JP2012130986A - Moving body - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a moving body efficiently performing a given task.SOLUTION: The moving body 1 includes: a movement obstacle detecting means 6, 7 detecting a movement obstacle; an approaching determining means 14 determining whether the movement obstacle approaches the moving body 1 within a prescribed clearance; a standard avoiding operation planning means 15 planning the standard avoiding movement C when determining that the movement obstacle approaches the moving body 1 within the predetermined clearance; a working avoiding operation planning means 16 planning a working avoiding operation E serving for both the avoiding operation of the moving body 1 for avoiding the movement obstacle and the working movement that promotes the movement obstacle to avoid on the basis of the standard avoiding operation C; and a control means 13 controlling the moving body 1 on the basis of the working avoiding operation E. The working avoiding operation planning means 16 plans, as the working avoiding operation E, the avoiding operation in which at least one of the speed of the operation change and a variation amount of the operation change in the standard avoiding operation C is increased.

Description

  The present invention relates to a moving body such as a robot, a manipulator, and mobility.

  Conventionally, Japanese Patent Laid-Open No. 2008-137127 is known as a technical document in such a field. This publication describes a robot control method that restricts the operation of a robot so that it does not come into contact with a human emergency when it is predicted that a contact between an autonomous mobile robot and a human will occur.

Japanese Patent No. 2008-137127

  By the way, based on the principle that the robot is not brought into contact with a human being as much as possible, robot control is performed such that the human operation is given top priority and the robot avoids this or restricts the robot operation. Also in the conventional robot control described above, when contact with a human is predicted, the robot side restricts the operation according to the human.

  However, in the conventional control in which the robot is unilaterally avoided or restricted in operation, a state where there is a high possibility of contact with a human in a congested situation continuously occurs, so the robot performs its task efficiently. It may not be possible.

  Therefore, the present invention can efficiently perform a given task by planning a work avoidance operation that serves as both an avoidance operation in which a moving body avoids a moving obstacle and a work operation that prompts the moving obstacle to avoid. The object is to provide a moving body.

  In order to solve the above-mentioned problem, the moving object according to the present invention is a moving obstacle detecting means for detecting a moving obstacle existing around the moving object, and whether the moving obstacle and the moving object approach within a predetermined interval. Plans a standard avoidance action that is an action that the moving object avoids the moving obstacle when the approach determining means determines whether or not the moving obstacle and the moving object approach within a predetermined interval. Based on the standard avoidance action planning means and the standard avoidance action planned by the standard avoidance action planning means, and the avoidance action in which the moving body avoids the moving obstacle and the action avoidance action that prompts the moving obstacle to avoid And a control means for controlling the moving body based on the work avoidance operation. The work avoidance operation plan means includes a speed change and a motion change in the standard avoidance operation. Wherein the planning as avoidance operation encourage the avoidance operation was increased at least one of the amount of change changes.

  According to the moving body according to the present invention, when a moving obstacle such as a human approaches within a predetermined interval of the moving body, instead of the standard avoidance operation, the obstacle avoidance operation is performed, so that the movement obstacle is simultaneously performed with its own avoidance. Can encourage things to avoid. Thereby, compared with the case where a mobile body performs unilateral avoidance operation | movement, since an efficient passing can be performed, the task given efficiently can be performed. Furthermore, in this moving body, the action is performed by the action avoidance operation with a larger speed and change amount of the operation change in the standard avoidance operation, so it is not necessary to use equipment such as a manipulator or a speaker for the action. Even if the equipment is not provided or the equipment is being used for another work, the work can be realized.

In the moving body according to the present invention, the action avoidance operation planning means avoids the action so that the operation speed is higher than the standard avoidance operation in the initial stage of operation and lower than the standard avoidance operation in the moving obstacle approach stage. It is preferable to plan the operation.
According to the moving body according to the present invention, it is possible to make the moving obstacle easily noticeable by increasing the operation speed and emphasize the movement in the initial operation stage where the moving obstacle is at a distance from the moving obstacle. By reducing the operation speed at the moving obstacle approaching stage that is closest to the object, a highly safe passing can be performed.

In the mobile body according to the present invention, it is preferable that the standard avoidance operation planning means plans an expected avoidance operation expected for the moving obstacle and also plans a standard avoidance operation based on the expected avoidance operation.
According to the moving body according to the present invention, when the moving body and the moving obstacle approach each other, the moving obstacle normally performs the avoiding operation. By planning the standard avoidance operation so as to obtain an avoidance amount considering the expected avoidance operation, an efficient passing between the moving obstacle and the moving object can be achieved.

In the moving body according to the present invention, the expected avoidance operation in which the avoidance operation amount of the moving obstacle is expected by the moving object based on the detection result of the moving obstacle detection unit and the expected avoidance operation planned by the standard avoidance operation planning unit. The control unit further includes an avoidance operation amount determination unit that determines whether or not the amount has reached the amount, and the control unit determines whether the avoidance operation amount determination unit has determined that the avoidance operation amount of the moving obstacle has reached the expected avoidance operation amount. When it is predicted to be clear, it is preferable to control the moving body based on the standard avoidance operation.
According to the moving body according to the present invention, when it is predicted that the avoidance operation amount of the moving obstacle has reached or is expected to reach the expected avoidance operation amount, it is not necessary to perform further action, so avoiding the action. By shifting from the control based on the operation to the control based on the standard avoidance operation, an efficient passing can be performed.

