Robot control system, robot control method, and recording medium
AssigneeNEC Corporation
InventorsMasatsugu Ogawa, Hisaya Wakayama, Nobuharu Kami
Life of this patent
Estimating the term…
Abstract
In a robot control system, a storage unit stores a plurality of programs corresponding to tasks for a robot to perform. A reception unit receives selections made by a user, which designate a task for the robot to perform and attribute information related to a program causing the robot to perform the task among the plurality of programs. An acquisition unit acquires the program for performing the task among the plurality of programs from the storage unit based on the task and the attribute information. A robot control unit controls the robot in accordance with the acquired program.
Claims11
A robot control system comprising:
- a memory storing instructions;
- and
- one or more processors configured to execute the instructions to:
- store a plurality of programs in the memory, the plurality of programs corresponding to tasks for an arm-type robot;
- receive selections made by a user, which designate a task for the robot to perform and attribute information related to a target program causing the robot to perform the task, from among the plurality of programs;
- acquire the target program for performing the task from the memory based on the task and the attribute information;
- control the robot in accordance with the target program as has been acquired;
- and
- perform learning of an environment of the robot while the robot is controlled by executing the target program and while preventing the robot from hitting an obstacle, to update the target program,
- wherein the learning is performed based on an optimization algorithm,
- wherein the optimization algorithm uses an interference function indicating an interference state between the obstacle and the robot, and
- wherein the optimization algorithm includes a constraint indicating that the robot is not to hit the obstacle.
A robot control method by performed by a computer and comprising:
- storing a plurality of programs in a memory, the plurality of programs corresponding to tasks for an arm-type robot;
- receiving selections made by a user, which designate a task for the robot to perform and attribute information related to a target program causing the robot to perform the task, from among the plurality of programs;
- acquiring the target program for performing the task from the memory based on the task and the attribute information;
- controlling the robot in accordance with the target program as has been acquired;
- and
- performing learning of an environment of the robot while the robot is controlled by executing the target program and while preventing the robot from hitting an obstacle, to update the target program,
- wherein the learning is performed based on an optimization algorithm,
- wherein the optimization algorithm uses an interference function indicating an interference state between the obstacle and the robot, and
- wherein the optimization algorithm includes a constraint indicating that the robot is not to hit the obstacle.
A non-transitory computer-readable recording medium storing instructions executable by a computer to perform a process comprising:
- storing a plurality of programs in a memory, the plurality of programs corresponding to tasks for an arm-type robot;
- receiving selections made by a user, which designate a task for the robot to perform and attribute information related to a target program causing the robot to perform the task, from among the plurality of programs;
- acquiring the target program for performing the task from the memory based on the task and the attribute information;
- controlling the robot in accordance with the target program as has been acquired;
- and
- performing learning of an environment of the robot while the robot is controlled by executing the target program and while preventing the robot from hitting an obstacle, to update the target program,
- wherein the learning is performed based on an optimization algorithm,
- wherein the optimization algorithm uses an interference function indicating an interference state between the obstacle and the robot, and
- wherein the optimization algorithm includes a constraint indicating that the robot is not to hit the obstacle.
Dependencies are inferred from the claim text.