (shown comprehensively in FIG. 3) is provided. The contact sensor 58 on the front surface of the trolley 30 detects the contact of a human 16 or another obstacle with the trolley 30. Therefore, when the robot 12 comes into contact with an obstacle during its own movement, the robot 12 can detect the contact and immediately stop driving the wheels 32 to suddenly stop the movement of the robot 12. In addition, the other contact sensor 58 detects whether or not the respective parts have been touched.

A neck joint 60 is provided at the upper center of the body 42 (a position corresponding to a person's neck), and a head 62 is provided above the neck joint 60. Although not shown, the neck joint 60 has three degrees of freedom, and the angle can be controlled around each of the three axes. A certain axis (yaw axis) is an axis that goes directly above the robot 12 (vertically upward), and the other two axes (pitch axis, roll axis) are axes that are orthogonal to each other in different directions.

A speaker 64 is provided on the head 62 at a position corresponding to a person's mouth. The speaker 64 is used by the robot 12 to communicate with people around it by voice. Further, a microphone 66R and a microphone 66L are provided at positions corresponding to human ears. Hereinafter, the right microphone 66R and the left microphone 66L may be collectively referred to as a microphone 66. The microphone 66 captures ambient sounds, especially those of the person with whom communication is to be performed. Further, a right eyeball portion 68R and a left eyeball portion 68L are provided at positions corresponding to the human eye. The right eyeball portion 68R and the left eyeball portion 68L include a right eye camera 70R and a left eye camera 70L, respectively. Hereinafter, the right eyeball portion 68R and the left eyeball portion 68L may be collectively referred to as an eyeball portion 68. Further, the right eye camera 70R and the left eye camera 70L may be collectively referred to as an eye camera 70.

The eye camera 70 captures a face of a person approaching the robot 12, other parts or objects, and captures a corresponding video signal. In this embodiment, the robot 12 detects the line-of-sight directions (vectors) of the left and right eyes of the person by the video signal from the eye camera 70. Specifically, the line-of-sight detection method includes a method using two cameras and a method using one camera, but since the details are not important here, detailed description thereof will be omitted.

Further, as the eye camera 70, the same camera as the omnidirectional camera 46 described above can be used. For example, the eye camera 70 is fixed in the eyeball portion 68, and the eyeball portion 68 is attached to a predetermined position in the head 62 via an eyeball support portion (not shown). Although not shown, the eyeball support portion has two degrees of freedom, and the angle can be controlled around each of these axes. For example, one of these two axes is an axis (yaw axis) in the upward direction of the head 62, and the other is orthogonal to one axis and orthogonal to the front side (face) of the head 62. This is the axis (pitch axis) in the direction of yaw. By rotating the eyeball support portion around each of these two axes, the tip (front) side of the eyeball portion 68 or the eye camera 70 is displaced, and the camera axis, that is, the line-of-sight direction is moved. The installation positions of the speaker 64, the microphone 66, and the eye camera 70 are not limited to the relevant parts, and may be provided at appropriate positions.

As described above, the robot 12 of this embodiment has independent 2-axis drive of the wheels 32, 3 degrees of freedom of the shoulder joint 48 (6 degrees of freedom on the left and right), 1 degree of freedom of the elbow joint 52 (2 degrees of freedom on the left and right). It has a total of 17 degrees of freedom, including 3 degrees of freedom for the neck joint 60 and 2 degrees of freedom for the eyeball support (4 degrees of freedom on the left and right).

FIG. 3 is a block diagram showing an electrical configuration of the robot 12. With reference to FIG. 3, the robot 12 includes a CPU 80. The CPU 80, also called a microcomputer or processor, is connected to a memory 84, a motor control board 86, a sensor input / output board 88, and a voice input / output board 90 via a bus 82.

The memory 84 includes a ROM, an HDD, and a RAM, although not shown. The robot action determination program for controlling the operation of the robot 12 is stored in the ROM and the HDD in advance. However, the action determination program of this type of robot has not yet been installed in the robot 12 simulated by the simulation device 10. That is, the simulation device 10 is a support device for developing and designing an optimum action determination program for such a robot 12.

The motor control board 86 is composed of, for example, a DSP, and controls the drive of each axis motor such as each arm, neck joint 60, and eyeball portion 68. That is, the motor control board 86 receives control data from the CPU 80 and controls the angles of the two axes of the right eyeball portion 68R (in FIG. 3, collectively referred to as “right eyeball motor 92”). Control the rotation angle. Similarly, the motor control board 86 receives control data from the CPU 80 and controls the angles of the two axes of the left eyeball portion 68L (in FIG. 3, collectively referred to as “left eyeball motor 94”). ) Rotation angle is controlled.