In the mobile body according to the present invention, the work avoidance operation planning unit performs the work avoidance operation in a state where the orientation of the mobile body is maintained so that the avoidance direction expected of the moving obstacle is on the back side of the mobile body. It is preferable to plan.
According to the moving body according to the present invention, when the moving object is avoided, the moving obstacle tries to turn to the back side instead of the front side which is normally considered as the traveling direction of the moving object. By moving in a state where the orientation of the moving body is maintained such that the direction of the moving object is on the back side of the moving body, it is possible to work to guide the avoiding direction of the moving obstacle.

  According to the present invention, a given task can be performed efficiently.

It is a block diagram which shows the structure of 1st Embodiment of the robot which concerns on this invention. It is a schematic plan view for demonstrating the operation | movement interference with a robot and a moving obstruction. It is a schematic plan view for demonstrating the standard avoidance operation | movement by a robot. It is a graph which shows the time change of the moving speed of the robot in standard avoidance operation | movement. It is a schematic plan view for demonstrating the action avoidance operation | movement by a robot. It is a graph which shows the time change of the moving speed of the robot in an approach avoiding operation. It is a graph which shows the other example of the time change of the moving speed of the robot in action avoiding operation | movement. It is a schematic plan view for demonstrating the other example of the work avoidance operation | movement by a robot. It is a graph which shows the time change of the moving speed of the robot in an approach avoiding operation. It is a graph which shows the other example of the time change of the moving speed of the robot in action avoiding operation | movement. It is a schematic plan view for demonstrating the other example of the work avoidance operation | movement by a robot. It is a schematic plan view for demonstrating the other example of the work avoidance operation | movement by a robot. It is a schematic plan view for demonstrating the other example of the work avoidance operation | movement by a robot. It is a flowchart which shows the control procedure of the control apparatus which concerns on 1st Embodiment. It is a block diagram which shows the structure of 2nd Embodiment of the robot which concerns on this invention. It is a schematic plan view for demonstrating the action avoidance operation | movement by the robot of 2nd Embodiment. It is a flowchart which shows the control procedure of the control apparatus which concerns on 2nd Embodiment. It is a schematic plan view for demonstrating the action avoidance operation | movement by the robot of 3rd Embodiment.

  DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In addition, in each figure, the same code | symbol is attached | subjected to the same or an equivalent part, and the overlapping description is abbreviate | omitted.

[First Embodiment]
In the first embodiment, the autonomous mobile robot 1 will be described as a mobile body according to the present invention. The use and form of the robot 1 are not particularly limited. For example, the robot 1 may be a cleaning robot or a transport robot car. Further, the moving method of the robot 1 is not limited, and may be a wheel or bipedal walking.

  In order to achieve a given operation target, the robot 1 performs an avoidance operation on these obstacles when it is determined that an obstacle such as a human being or a building around the robot 1 and the operation of the robot 1 interfere with each other. Here, of the obstacles, movable obstacles such as humans, animals, automobiles, and other robots are referred to as moving obstacles. Hereinafter, a human will be described as an example of a moving obstacle.

  When the robot 1 determines that the movement interferes with a human being who is a moving obstacle, the robot 1 performs the work avoidance action that serves as both the self avoidance action and the action action that prompts the human to avoid the movement. This is a way to realize passing between people and robots.

  FIG. 2 is a schematic plan view for explaining the operation interference between the robot 1 and the human M. An arrow V shown in FIG. 2 represents the moving speed and moving direction of the robot 1, and A represents the movement target trajectory of the robot 1. Similarly, the arrow VM represents the moving speed and direction of the human M, and B represents the motion target trajectory of the human M.

  The robot 1 predicts the motion target trajectory B of the human M from the detection results of the human M by various sensors, thereby determining whether the robot 1 and the human M approach within a predetermined safety margin. As shown in FIG. 2, when the robot 1 and the human M approach each other within a predetermined safety margin, such as when the movement target trajectory A of the robot 1 and the movement target trajectory B of the human M intersect at a short distance, as shown in FIG. If it is determined, it is determined that the operation of the robot 1 and the human M interferes.

  When the robot 1 determines that the human M and the operation of the robot 1 interfere with each other, the robot 1 plans a standard avoidance operation C for avoiding the human M. The standard avoidance operation C is a minimum avoidance operation necessary to avoid the human M, for example.

  Here, FIG. 3 is a schematic plan view for explaining the standard avoidance operation C by the robot 1. D shown in FIG. 3 represents an expected avoidance operation that the robot 1 expects from the human M. Also, t1 to t3 shown in FIG. 3 represent stages of avoidance operations of the robot 1 and the human M. t1 represents the initial stage of the operation when the avoidance operation is started, t2 represents the moving obstacle approach phase where the robot 1 and the human M are closest, and t3 represents the operation end phase where the avoidance operation is completed. FIG. 4 is a graph showing a change with time of the moving speed of the robot 1 in the standard avoidance operation C of FIG.

  As shown in FIG. 3, the robot 1 plans the standard avoidance operation C so that the robot 1 moves as smoothly as possible and moves in a manner that is natural and uncomfortable for humans. Specifically, the robot 1 performs the optimization calculation by setting the speed, acceleration, change of trajectory, energy consumption efficiency, required time, etc. as the cost based on the current movement state and the operation target, so that it is as smooth as possible. Plan standard avoidance action C to move.

  Further, the robot 1 plans the standard avoidance operation C using the stored past human avoidance operation data. The robot 1 extracts parameters such as the trajectory shape, speed, acceleration, and relative distance of the human avoidance motion from the past human avoidance motion data. The standard avoidance operation C is planned so as not to feel a feeling.

