JP2020006489A - Track formation device, track formation method and track formation program - Google Patents

Abstract

To provide a technique for forming a track for a movable body moving to a target position, by avoiding a situation in which the track for the movable body moving to the target position cannot be formed.SOLUTION: A track formation device according to the present disclosure, which forms a track for a movable body moving to a target position, comprises: a track formation unit, a determination unit, and a direction change unit. In S402 and S414, the track formation unit searches and forms a track through a predetermined algorithm. In S408, the determination unit determines whether the track formation unit can form the track or not. In S410, if the determination unit determines that the track formation unit cannot form the track, the direction change unit changes a search direction of the algorithm through which the track formation unit searches the track.SELECTED DRAWING: Figure 4

Description

  The present disclosure relates to a technique for generating a trajectory of a moving object that moves to a target position.

  A technique for generating a trajectory of a moving body that moves to a target position is known. For example, according to the technology described in Patent Literature 1, when a hand of a manipulator, which is a moving body, searches for and generates a trajectory that reaches a target position from a start position, the potential of the target position is minimized as a search algorithm, and the potential of the target is minimized. A map is generated in which the potential increases as the distance from the position increases. Then, a trajectory is searched from the starting position in a direction having a low potential, and a trajectory that reaches the target position where the poncial is lowest is generated.

JP 2003-280710 A

  However, according to the technique described in Patent Literature 1, when a moving object enters a concave obstacle by searching for a direction having a low potential, a moving direction having a lower potential than the current position does not exist. Sometimes. In a policy of simply searching for a trajectory in a direction having a low potential as in the technique described in Patent Literature 1, this state cannot be avoided, so that there is a problem that a trajectory of a moving object that moves to a target position cannot be generated.

  The present disclosure provides a technique for avoiding such a state and generating a trajectory of a moving body that moves to a target position even when the state becomes unable to generate a trajectory of the moving body moving to the target position.

  A trajectory generation device (10) of the present disclosure is a trajectory generation device that generates a trajectory (210, 230) of a moving object (100) moving to a target position (120), and includes a trajectory generation unit (12, S402, S414). ), A determination unit (16, S408), and a policy change unit (18, S410).

  The trajectory generation unit searches for and generates a trajectory using a predetermined algorithm. The determination unit determines whether the trajectory generation unit can generate a trajectory. If the determination unit determines that the trajectory generation unit cannot generate the trajectory, the policy change unit changes the search policy of the algorithm by which the trajectory generation unit searches for the trajectory.

  According to such a configuration, when it becomes impossible to generate the trajectory of the moving object moving to the target position, the search policy of the algorithm for searching the trajectory is changed, so that the trajectory of the moving object moving to the target position cannot be generated. And the trajectory of the moving object moving to the target position can be generated.

  It should be noted that reference numerals in parentheses described in this column and in the claims indicate a correspondence relationship with specific means described in an embodiment described below as one aspect, and the technical scope of the present disclosure will be described. There is no limitation.

FIG. 2 is a block diagram showing a trajectory generation device according to the embodiment. 9 is a flowchart illustrating a trajectory generation process. FIG. 4 is a schematic diagram showing a process of generating a trajectory. FIG. 9 is a schematic diagram showing another generation process of the trajectory.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[1. Constitution]
The trajectory generating device 10 illustrated in FIG. 1 generates a trajectory of a moving body that moves to a target position. The trajectory generating device 10 is mainly configured by a known microcomputer having a CPU and a semiconductor memory such as a RAM, a ROM, and a flash memory.

  Various functions of the trajectory generation device 10 are realized by the CPU executing a program stored in a non-transitional substantial recording medium. In this example, the semiconductor memory corresponds to a non-transitional substantial recording medium storing a program. When this program is executed, a method corresponding to the program is executed. The trajectory generating device 10 may be constituted by one microcomputer, or may be constituted by a plurality of microcomputers.

The trajectory generation device 10 includes a trajectory generation unit 12, a movement control unit 14, and a determination unit 16 and a policy change unit 18 as a configuration of functions realized by the CPU executing the program.
The method of realizing these elements constituting the trajectory generating device 10 is not limited to software, and one or a plurality of hardware may be realized for one or all of the elements. For example, when the above function is implemented by an electronic circuit that is hardware, the electronic circuit may be implemented by a digital circuit including a large number of logic circuits, an analog circuit, or a combination thereof.

  The trajectory generation unit 12 searches for and generates a trajectory 210 and a rotation direction 230 where the mobile unit 100 moves to the target position 120 as shown in FIGS. 2 and 3, for example. The moving body 100 is, for example, a transfer robot that transfers materials at a robot arm or a factory.