Further, the motor control board 86 receives control data from the CPU 80, and includes three motors that control the angles of the three orthogonal axes of the shoulder joint 48R and one motor that controls the angles of the elbow joint 52R, for a total of four. The rotation angle of one motor (collectively referred to as "right arm motor 96" in FIG. 3) is controlled. Similarly, the motor control board 86 receives control data from the CPU 80, and has three motors that control the angles of the three orthogonal axes of the shoulder joint 48L and one motor that controls the angles of the elbow joint 52L. The rotation angles of a total of four motors (collectively referred to as "left arm motor 98" in FIG. 3) are controlled.

Further, the motor control board 86 receives control data from the CPU 80 and controls the angles of the three orthogonal axes of the neck joint 60 (in FIG. 3, collectively referred to as “head motor 100”). Control the rotation angle of. Then, the motor control board 86 receives the control data from the CPU 80 and controls the rotation angles of the two motors (collectively referred to as “wheel motor 36” in FIG. 3) for driving the wheels 32.

A hand actuator 108 is further coupled to the motor control board 86, and the motor control board 86 receives control data from the CPU 80 and controls the opening and closing of the hands 56R and 56L.

Similar to the motor control board 86, the sensor input / output board 88 is composed of a DSP, and takes in signals from each sensor and gives them to the CPU 80. That is, data regarding the reflection time from each of the distance sensors 40 is input to the CPU 80 through the sensor input / output board 88. Further, the video signal from the omnidirectional camera 46 is input to the CPU 80 after performing predetermined processing on the sensor input / output board 88 as needed. The video signal from the eye camera 70 is also input to the CPU 80 in the same manner. Further, signals from the plurality of contact sensors 58 (collectively referred to as “contact sensor 58” in FIG. 3) described above are given to the CPU 80 via the sensor input / output board 88. Similarly, the voice input / output board 90 is also composed of a DSP, and a voice or a voice according to the voice synthesis data given from the CPU 80 is output from the speaker 64. Further, the voice input from the microphone 66 is given to the CPU 80 via the voice input / output board 90.

Further, the CPU 80 is connected to the communication LAN board 102 via the bus 82. The communication LAN board 102 is composed of, for example, a DSP, and gives transmission data given by the CPU 80 to the wireless communication device 104, and the wireless communication device 104 transmits the transmission data to a server (not shown) or the like via a network. .. Further, the communication LAN board 102 receives data via the wireless communication device 104 and gives the received data to the CPU 80.

Returning to FIG. 1, the simulation apparatus 10 includes two databases 14 and 16. The person pattern database 16 is a database in which the shape of a person, that is, the shape (pattern) of a person such as an adult, a child, a man, or a woman is preset for a person agent to be input to the virtual simulation space of the simulation device 10. is there. In addition to these four basic patterns, it is possible to add the shape (pattern) of a person carrying a rucksack, the shape of a person when pulling a cart, and the like.

The person pattern database 14 may hold different data sets for each day of the week and time so that the simulation can be performed by designating the day of the week and time. Then, you can simulate various things such as what happens when you move the robot at a certain time.

The anti-robot behavior pattern database 16 is a database in which behavior patterns that a person performs on a robot in an environment are preset. In other words, the anti-robot behavior patterns that a person can actually take in real space are registered.

For example, as an action pattern that a person takes when the robot enters his / her field of view, a pattern of interacting with the robot and a pattern of observing the robot shown in FIGS. 4 and 5 are registered in advance as an example.

FIG. 4 shows an action pattern in which when a person (indicated by “i”) finds a robot 12 in his / her field of view, he / she stops around the robot 12 at a stop distance Dstop from the robot 12 and interacts with the robot 12. Illustrated.

In FIG. 5, when a person (indicated by “i”) finds the robot 12 in his / her field of view, the robot stops around the robot 12 with an observation distance Dobserve slightly larger than the stop distance Dstop. The behavioral pattern of observing 12 is illustrated.

In the anti-robot behavior pattern database 16, many anti-robot behavior patterns including such behavior patterns are pre-registered. The simulation device 10 simulates that the human agent behaves according to one of the registered behavior patterns.

For example, in addition to the standard human behavior patterns against robots as shown in FIGS. 4 and 5, data on the behavior patterns of human agents who perform strange or difficult behaviors may be set in this database 16. .. For example, set a difficult anti-robot behavior pattern that is common, such as trying out the reaction of the robot (trying around), bullying (persistent behavior), talking all the time, running when you see the robot, etc. You may try to simulate according to such a special behavior pattern.