  Further, the robot 1 can plan the standard avoidance operation C in consideration of the possibility of collision with the human M, safety at the time of collision, personal space of the human M, and the like. The standard avoidance operation C basically includes a simple trajectory such as a constant velocity motion or a linear motion when no other obstacle exists around the robot 1. The content of the standard avoidance operation C varies depending on the hardware configuration such as the moving mechanism of the robot 1.

  FIG. 5 is a schematic plan view for explaining an action avoiding operation by the robot 1. After planning the standard avoidance action C, the robot 1 plans the avoidance action E that combines the avoidance action of the robot 1 and the action that prompts the human M to expect the avoidance action D. The robot 1 plans the avoidance operation E as an avoidance operation E by increasing at least one of the speed and the amount of change of the operation change in the standard avoidance operation C.

  The robot 1 in FIG. 5 plans a work avoidance operation E in which the trajectory change amount and the speed change amount of the operation change in the standard avoidance operation C are increased. The robot 1 plans an action avoiding operation E in which the amount of change in trajectory and the amount of change in speed at the initial operation stage t1 are increased. In addition to this, the robot 1 can plan the avoidance operation E as an avoidance operation E by increasing the change speed and change amount when changing the course, and the moving speed when changing the course. The robot 1 can also plan an action avoiding operation E in which the course is changed in a plurality of stages.

  In this way, the robot 1 makes the human M aware of the presence of the robot 1 by making the change in the avoidance motion abrupt or larger than that assumed by the human, and the robot 1 recognizes the change in the emphasized motion. By clearly communicating the intention to the person M, the direction of avoidance can be acted on the person M. Then, the action of the robot 1 can guide the human M to the expected avoidance operation expected by the robot 1.

  FIG. 6 is a graph showing the change over time of the moving speed of the robot 1 in the work avoidance operation E shown in FIG. As shown in FIG. 6, the robot 1 has a higher moving speed than the standard avoidance action C in the vicinity of the movement initial stage t1, and a lower moving speed than the standard avoidance action C in the vicinity of the moving obstacle approach stage t2. In this way, the avoidance operation E is planned.

  Further, the robot 1 plans the avoidance operation E so that the state (arrival position, movement speed, arrival time) at the operation end stage t3 matches the standard avoidance operation C. The robot 1 sets the acceleration / deceleration parameters of the avoidance operation E so that the movement advanced ahead of the standard avoidance operation C by the acceleration performed in the operation initial stage t1 is delayed by the subsequent deceleration.

  FIG. 7 is a graph showing another example of the change over time of the moving speed of the robot 1 in the work avoidance operation E. As shown in FIG. 7, it is not always necessary for the robot 1 to change the movement speed from the standard avoidance action C in both the operation initial stage t1 and the moving obstacle approach stage t2, and the movement speed only in the vicinity of the operation initial stage t1. It is also possible to plan a work avoidance operation E that is changed. Further, the robot 1 may plan the action avoiding operation E in which the moving speed is changed only in the vicinity of the moving obstacle approaching stage t2.

  FIG. 8 is a schematic plan view for explaining another example of the work avoidance operation E. As shown in FIG. 8, the robot 1 does not necessarily need to deviate from the original motion target trajectory even when the motion interferes with the human M, and a standard that avoids the human M only by changing the speed. The avoidance operation C and the work avoidance operation E can be planned.

  Here, the difference between the standard avoidance operation C and the action avoidance operation E in the case of FIG. 8 will be described. FIG. 9 is a graph showing temporal changes in the moving speed of the robot 1 in the standard avoidance operation C and the work avoidance operation E in the case of FIG.

  As shown in FIG. 9, the robot 1 first plans a standard avoidance operation C that performs a speed change necessary to avoid the human M. The robot 1 plans the standard avoidance operation C so that the human M can perform a smooth operation without any sense of incongruity.

  Thereafter, the robot 1 plans a work avoidance operation E in which the speed change in the standard avoidance operation C is greatly performed at the operation initial stage t1. In the work avoiding operation E, the movement speed of the robot 1 is planned to be maximized in the vicinity of the operation initial stage t1. In the work avoidance operation E shown in FIG. 9, the work avoidance operation E is planned so as to maintain the maximum moving speed of the robot 1 in the standard avoidance operation C. As described above, the robot 1 plans the action avoiding operation E in which the speed change in the initial operation stage t1 is increased and accelerates suddenly, thereby emphasizing the movement of the robot 1 with respect to the human M and causing the human M to You can notice the existence.

  Here, FIG. 10 is a graph showing another example of the time change of the moving speed of the robot 1 in the action avoiding operation E of FIG. In FIG. 10, acceleration is performed more rapidly than in the case of FIG. 9 without maintaining the maximum movement speed of the robot 1 in the standard avoidance operation C.

  As shown in FIG. 10, when the robot 1 can ensure safety from the surrounding environmental conditions and the performance of the robot 1, the work avoidance operation in which the change amount of the speed change at the initial operation stage t 1, that is, the acceleration is further increased. E can be planned. Thereby, the robot 1 can plan the action avoiding operation E that is more easily noticed by the human M and has a strong action effect.

  FIGS. 11 to 13 are schematic plan views for explaining still another example of the work avoidance operation E. FIG. As shown in FIGS. 11 to 13, the robot 1 can plan the action avoiding operation E of various patterns in addition to the above.