The movement control unit 14 moves the moving body 100 along the trajectory searched by the trajectory generation unit 12.
The determination unit 16 determines whether or not the movement control unit 14 moves the moving body 100 along the trajectory searched by the trajectory generating unit 12 and the trajectory generating unit 12 can generate a trajectory at which the moving body 100 reaches the target position 120. Is determined.

  When the determination unit 16 determines that the trajectory generation unit 12 cannot generate the trajectory of the moving object 100 that moves to the target position 120, the policy change unit 18 changes the search policy of the algorithm by which the trajectory generation unit 12 generates the trajectory. When the search policy of the algorithm for generating the trajectory is changed by the policy changing unit 18, the trajectory generation unit 12 searches for the trajectory of the moving body 100 moving to the target position 120 based on the changed search policy.

[2. processing]
Next, the trajectory generation processing executed by the trajectory generation device 10 will be described with reference to the flowchart of FIG.

  In S400, the trajectory generation unit 12 initializes settings for generating the trajectory of the moving body 100. For example, the trajectory generation unit 12 sets a moving speed and an acceleration, or a rotational angular velocity and an operation limit angle that can be rotated from the initial position, based on the initial position of the moving body 100. Furthermore, the coordinates of the trajectory specifying the trajectory of the moving body 100 and the moving direction of the moving body 100 at the coordinates, or the state parameters related to the movement of the moving body 100 such as the rotation angle position and the rotating direction of the moving body 100 are initialized.

  In S402, the trajectory generating unit 12 searches for an optimal trajectory so that the mobile unit 100 moves to the target position based on preset trajectory search constraints and conditions. The constraints and conditions for the trajectory search include, for example, a predetermined range in which the moving body 100 can move, and a direction in which the cost required for the moving body 100 to move to the target position 120 as an optimal trajectory is minimized. Moving.

  The cost required for the moving body 100 to move to the target position 120 is, for example, a distance, energy, time, and the like for the moving body 100 to move to the target position 120. The cost is calculated based on, for example, a cost function.

  In S404, the movement control unit 14 moves the moving body 100 based on the trajectory searched by the trajectory generation unit 12. Note that the movement control unit 14 may actually move the moving body 100 based on the trajectory searched by the trajectory generation unit 12, or may simulate the moving body 100 based on the trajectory searched by the trajectory generation unit 12 by simulation. May be moved.

  In S406, the determination unit 16 determines whether the moving body 100 has reached the target position 120. The determining unit 16 determines that the moving body 100 has reached the target position 120 if the difference between the distance between the moving body 100 and the target position 120 is equal to or less than a predetermined distance. If the determination in S406 is Yes, that is, if the moving body 100 has reached the target position 120, the process proceeds to S412.

  If the determination in S406 is No, that is, if the moving body 100 has not reached the target position 120, in S408, the determination unit 16 repeats the state in which the moving body 100 has been stopped at the same position for a predetermined number of times. Is determined. For example, when the amount of change in the position of the moving body 100 is equal to or less than a predetermined amount, or the amount of change in the cost calculated by the cost function is equal to or less than the predetermined amount, the determining unit 16 stops the moving body 100 at the same position. It is determined that there is.

  For example, in FIG. 2, when the moving body 100 moves to the target position 120, moves in the gradient vector direction indicated by the arrow 200 where the cost decreases fastest, and enters the recess 112 of the concave obstacle 110, the dotted line indicates In any direction indicated by the arrow 202, the cost is the same or higher. As a result, it may not be possible to search for a moving direction in which the cost decreases and the gradient decreases. In this case, since the moving body 100 is at the immovable position where it cannot move, the amount of change in the position or the amount of change in the cost is equal to or less than a predetermined amount, and the vehicle 100 is stopped at the same position.

  In addition, as shown in FIG. 3, since the moving body 100 moves to the target position 120, the moving body 100 rotates around the shaft 102 in the counterclockwise rotation direction 220 where the cost is fastest, and moves with the operation limit angle 130. When the difference from the rotation angle of the body 100 becomes equal to or less than the predetermined angle and the body 100 rotates to the operation limit angle 130, the moving body 100 cannot rotate in the rotation direction 220 beyond the operation limit angle 130.

  In this case, since the moving body 100 is in the immovable position where it cannot rotate in the rotation direction 220 beyond the operation limit angle 130, the change amount of the rotation angle or the change amount of the cost is not more than a predetermined amount as the position change amount. And stop at the same position.