In the virtual simulation space, the behavior of agents changes under the influence of other agents (for example, avoiding people from colliding with each other, approaching robots when they see them, etc.).

For example, as shown in FIG. 6, two people i and j may influence each other to determine the movement position in the next step. Since the person agent i originally moves in the direction of di, j. However, since the human agent j is trying to move in the direction of the angle θi, j with respect to the human agent i at a velocity vi, j, the human agent i will eventually move in the direction of d'i, j. ..

Further, the human agent's behavior toward the robot may be set to change depending on the interaction between the human agents. For example, when a human agent talks to a robot, the probability that another human agent will approach the robot increases, and when the preceding human agent is influenced by the robot (getting a leaflet, being guided to a store, etc.), the subsequent person agent It is also possible to set in this database 16 so that the probability that the human agent is also affected increases.

The human simulation unit 18 includes human pattern data received from the human pattern database 14, behavior pattern data received from the robot behavior pattern database 16, map data of the virtual simulation space received from the environment information presentation unit 20, humans and robots. According to the data such as the position of, the behavior of the human agent in the virtual simulation space (virtual space) in the next step is simulated.

In this way, the environment information presentation unit 20 simulates the behavior of the robot, such as map data indicating the virtual simulation space (virtual space) in which the behavior of the robot should be simulated, and the positions of human agents and the robot in the virtual simulation space. Present the data needed to do this.

FIG. 7 shows an example of the virtual simulation environment (space) as described above, and the virtual space 200 assumes a closed space partitioned by a wall 202 provided with an entrance 204 and an exit 206. However, the pillar 208, the desk 210, and the like are written in the map data as obstacles. Further, although this virtual space is drawn in two dimensions in FIG. 7, it is actually displayed as a three-dimensional image on the display 29 (FIG. 1).

The robot behavior simulation unit 22 simulates the behavior of the robot in the next step according to the robot behavior determination program 24 based on the data such as the position of the robot received from the environment information presentation unit 20. Here, the action determination program 24 is a program that determines the behavior of the robot by controlling the motors 36, 92-106 and the actuator 108 in response to the sensor inputs of the sensors 40, 46, 58, 70 and the like shown in FIG. Is. As an example, the robot action determination program 24 is set to move according to the movement path 212 from the entrance 204 until the robot 12 enters the virtual space 200 and exits the exit 206, for example, in FIG. 7. In FIG. 7, reference numeral 214 indicates a human agent put into the virtual space 200 for simulation.

The physics engine 26 calculates what kind of movement path each agent follows based on physical forces, laws, and the like. More specifically, the physics engine 26 executes a process of executing the movement intended by each agent (including the robot) in the virtual simulation space 200 (FIG. 7). At this time, if there is no interference with other obstacles, structures or other agents, the intended movement is generated. However, if there is interference, it processes the actual movement of each agent according to the laws of physics (laws of mechanics).

The sensor output simulation unit 28 executes simulation of an image sensor such as a camera or a distance sensor such as a laser range finder (LRF) arranged in the virtual simulation space 200. In the case of a camera image, optical calculation is performed from the camera viewpoint to generate image information for each camera viewpoint. In the case of a distance sensor, the distance is measured from the sensor in each direction by imitating a scan by a laser, and the result is saved as a sensor value.

The simulation device 10 is provided with a display 29, and the display 29 displays an image of a virtual simulation space as shown in FIG. 7, for example.

With reference to FIG. 8, the simulation apparatus 10 shown in FIG. 1 is set to move on the movement path 212 shown in FIG. 7 as an example in the first step S1, the robot action determination program 24. From the environment information presentation unit 20, the map data of the virtual simulation space as shown in FIG. 7 that simulates the behavior of the robot 12, and the parameters (appearance frequency, movement route, group) related to the generation of the human agent to be input to the virtual simulation space. The number of people, the ratio of adults and children and men and women, height, etc.) and the position data of each agent (including the robot 12) are read.

In the next step S3, one or more human agents are generated in the virtual simulation space according to the above-mentioned parameters regarding the generation of the human agent presented by the environment information presentation unit 20 and the human pattern data from the human pattern database 14. And place it. Since various internal patterns are registered in the human pattern database 14 and human agents are generated in the virtual simulation space based on the human patterns given from the internal patterns, various human agents can be generated.

The parameters related to the generation of the human agent in step S3 may be set based on the actual observation data.