  For example, as shown in FIG. 11, the robot 1 can also plan a work avoidance operation E in which the avoidance width (horizontal offset) for the human M is larger than that in the case of FIG. . When the avoidance width is increased in this way, it is possible to emphasize the change in the course of the robot 1 and convey it to the human M. Further, when the robot 1 changes the course earlier than the standard avoidance operation C, the robot 1 can be informed of the intention of the robot 1 to the human M earlier. As a result, since the human M can perform the avoidance with the minimum necessary amount of avoidance operation, the burden on the human M in the transfer can be reduced.

  In addition, as shown in FIG. 12, the robot 1 can also plan a work avoidance operation E in which the course change timing is delayed compared to the standard avoidance operation C. In other words, the robot 1 in FIG. 12 plans the action avoiding operation E in which the amount of trajectory change is larger on the operation end stage t3 side than in FIG. Thus, when the course change timing is delayed, the avoidance operation amount of the expected avoidance operation D urged by the human M can be increased. As a result, the robot 1 can reduce its own avoidance motion amount, so that the given motion target can be efficiently performed.

  Further, as shown in FIG. 13, the robot 1 can also plan an action avoiding operation E meandering with respect to the trajectory of the standard avoiding operation C. In other words, the robot 1 in FIG. 13 plans an action avoiding operation E having alternately locations where the trajectory change amount is larger than the standard avoidance operation C and locations where the trajectory change amount in the reverse direction is added. When the robot 1 meanders in this way, the human M is wary without being able to read the intention of the robot 1, so that the avoidance operation amount of the expected avoidance operation D urged by the human M can be increased.

  Next, the configuration of the robot 1 will be described.

  As shown in FIG. 1, the robot 1 includes a control device 2 that performs overall control of the entire robot. The control device 2 is configured by using microcomputer hardware and software such as an ECU [Electric Control Unit]. The control device 2 is connected to the communication unit 3, the position detection unit 4, the map information storage unit 5, the radar sensor 6, and the camera 7.

  The communication unit 3 performs wireless communication with an external terminal for robot control. The communication unit 3 receives a task (operation target) input to an external terminal by wireless communication. The communication unit 3 transmits the received task information to the control device 2.

  The position detection unit 4 detects the current location of the robot 1. The position detection unit 4 performs position detection using a GPS [Global Positioning System] or a wireless LAN [Local Area Network]. In position detection, the current location may be obtained by comparing the relative position information of the peripheral obstacles and landmarks acquired by an external sensor such as a radar sensor or camera with the map information. Alternatively, a method of simultaneously creating a map by an external sensor and estimating the current location, such as SLAM [Simultaneously Localization and Mapping], may be used.

  The map information storage unit 5 stores a wide range of map information. The position detection unit 4 recognizes the current location based on the map information in the map information storage unit 5. The position detection unit 4 transmits the current location information of the robot 1 and map information around the current location to the control device 2.

  The radar sensor 6 detects obstacles around the robot based on the reflected wave of the emitted radio wave. Obstacles include fixed obstacles such as buildings, and moving obstacles such as humans, animals, cars, and other robots. The radar sensor 6 transmits the detected position information of the obstacle to the control device 2.

  The camera 7 captures an image around the robot 1. The camera 7 transmits image information around the captured robot to the control device 2. The radar sensor 6 and the camera 7 function as a moving obstacle detection unit described in the claims. The radar sensor in the present invention may be any radar sensor that can measure the distance to an obstacle using similar means such as light, electromagnetic waves, and sound waves, as well as radio waves.

  The control device 2 includes a robot motion target generation unit 10, a human motion prediction unit 11, a motion plan unit 12, and a movement control unit 13.

  The robot operation target generator 10 generates an operation target for the robot 1. The robot operation target generation unit 10 generates an operation target based on the task information of the communication unit 3 and the current location information and map information of the position detection unit 4.

  Specifically, when a destination is given as a task, the robot motion target generation unit 10 sets a control target value of a robot moving actuator (wheel or the like) so as to reach the given destination efficiently. Generate. Examples of such target values include a route (arrangement of positions), a course (arrangement of positions and time or velocity), an acceleration / deceleration pattern (arrangement of acceleration), and the like. In addition, the operation target includes a work that does not involve moving or staying in a specific place (such as cleaning the floor under the feet).

  The human motion prediction unit 11 predicts the motion of the human M existing around the robot 1. The human motion prediction unit 11 first determines whether or not the human M is detected around the robot 1 based on the position information of the obstacle of the radar sensor 6 and the image information around the robot of the camera 7. The position information of the person M may be detected using only one of the position information of the obstacle of the radar sensor 6 and the image information around the camera robot. Or you may acquire the GPS data which the person M has by communication.

  When it is determined that the human M has been detected, the human motion prediction unit 11 recognizes the traveling direction and the moving speed from the history of the position information of the human M. The human motion prediction unit 11 predicts a future motion when the human M maintains the traveling direction and the moving speed. This future operation includes an operation of continuing to advance in a predetermined direction, an operation of decelerating and stopping, an operation of changing the direction, and an operation of staying on the spot. Future motions include prediction of motion target trajectories. Note that the human motion prediction unit 11 predicts the motion of the human M with higher accuracy by recognizing the appearance of the target human M and the surrounding circumstances (positional relationship with an obstacle, congestion, etc.). be able to.

  The motion planning unit 12 plans the motion of the robot 1. The operation planning unit 12 includes an operation interference determination unit 14, a transfer operation planning unit 15, and a work avoidance operation planning unit 16.