  If the determination in S408 is No, that is, if the state where the moving body 100 has stopped at the same position for the predetermined time has not been repeated the predetermined number of times, the process proceeds to S402. In this case, the trajectory generation unit 12 continues searching for the trajectory according to the search policy of the current algorithm.

  If the determination in S408 is Yes, that is, if the state in which the moving body 100 has been stopped at the same position for the predetermined time is repeated a predetermined number of times, the policy change unit 18 searches the trajectory generation unit 12 for an algorithm for searching the trajectory. You decide that you need to change your policy.

  In this case, in S410, the policy changing unit 18 changes the search policy of the algorithm for searching for the trajectory, and shifts the processing to S402. Accordingly, the trajectory generation unit 12 continues searching for the trajectory using the algorithm whose search policy has been changed.

  For example, in FIG. 2, the policy change unit 18 changes the search policy of the algorithm by setting a cost 114 larger than the surroundings at the entrance of the recess 112 of the obstacle 110. Then, the trajectory generator 12 searches for the trajectory 210 avoiding the entrance of the obstacle 110.

  In the case of FIG. 3, the policy change unit 18 changes the algorithm search policy by reversing the counterclockwise rotation direction 220 toward the operation limit angle 130 into the clockwise rotation direction 230. Then, the trajectory generation unit 12 searches for the trajectory of the moving body 100 with the rotation direction 230 changed clockwise as the trajectory.

  In S412, the trajectory generating unit 12 sets the trajectory at which the mobile unit 100 has reached the target position as the optimum trajectory, and sets the trajectory in each time step in the time series of the optimum trajectory, that is, at the processing time interval between S402 to S406. The position, the moving speed, the moving state of the moving body 100 such as whether or not the movement is interrupted by an obstacle is input, and the moving direction, the moving speed, and the like of the moving body 100 moving from the state are output as the output of the neural network. Create training data.

  In S414, the trajectory generation unit 12 subsequently generates a trajectory in which the mobile unit 100 moves to the target position in real time by supervised machine learning of the neural network based on the learning data of the optimal trajectory generated multiple times in S412.

[3. effect]
According to the embodiment described above, the following effects can be obtained.
(1) When the trajectory of the moving body 100 moving to the target position 120 cannot be generated, the trajectory generation device 10 changes the search policy in the algorithm for searching for the trajectory.

As a result, even when a state in which the trajectory of the moving body 100 that moves to the target position 120 cannot be generated can be avoided, the trajectory of the moving body 100 that moves to the target position 120 can be generated.
(2) If the costs of a wide range of positions around the moving body 100 are stored, the trajectory where the amount of change in the position or the amount of change in the cost is equal to or less than a predetermined amount before the moving body 100 reaches the target position 120 is determined. By avoiding the trajectory, the trajectory of the moving body 100 moving to the target position 120 can be generated. However, a large storage capacity is required to store the cost of a wide range of positions around the mobile unit 100.

  On the other hand, in the present embodiment, when the trajectory of the moving body 100 moving to the target position 120 cannot be generated, the trajectory generation device 10 changes the search policy in the algorithm for searching the trajectory. The trajectory of the moving body 100 moving to the target position 120 can be generated by storing the cost of the position in the narrowest possible range. Therefore, the storage capacity for storing costs can be reduced.

  (3) By learning data of a plurality of optimal trajectories with a neural network, even if the moving environment of the moving body 100 changes due to a change in the position, size, or the like of the obstacle, the trajectory generating device 10 can perform real-time processing. In response to the change in the moving environment of the moving body 100, the trajectory of the moving body 100 moving to the target position 120 can be generated.

In the above embodiment, S402 and S414 correspond to the processing of the trajectory generation unit, S408 corresponds to the processing of the determination unit, and S410 corresponds to the processing of the policy change unit 18.
[4. Other Embodiments]
Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above embodiments, and can be implemented with various modifications.

  (1) In the above embodiment, when the moving body 100 rotates in the counterclockwise rotation direction 220 and stops at the operation limit angle 130 in FIG. 3, the mobile body 100 rotates in the clockwise rotation direction 230 opposite to the counterclockwise rotation. The trajectory search policy was changed by changing the direction.

  On the other hand, in FIG. 3, when the moving body 100 rotates in the counterclockwise rotation direction 220 and stops at the operation limit angle 130, the cost of the operation limit angle 130 is made larger than the cost of the surroundings, so that the trajectory search policy is set. And the rotation direction of the moving body 100 may be changed to the clockwise rotation direction 230.

  (2) In the above-described embodiment, a robot arm or a transfer robot that transfers materials is illustrated as the moving body 100. In addition to this, any moving object may be used to generate the trajectory that moves to the target position.