In step S5, the human simulation unit 16 is next to the human agent based on the current positions of the human agent and the robot 12 indicated by the environmental information presentation unit 20, and based on the anti-robot behavior pattern indicated by the anti-robot behavior pattern database 16. Determine what to do in the steps. The type and frequency of actions that act on the robot 12 here may be set based on actual observation data. The behavior pattern registered in the anti-robot behavior pattern database 16 is an anti-robot behavior pattern that a person can actually take in real space, and the simulation is performed according to such an behavior pattern, so that the simulation result is a valid result. ..

In step S7, the robot behavior simulation unit 22 determines the behavior of the robot 12 in the next step according to the robot behavior determination program 24 based on the current positions of the human agent and the robot 12 indicated by the environment information presentation unit 20. ..

Specifically, in step S7, when the robot 12 is a robot whose main purpose is to go to the destination, such as a transfer robot or a boarding robot, it is possible to consider another person from the current position and movement speed of other people. Performs a process of calculating a movement route that does not interfere with the movement of the robot. Further, when the robot 12 is a robot whose purpose is to provide services to people, for example, a person who is stopped near the robot in a simple case in order to approach a person who may be interested in the robot. Choose to go towards that person, choose a place where people tend to gather, wait for people to come in such a place, or get too close to a place where people pass so as not to cause congestion. It performs processing such as preventing it from occurring.

In the case of this step S7, since the action of the person in the next step in step S5 is determined, the above-mentioned process is executed in consideration of it. Therefore, it is possible to judge in the virtual simulation space whether or not the action determination program of the robot that operates in the real space coexisting with humans is appropriate.

After that, in step S9, the action of each agent is realized in the physics engine 26. More specifically, in step S9, as described above, the process of executing the movement intended by each agent (including the robot) in the virtual space 200 is performed, and the process is performed with other obstacles, structures, and other agents. Without any interference, each agent works as intended. For example, if the agent tries to take one step forward, the agent takes one step forward as intended.

However, when there is interference, the actual movement of each agent is processed according to the laws of physics (law of mechanics). In a simple case, for example, if you try to move toward a wall, it will collide with the wall, you will not be able to move forward any further, and you will receive a reaction force according to a constant coefficient of restitution. In the case of a collision between agents, the forward movement and the movement due to the reaction force overlap, for example, considering a simple example, if a collision occurs from the side while moving forward, it will actually move diagonally forward. ..

In the next step S11, the sensor simulation unit 24 updates the environmental information, the output information of the sensor, and the like as described above.

Then, it is determined whether or not to end the simulation in step S13, and if the simulation is to be continued, the process returns to the previous step S3 again.

In this embodiment, the behavior of the robot is simulated by simulating the behavior of the person around the robot, so that the robot coexists with the person without actually moving it in the real space. It is possible to efficiently develop an action determination program 24 for a robot that operates in space.

The action of the human agent acting on the robot in step S5 described above depends on the appearance and behavior of the robot. The dependency may be stored in advance as a database (eg, when the color is white, people approach twice as often). Also, by editing this parameter, what would happen if the robot could double the number of people (that is, if the design could be changed that way) (sales increase, overcrowding, or safety? , Etc.), may be able to simulate.

Further, in the above-described embodiment, only one robot 12 moves in the virtual simulation space, but a plurality of robots may move at the same time. In this case, each robot acts according to a different action determination program 24.

In the above embodiment, each person agent is automatically determined by the person simulation unit 18 in the next step in steps S3 and S5, but when a large number of person agents are simulated, one of them ( Alternatively, the person who operates the simulation device 10 may operate and move the simulation device 10. In addition, sensor data generated as a result of human movement in an actual environment may be input (synthesized). For example, you can debug the result of your movement in a virtual simulation.

A part of the sensor data simulated by the sensor output simulation unit 28 in the simulation device 10 described above, for example, a voice spoken by the user, may be reproduced by reproducing the actually measured sensor data.

In the simulation device 10 of this embodiment, for the human agent in the virtual simulation space, a human pattern database 14 in which a human shape (pattern) model is registered in advance is provided, and attributes such as adults and children, men and women, etc. are matched from the human pattern database 14. The data was put into the virtual simulation space, but a mechanism was set up to replace this data with the data of the people who actually measured it, for example, many people have a backpack, winter clothes, summer clothes, etc. May be good.

As the output of the simulation device 10, data (sensor data and labeled data) useful for various pattern recognition may be output. For example, if a large amount of data is output as to whether or not there is a person in the field of view of the robot's range image sensor, it is possible to realize a recognition program that detects the person by a method such as deep learning.