  The motion interference determination unit 14 determines whether the motion of the human M and the robot 1 interferes based on the motion target generated by the robot motion target generation unit 10 and the motion of the surrounding human M predicted by the human motion prediction unit 11. Determine whether. The motion interference in the motion interference determination unit 14 includes not only the contact between the human M and the robot 1 but also a state where the human M and the robot 1 are so close that a safety margin cannot be secured. The safety margin is set in advance according to the external shape and application of the robot 1.

  The motion interference determination unit 14 determines that the motion interferes when it is determined that the human and the robot 1 approach within the safety margin. The motion interference determination unit 14 functions as an approach determination unit described in the claims. The safety margin of the robot 1 corresponds to the predetermined interval described in the claims.

  The transfer operation planning unit 15 plans a transfer operation when the operation interference determination unit 14 determines that the surrounding human M and the operation of the robot 1 interfere with each other. This transfer operation is an operation including the standard avoidance operation C of the robot 1 and the expected avoidance operation D expected from the human M (see FIG. 5). The standard avoidance operation C is an operation of the robot 1 for the purpose of avoiding operation interference with the human M. The expected avoidance operation D is an avoidance operation that can be expected to be performed by the human M in order to avoid unnecessary approach with the robot 1. The transfer operation planning unit 15 plans the expected avoidance operation D using, for example, past statistical data.

  Based on the planned expected avoidance operation D and the operation target generated by the robot operation target generation unit 10, the transfer operation planning unit 15 can avoid the standard avoidance operation most efficiently with the minimum necessary operation while ensuring a safety margin. Plan C. Further, after planning the standard avoidance operation C of the robot 1, the transfer operation planning unit 15 can also plan the expected avoidance operation D of the human M based on the planned standard avoidance operation C. The transfer operation planning unit 15 functions as a standard avoidance operation planning unit described in the claims.

  Based on the standard avoidance operation C and the expected avoidance operation D planned by the transfer operation planning unit 15, the action avoidance operation planning unit 16 performs an avoidance operation for avoiding the person M and an action operation for encouraging the person M to expect the avoidance operation D. Plan a work avoidance action E that doubles as well. The action avoidance operation planning unit 16 plans an avoidance operation in which the speed and change amount of the operation change in the standard avoidance operation C are increased as the action avoidance operation E.

  As described above, the action avoidance operation planning unit 16 can plan the action avoidance operation E of various patterns (see FIGS. 5 to 13). The action avoidance operation planning unit 16 plans the action avoidance operation E having the most appropriate pattern based on the hardware configuration of the robot 1, the surrounding environment, the current operation target, and the like. The action avoidance operation planning unit 16 functions as an action avoidance operation unit described in the claims.

  The movement control unit 13 controls the movement mechanism of the robot 1. The movement control unit 13 executes the standard avoidance operation C and the work avoidance operation E of the robot 1 by controlling the movement mechanism based on the avoidance operation planned by the transfer operation planning unit 15. In addition, the movement control unit 13 controls the movement mechanism along the operation target generated by the robot operation target generation unit 10 to achieve the operation target. The movement control unit 13 functions as a control unit described in the claims.

  Next, a control procedure in the control device 2 described above will be described.

  As shown in FIG. 2, in the control device 2, first, the robot operation target generation unit 10 performs an operation target generation process for generating an operation target of the robot 1 (S <b> 1). The robot operation target generation unit 10 generates an operation target based on the task information of the communication unit 3 and the current location information and map information of the position detection unit 4.

  Next, the human motion prediction unit 11 determines whether or not the human M around the robot 1 has been detected based on the position information of the obstacle of the radar sensor 6 and the image information around the robot of the camera 7 ( S2). When the human motion prediction unit 11 determines that the human M around the robot 1 is not detected, the human motion prediction unit 11 proceeds to step S5. The target of the operation of the present invention is not limited to the human M, and the present invention can be applied to all moving obstacles such as robots and automobiles that move autonomously.

  When it is determined that the human M around the robot 1 has been detected, the human motion prediction unit 11 performs a motion prediction process for predicting the motion of the detected human M around the robot 1 (S3). The human motion prediction unit 11 recognizes the traveling direction and moving speed of the surrounding human M based on the position information of the obstacle of the radar sensor 6 and the image information around the robot of the camera 7. The human motion prediction unit 11 predicts the future motion of the human M when the traveling direction and the moving speed are maintained.

  The motion interference determination unit 14 interferes with the motion of the surrounding human M and the robot 1 based on the motion target generated by the robot motion target generation unit 10 and the motion of the surrounding human M predicted by the human motion prediction unit 11. Whether or not (S4). If the operation interference determination unit 14 determines that there is no operation interference, the operation interference determination unit 14 proceeds to step S5.

  In step S <b> 5, the movement control unit 13 performs a normal time robot control process for controlling the moving mechanism of the robot 1 along the operation target generated by the robot operation target generation unit 10.

  On the other hand, when the operation interference determination unit 14 determines that there is an operation interference in step S4, the transfer operation planning unit 15 performs a transfer operation planning process for planning a transfer operation (S6). The transfer operation planning unit 15 plans the standard avoidance operation C of the robot 1 and the expected avoidance operation D expected of the human M as the transfer operation.

  Thereafter, the work avoidance operation planning unit 16 performs a work avoidance operation planning process for planning the work avoidance operation E based on the standard avoidance operation C and the expected avoidance operation D planned by the transfer operation planning unit 15 (S7).

