Granted Patent B2

US 12,228,893 · App. 17/635,585 · Granted Feb 18, 2025

Inventor: Hiroyuki Oyama (Tokyo, JP)

Assignee: NEC CORPORATION

G05B13/042B25J13/082G05B13/024

Abstract

Claims (51)

1. An information processing device comprising:

at least one memory storing instructions; and

at least one processor configured to execute the instructions to:

acquire measurement information indicating measurement results in a workspace where a robot performs a task;

acquire abstract model information indicating:

an abstract model in which the measurement results in the workspace are abstracted by a hybrid system, the hybrid system mixing continuous and discrete dynamics; and

switching conditions of the continuous and discrete dynamics in the hybrid system;

generate the abstract model based on the abstract model information and the measurement information;

acquire a target logical formula in temporal logic representing a final state to be achieved in performing the task;

acquire a time step logical formula indicating a state for each time step to perform the task;

determine a control input for controlling the robot for each time step, based on the abstract model, the target logical formula, and the time step logical formula; and

control the robot in accordance with the determined control input.

2. The information processing device according to claim 1 ,

wherein the at least one processor is configured to generate the abstract model based on:

information relating to the robot included in the measurement information; and

information relating to a target object subjected to operation by the robot.

3. The information processing device according to claim 1 ,

wherein the at least one processor is configured to, in a case where the robot performs a task relating to pick-and-place, generate the abstract model of the hybrid system in which an operation of grasping a target object by the robot is abstractly expressed by a logical variable.

4. The information processing device according to claim 1 ,

wherein the at least one processor is configured to determine an evaluation function in accordance with the task, and

wherein the at least one processor is configured to determine the control input which optimizes the evaluation function while satisfying the abstract model and the time step logical formula.

5. The information processing device according to claim 1 ,

wherein the at least one processor is configured to generate the control input for time steps required for an occurrence of an event that is necessary to occur by time of completion of the task and generate the control input every time the event occurs.

6. The information processing device according to claim 1 ,

wherein the at least one processor is further configured to generate the time step logical formula with a number of time steps necessary to complete the task or required for an occurrence of an event that is necessary to occur by time of completion of the task.

7. The information processing device according to claim 6 ,

wherein the at least one processor is configured to set the number of the time steps to a predetermined initial value and increase the number of the time steps until the time step logical formula with which the control input is able to be determined is generated.

8. The information processing device according to claim 7 ,

wherein the at least one processor is configured to set the initial value to a value smaller than the number of the time steps corresponding to a prospective work time for the task, the prospective work time being determined by a user.

9. The information processing device according to claim 1 ,

wherein the at least one processor is further configured to generate, on a basis of the control input for each time step, a sequence of sub-tasks that are each a task in units of command that the robot can accept.

10. A control method performed by an information processing device and comprising:

acquiring measurement information indicating measurement results in a workspace where a robot performs a task;

acquiring abstract model information indicating:

an abstract model in which the measurement results in the workspace are abstracted by a hybrid system, the hybrid system mixing continuous and discrete dynamics; and

switching conditions of the continuous and discrete dynamics in the hybrid system;

generating the abstract model based on the abstract model information and the measurement information;

acquiring a target logical formula in temporal logic representing a final state to be achieved in performing the task;

acquiring a time step logical formula indicating a state for each time step to perform the task;

determining a control input for controlling the robot for each time step, based on the abstract model, the target logical formula, and the time step logical formula; and

controlling the robot in accordance with the determined control input.

11. A non-transitory computer readable storage medium storing a program executed by a computer, the program causing the computer to:

acquire measurement information indicating measurement results in a workspace where a robot performs a task;

acquire abstract model information indicating:

an abstract model in which the measurement results in the workspace are abstracted by a hybrid system, the hybrid system mixing continuous and discrete dynamics; and

switching conditions of the continuous and discrete dynamics in the hybrid system;

generate the abstract model based on the abstract model information and the measurement information;

acquire a target logical formula in temporal logic representing a final state to be achieved in performing the task;

acquire a time step logical formula indicating a state for each time step to perform the task;

determine a control input for controlling the robot for each time step, based on the abstract model, the target logical formula, and the time step logical formula; and

control the robot in accordance with the determined control input.

Assignments (1)

Continuity (1)

References Cited (24)