  (3) A plurality of functions of one component in the above embodiment may be realized by a plurality of components, or one function of one component may be realized by a plurality of components. . Also, a plurality of functions of a plurality of components may be realized by one component, or one function realized by a plurality of components may be realized by one component. Further, a part of the configuration of the above embodiment may be omitted. Further, at least a part of the configuration of the above-described embodiment may be added to or replaced by the configuration of another above-described embodiment. It should be noted that all aspects included in the technical idea specified by the language described in the claims are embodiments of the present disclosure.

  (4) In addition to the trajectory generation device 10 described above, a system including the trajectory generation device 10 as a component, a program for causing a computer to function as the trajectory generation device, and a non-transitional entity such as a semiconductor memory storing the program. The present disclosure can be realized in various forms, such as a dynamic recording medium and a trajectory generation method.

10: trajectory generation device, 12: trajectory generation unit, 16: determination unit, 18: policy change unit, 100: moving body, 102: axis, 120: target position, 210: trajectory, 230: rotation direction (trajectory)

Claims (12)

A trajectory generator (10) for generating a trajectory (210, 230) of a moving object (100) moving to a target position (120),
A trajectory generator (12, S402, S414) configured to search for and generate the trajectory by a predetermined algorithm;
A determination unit (16, S408) configured to determine whether the trajectory generation unit can generate the trajectory;
A policy change unit configured to change a search policy of the algorithm for searching for the trajectory when the determination unit determines that the trajectory generation unit cannot generate the trajectory (18, S410); When,
A trajectory generation device comprising: The trajectory generation device according to claim 1,
The determination unit is configured to determine that the trajectory generation unit cannot generate the trajectory when the amount of change in the position of the mobile body is equal to or less than a predetermined amount and the mobile body has not reached the target position. Have been
Orbit generator. The trajectory generating device according to claim 1 or 2,
The moving body is a robot arm having at least one axis (102);
When the difference between the operation limit angle of the axis and the rotation angle of the axis is equal to or less than a predetermined angle and the axis cannot rotate any more, when the robot arm has not reached the target position, The trajectory generating unit is configured to determine that the trajectory cannot be generated,
Orbit generator. The trajectory generation device according to claim 3, wherein
The policy changing unit, when the determination unit determines that the trajectory generation unit can not generate the trajectory, is configured to change the search policy by reversing the rotation direction of the axis,
Orbit generator. The trajectory generation device according to any one of claims 1 to 4,
The trajectory generating unit generates the trajectory by searching for a direction in which the cost of generating the trajectory is minimized,
The determination unit may determine that the trajectory generation unit cannot generate the trajectory when the change amount of the cost is equal to or less than a predetermined amount and the moving body is at a non-movable position that has not reached the target position. Is configured to
Orbit generator. The trajectory generation device according to claim 5, wherein
The policy change unit, when the determination unit determines that the moving object is at the immovable position, changes the cost gradient that changes when the moving object moves, and changes the search policy. It is configured,
Orbit generator. The trajectory generation device according to claim 6, wherein
The policy change unit is configured to set the cost of the immovable position to a value larger than surroundings when the determination unit determines that the moving object is at the immovable position,
Orbit generator. The trajectory generation device according to any one of claims 1 to 7,
The trajectory generation unit (S414) is configured to search and generate the trajectory in real time by a neural network based on the trajectory determined by the determination unit that the trajectory generation unit was able to generate the trajectory. ing,
Orbit generator. A trajectory generation method for generating a trajectory of a moving object moving to a target position,
Determine whether the trajectory can be searched and generated by a predetermined algorithm,
If the trajectory cannot be generated, search the trajectory by changing the search policy of the algorithm,
Orbit generation method. The trajectory generation method according to claim 9, wherein
If the amount of change in the position of the moving body is equal to or less than a predetermined amount and the moving body has not reached the target position, it is determined that the trajectory cannot be generated.
Orbit generation method. A trajectory generation program that generates a trajectory of a moving object that moves to a target position,
A trajectory generator (12, S402, S414) configured to search for and generate the trajectory by a predetermined algorithm;
A determination unit (16, S408) configured to determine whether the trajectory generation unit can generate the trajectory;
A policy change unit configured to change a search policy of the algorithm for searching for the trajectory when the determination unit determines that the trajectory generation unit cannot generate the trajectory;
Trajectory generation program that makes the computer function The trajectory generation program according to claim 11, wherein
The determination unit determines that the trajectory generation unit cannot generate the trajectory when the amount of change in the position of the mobile body is equal to or less than a predetermined amount and the mobile body has not reached the target position. Trajectory generation program that makes a computer function