  In step S <b> 8, the movement control unit 13 performs an on-interference robot control process for realizing a transfer operation between the human M and the robot 1. In the robot control process at the time of interference, the movement control unit 13 controls the movement mechanism of the robot 1 based on the action avoiding action E, so that the avoidance action of the robot 1 and the action action that prompts the human M to avoid are executed simultaneously. . In the robot control process at the time of interference, the robot 1 is controlled along the operation target after the transfer operation is completed.

  Thereafter, the control device 2 determines whether or not the operation target is achieved and the operation of the robot 1 is completed (S9). When it is determined that the operation of the robot 1 is not completed, the control device 2 repeats the process from step S1. When it is determined that the operation of the robot 1 is completed, the control device 2 ends the control.

  Then, the effect of the robot 1 mentioned above is demonstrated.

  According to the robot 1 according to the first embodiment, when a moving obstacle such as a human approaches within a predetermined interval of the robot 1, by performing the work avoidance operation E instead of the standard avoidance operation C, the robot 1 At the same time as avoidance, it is possible to encourage the moving obstacles to avoid. Thereby, compared with the case where the robot 1 unilaterally performs the avoidance operation, it is possible to perform an efficient passing, so that a given task can be performed efficiently.

  Further, according to the robot 1, by acting on the human M to perform the expected avoidance operation D, it is possible to increase the possibility that the human M side understands the intention of the robot 1 and performs the expected avoidance operation D. . Then, when the robot 1 performs the work avoidance operation E corresponding to the expected avoidance operation D, it is possible to realize passing between the human M and the robot 1 with concession performed between humans. Therefore, according to this robot 1, even in a crowded environment, it is possible to achieve its own tasks by giving up with surrounding people, and coexistence between the human and the robot can be realized.

  Further, in this robot 1, since the action is performed by the action avoiding action E in which the speed and change amount of the action change in the standard avoidance action C is increased, it is not necessary to use equipment such as a manipulator or a speaker for the action. Even when the equipment is not provided or when the equipment is being used for another work, the work can be realized.

  Therefore, the robot 1 does not need to be provided with facilities such as a manipulator, a speaker, a light, and a display for working, so that the cost of the hardware configuration can be reduced. Further, even with an existing robot that does not include equipment such as a manipulator, the function of the present invention can be added without adding equipment. In addition, since it is possible to work without using equipment such as a manipulator, it is possible to realize a transfer that is coordinated with surrounding people while carrying out other operations such as transporting luggage. This also realizes further efficiency improvement of task processing.

  Further, according to the robot 1, the action avoiding operation E is planned in which the operation speed is higher than the standard avoidance operation C in the vicinity of the initial operation stage t1 and lower than the standard avoidance operation C in the moving obstacle approach stage t2. Therefore, by emphasizing the movement at the movement initial stage t1 where there is a distance from the moving obstacle, the movement obstacle can be easily noticed and the movement speed is reduced at the movement obstacle approach stage t2. Highly safe passing can be performed against moving obstacles. Further, by appropriately selecting the acceleration at the operation initial stage t1 and the deceleration at the moving obstacle approach stage t2, the motion state at the operation end stage t3 can be made the same as the standard avoidance action C.

  Further, according to this robot 1, when the robot 1 and the moving obstacle approach each other, the moving obstacle also normally performs an avoiding operation. Therefore, the expected avoiding operation D expected for the moving obstacle is planned, and this By planning the standard avoidance operation C so as to have an avoidance amount considering the expected avoidance operation D, an efficient passing between the moving obstacle and the robot 1 can be achieved.

[Second Embodiment]
As shown in FIG. 15, in the robot 21 according to the second embodiment, the avoidance motion amount of the human M reaches the expected avoidance motion amount expected by the robot 21 compared to the robot 1 according to the first embodiment. Control based on the avoidance operation E, in which the robot 21 is controlled based on the determination result of the avoidance operation amount determination unit 24 and the avoidance operation amount determination unit 24 that determines whether or not it is predicted that it will be clear. The main difference is that the control shifts to the control based on the standard avoidance operation C.

  FIG. 16 is a schematic plan view for explaining the action avoiding operation E of the robot 21 according to the second embodiment. T4 shown in FIG. 16 represents a normal operation return stage in which the normal operation is restored after the operation end stage t3 in which the avoidance operation is finished. Further, Q shown in FIG. 16 represents the avoidance operation amount in the first course change of the robot 21, and QM represents the expected avoidance operation amount that the robot 21 expects from the human M.

  As shown in FIG. 16, the robot 21 plans a trajectory action avoiding operation E in which a course change is performed in a plurality of stages. The robot 21 performs an action for prompting the human M to perform the expected avoidance action D by the first course change in the action avoidance action E.

  Specifically, the robot 21 transmits the expected expected avoidance motion amount QM to the human M based on the avoidance motion amount Q in the first course change. The avoidance motion amount Q can be expressed, for example, as the shortest distance between the route of the robot 21 before the start of the avoidance avoidance operation E and the route of the robot 21 immediately after the completion of the first route change. The expected avoidance motion amount QM corresponds to an avoidance motion amount that sufficiently satisfies the expected avoidance motion D planned by the robot 21, for example. Further, as the expected avoidance operation amount QM, a value smaller than the avoidance operation amount that satisfies the expected avoidance operation D may be selected.

  When the expected avoidance motion amount expected for the human M is large, the robot 21 works to make the avoidance motion amount Q small and plans the avoidance motion E. In addition, when the expected avoidance motion amount QM expected from the human M is small, the robot 21 plans the avoidance motion E so as to increase the avoidance motion amount Q.