In the above embodiment, the robot action determination program 24 has been described as programming all the series of actions of the robot 12 in the virtual simulation space, but when only a part of the series of actions is simulated. However, this simulation device 10 can be used. That is, the robot action determination program 24 may be a program that determines at least one action of the robot, and the simulation device 10 gives an index for determining the quality of such a program.

The simulation device 10 of the embodiment can also be used to measure the performance of the robot 12. As an example in FIG. 9, the movement efficiency (average speed) when the robot 12 moves from the point A (for example, the entrance 202) to the point B (for example, the exit 204) of the virtual simulation space 200 (see FIG. 7). The operation of the simulation device 10 in the case of measuring (N = 1000 when simulating 1000 times to obtain the average) is shown.

In the first step S21 of FIG. 9, the variable n for counting the number of simulations and v_sum indicating the total velocity are initialized (n = 0, v_sum = 0). However, the number of simulations is counted by a separately provided counter (not shown), and the total speed v_sum is accumulated by a separately provided buffer (not shown).

Then, in the next step S23, it is determined whether or not the number of simulations n has reached the predetermined number of times N (n> N?).

If "NO" in step S23, the process proceeds to step S25, and the initial value is set in the simulation device 10. The current time is set to t0, and the position of the robot 12 is set to point A (entrance 202). Then, the number of simulations is incremented (n = n + 1).

In step S27, the simulation is executed until the robot 12 reaches the point B (exit 204).

In step S29, the moving speed v = AB / (t1-t0) of the robot 12 in this simulation is calculated from the simulation end time t1 in the simulation device 10. However, AB is the distance from point A to point B.

In step S31, the total velocity v_sum is updated in the buffer (v_sum + v).

Then, in step S23, when it is determined that the number of simulations n has reached the specified number of times N, the average velocity v_avg is calculated in the next step S33 (v_avg = v_sum / N). The moving efficiency (average moving speed) of the robot 12 simulated in this way from the point A to the point B can be an index for the safe operation of the robot 12 in the human coexistence environment.

If the simulation device 10 of the embodiment is used, in addition to the above-mentioned average moving speed, as an index related to safety, the number of times the robot has touched a person and the person approaches within a certain distance (for example, 50 cm) from the robot. As an index related to the number of times and performance, the number of people who talked to the robot and the number of people who visited the store after hearing the advertisement of the robot can also be used.

The robot capable of simulating the behavior of the simulation device 10 is not limited to the robot 12 shown in FIGS. 3 and 4 described in the embodiment. This simulation device can also be applied to communication robots of other types and structures.

10 ... Simulation device 12 ... (Communication) Robot 14 ... Human pattern database 16 ... Anti-robot behavior pattern database 18 ... Human simulation department 20 ... Robot behavior simulation department 22 ... Environmental information presentation department 24 ... Robot behavior decision program 26 ... Physics Engine 28 ... Sensor output simulation unit

Claims (4)

A robot behavior simulation device that simulates behavior in a virtual simulation space of a robot that operates in an environment that coexists with humans, such as a communication robot, according to a robot behavior determination program.
A person agent position presentation unit that presents position data indicating the position of a person agent in the virtual simulation space.
A robot position presentation unit that presents position data indicating the position of a robot in the virtual simulation space.
An anti-robot behavior pattern database that presets the behavior patterns that a person takes when a robot enters the field of view,
A human agent generation unit that generates one or more human agents in the virtual simulation space according to the position data of the human agent presented by the human agent position presentation unit.
The current position of the robot in the virtual simulation space according to the robot position data presented by the robot position presentation unit, and the human agent in the virtual simulation space according to the position data presented by the human agent position presentation unit. The human agent behavior determination unit that determines the behavior of the human agent in the next step based on the position and the behavior pattern data indicated by the robot behavior pattern database, and the said in the next step in the virtual simulation space. A robot behavior simulation device including a robot behavior simulation unit that determines a robot's behavior according to the robot behavior determination program. A human pattern database in which patterns of adults, children, men, women, etc. for human agents to be input to the virtual simulation space are registered in advance is further provided, and the human agent generation unit is given from the human pattern database. The robot behavior simulation device according to claim 1, which generates a human agent with a human pattern. The robot behavior simulation device according to claim 1 or 2, further comprising a physics engine unit that realizes the behavior of each agent according to a physical law. The behavior simulation of the robot according to claim 3 , further comprising a sensor output simulation unit that executes a simulation of sensors arranged in the virtual simulation space and saves sensor values according to the behavior of each agent by the physics engine unit. apparatus.