  The avoidance operation amount determination unit 24 of the robot 21 detects the avoidance operation amount of the human M when it is determined that the robot 1 and the operation of the human M interfere with each other. The avoidance movement amount of the human M can be detected as, for example, the shortest distance between the path of the human M before the avoidance movement and the current path of the human M.

  The avoidance motion amount determination unit 24 determines whether or not it is predicted that the detected avoidance motion amount of the person M has reached or is expected to reach the expected avoidance motion amount QM. The avoidance motion amount determination unit 24 predicts that it is clear that the avoidance motion amount of the human M reaches the expected avoidance motion amount QM from changes in the speed and acceleration of the human M, for example. If the avoidance motion amount determination unit 24 determines that it is predicted that the avoidance motion amount of the human M has reached or is expected to reach the expected avoidance motion amount QM, no further action is required. The control based on the operation E is stopped, and the control is shifted to the control based on the standard avoidance operation C.

  Next, a control procedure in the control device 22 of the robot 21 described above will be described.

  As shown in FIG. 17, in the control device 22, the robot operation target generation unit 10 first performs an operation target generation process for generating an operation target of the robot 1 (S11). The robot operation target generation unit 10 generates an operation target based on the task information of the communication unit 3 and the current location information and map information of the position detection unit 4.

  Next, the human motion prediction unit 11 determines whether or not the human M around the robot 1 has been detected based on the position information of the obstacle of the radar sensor 6 and the image information around the robot of the camera 7 ( S12). When the human motion prediction unit 11 determines that the human M around the robot 1 is not detected, the human motion prediction unit 11 proceeds to step S15.

  When it is determined that the human M around the robot 1 has been detected, the human motion predicting unit 11 performs a motion prediction process for predicting the detected motion of the surrounding human M (S13). The human motion prediction unit 11 recognizes the traveling direction and moving speed of the surrounding human M based on the position information of the obstacle of the radar sensor 6 and the image information around the robot of the camera 7. The human motion prediction unit 11 predicts the future motion of the human M when the traveling direction and the moving speed are maintained.

  The motion interference determination unit 14 interferes with the motion of the surrounding human M and the robot 1 based on the motion target generated by the robot motion target generation unit 10 and the motion of the surrounding human M predicted by the human motion prediction unit 11. Whether or not (S14). If the operation interference determination unit 14 determines that there is no operation interference, the operation interference determination unit 14 proceeds to step S15.

  In step S <b> 15, the movement control unit 13 performs a normal time robot control process for controlling the movement mechanism of the robot 1 along the operation target generated by the robot operation target generation unit 10.

  On the other hand, when the operation interference determination unit 14 determines that there is an operation interference in step S14, the transfer operation planning unit 15 performs a transfer operation planning process for planning the transfer operation (S16). The transfer operation planning unit 15 plans the standard avoidance operation C of the robot 1 and the expected avoidance operation D expected of the human M as the transfer operation. Thereafter, the work avoidance operation planning unit 16 performs a work avoidance operation planning process for planning the work avoidance operation E based on the standard avoidance operation C and the expected avoidance operation D planned by the transfer operation planning unit 15 (S17).

  In step S <b> 18, the movement control unit 13 performs a work avoidance operation control process for performing movement control of the robot 21 based on the work avoidance operation E. When the movement control unit 13 starts the work avoidance operation control process, the control device 22 determines whether or not the work avoidance operation E of the robot 21 is incomplete (S19). When it is determined that the work avoidance operation E has been completed, the control device 22 repeats the processing from step S11.

  When the control device 22 determines that the work avoidance operation E is incomplete, the avoidance operation amount determination unit 24 predicted that the avoidance operation amount of the human M has reached or reached the expected avoidance operation amount QM. It is determined whether or not (S20). When the avoidance motion amount determination unit 24 determines that the avoidance motion amount of the human M has not reached the expected avoidance motion amount QM and has not predicted it, the avoidance motion amount determination unit 24 returns to step S18 and continues the work avoidance motion control process.

  When it is determined that the avoidance motion amount determination unit 24 has predicted that the avoidance motion amount of the human M has reached or is expected to reach the expected avoidance motion amount QM, the movement control unit 13 determines that the avoidance motion amount determination unit 24 is based on the standard avoidance operation amount C. A standard avoidance operation control process for controlling the moving mechanism of the robot 21 is performed (S21). In the standard avoidance operation control process, the movement control unit 13 controls the movement mechanism of the robot 1 based on the standard avoidance operation C, thereby realizing an efficient avoidance operation only for the avoidance of the robot 21. In the standard avoidance operation control process, the robot 21 is controlled in accordance with the operation target after the avoidance operation is completed.

  Thereafter, the control device 22 determines whether or not the operation target is achieved and the operation of the robot 1 is completed (S19). When it is determined that the operation of the robot 1 is not completed, the control device 2 repeats the process from step S11. When it is determined that the operation of the robot 1 has been completed, the control device 22 ends the control.

  According to the robot 21 according to the second embodiment described above, it is predicted that the avoidance operation amount of the moving obstacle such as the human M reaches or reaches the expected avoidance operation amount QM expected by the robot 21. In this case, it is not necessary to perform further action. Therefore, by shifting from the action avoidance operation to the control based on the standard avoidance operation, an efficient passing can be performed.

  Also, according to this robot 21, by planning a trajectory action avoiding operation E in which a course change is performed in a plurality of stages, anticipation avoidance that works on the human M according to the magnitude of the avoidance operation amount Q of the first course change. The magnitude of the operation amount QM can be adjusted.

  In the robot 21, if it is impossible to predict that the avoidance operation amount of the human M has not reached the expected avoidance operation amount QM even after the first change of route, a new route for further action is provided. It may be an aspect in which a stage of change is added. Thereby, the degree and length of the action with respect to the human M can be adjusted appropriately.

[Third Embodiment]
As shown in FIG. 18, the robot 31 according to the third embodiment has a clearer direction (front direction) of the robot 31 than the robot 1 according to the first embodiment. The main difference is that the human M is acted on depending on the orientation.

  Specifically, the robot 31 according to the third embodiment has an appearance with a clear front direction when viewed from a human, such as having a facial part. The robot 31 has a moving mechanism that can shift the front direction and the traveling direction of the robot 31. Examples of such a moving mechanism include an omni wheel and a biped walking mechanism.

  The robot 31 having such a configuration plans a work avoidance operation E in which the amount of change in the direction among the operation changes of the standard avoidance operation C is increased. The robot 31 changes the orientation of the robot 31 so that the avoidance direction expected of the human M is on the back side of the robot 31, and plans a work avoidance operation E in which the operation is performed while maintaining this orientation.

  According to the robot 31, a moving obstacle such as the human M tries to turn to the back side instead of the front side which is normally considered as the traveling direction of the robot 31 when avoiding the robot 31. By moving the robot 31 while maintaining the orientation of the robot 31 so that the direction of avoidance is on the back side of the robot 31, an action for guiding the avoidance direction of the moving obstacle can be performed. As a result, it is possible to work on the human M while performing the most efficient avoidance without changing the trajectory of the standard avoidance action C. Therefore, an efficient passing can be achieved, and a given task can be performed. It can be done efficiently.

  Note that the robot 31 may stop working when it is determined that the human M has sufficiently avoided, like the robot 21 according to the second embodiment. That is, when the robot 31 predicts that the human M's avoidance motion amount has reached or is expected to reach the expected avoidance motion amount QM, the robot 31 stops working and matches the front direction and the traveling direction. The mode which returns direction may be sufficient.

  The present invention is not limited to the embodiment described above.

  For example, the moving body according to the present invention is not limited to a robot, and may be a manipulator, an automobile, personal mobility, or the like. Further, the moving body is not necessarily limited to the one that moves on the ground, and may be one that moves in the air or underwater.

  Further, in the above-described embodiment, the case of operating with a human being is described, but the present invention can be suitably applied to a partner other than a human being. That is, the moving obstacles described in the claims include humans, animals such as dogs, mobility such as automobiles, and other robots.

DESCRIPTION OF SYMBOLS 1, 21 ... Robot (moving body) 2, 22 ... Control apparatus 6 ... Radar sensor (moving obstacle detection means) 7 ... Camera (moving obstacle detection means) 10 ... Robot motion target generation part 11 ... Human motion prediction part DESCRIPTION OF SYMBOLS 14 ... Operation | movement interference judgment part (approach determination means) 15 ... Transfer operation plan part (standard avoidance action plan means) 16 ... Action avoidance action plan part (work avoidance action plan means) 24 ... Avoidance action amount determination part (avoidance action determination means) A ... Target motion trajectory B ... Target motion trajectory C ... Standard avoidance motion D ... Expected avoidance motion E ... Work avoidance motion M ... Human Q ... Avoidance motion amount QM ... Expected avoidance motion amount t1 ... Operation initial stage t2 ... Movement obstacle Approaching stage t3 ... Operation end stage t4 ... Normal operation return stage

Claims (5)

A moving obstacle detecting means for detecting a moving obstacle existing around the moving body;
An approach determining means for determining whether or not the moving obstacle and the moving body approach within a predetermined interval;
A standard avoidance operation plan for planning a standard avoidance operation in which the moving object avoids the moving obstacle when the approach determining means determines that the moving obstacle and the moving object approach within a predetermined interval. Means,
Based on the standard avoidance operation planned by the standard avoidance operation planning means, the avoidance operation in which the moving body avoids the moving obstacle and the action operation that prompts the moving obstacle to avoid is planned. A work avoidance action planning means;
Control means for controlling the moving body based on the work avoidance operation;
With
The moving object avoiding action planning unit plans an avoiding action in which at least one of a speed of an action change in the standard avoiding action and a change amount of the action change is increased as the action avoiding action.   The action avoidance operation planning means plans the action avoidance operation so that the operation speed is higher than the standard avoidance operation in the initial stage of operation, and the operation speed is lower than that of the standard avoidance operation in the moving obstacle approach stage. The moving body according to claim 1.   3. The standard avoidance operation planning unit plans an expected avoidance operation expected for the moving obstacle and plans the standard avoidance operation based on the expected avoidance operation. The moving body described. Based on the detection result of the moving obstacle detection means and the expected avoidance action planned by the standard avoidance action planning means, the avoidance action amount of the moving obstacle reaches the expected avoidance action amount expected for the moving obstacle. An avoidance operation amount determination means for determining whether or not
The control unit controls the moving body based on the standard avoidance operation when the avoidance operation amount determination unit determines that the avoidance operation amount of the moving obstacle has reached the expected avoidance operation amount. The moving body according to claim 3.   The action avoidance operation planning means plans the action avoidance operation for performing an operation while maintaining the orientation of the moving body so that the avoidance direction expected for the moving obstacle is on the back side of the moving body. The moving body according to claim 3 or 4, characterized by the above.

Images