公开(公告)号：US20240123614A1

公开(公告)日：2024-04-18

申请号：US18278305

申请日：2021-02-26

Applicant: NEC Corporation

Inventor： Rin TAKANO ,  Hiroyuki OYAMA

IPC: B25J9/16

CPC classification number: B25J9/163 ,  B25J9/161

Abstract: A learning device 1X mainly includes an optimization problem calculation means 51X and an executable state set learning means 52X. The optimization problem calculation means 51X calculates a function value to be a solution for an optimization problem which uses an evaluation function for evaluating reachability to a target state, based on an abstract system model and a detailed system model concerning a system in which a robot operates. The executable state set learning means 52X learns an executable state set of an action of the robot to be executed by a controller based on a function value.

公开(公告)号：US20240139950A1

公开(公告)日：2024-05-02

申请号：US18278286

申请日：2021-02-26

Applicant: NEC Corporation

Inventor： Hiroyuki OYAMA ,  Rin TAKANO

IPC: B25J9/16

CPC classification number: B25J9/1666 ,  B25J9/163

Abstract: A constraint condition learning device 1X mainly includes a conversion means 15X and a constraint condition estimation means 16X. The conversion means 15X converts first time series data regarding a state and an input of a robot system during an execution period of a task executed by the robot system into second time series data represented by propositions. The constraint condition estimation means 16X estimates a constraint condition on the task as a logical formula, based on the second time series data and the information regarding whether or not the task succeeded.

公开(公告)号：US20250165860A1

公开(公告)日：2025-05-22

申请号：US18841436

申请日：2022-03-01

Applicant: NEC Corporation

Inventor： Rin TAKANO ,  Hiroyuki OYAMA

IPC: G06N20/00

Abstract: A learning device performs learning of a value of a meta parameter based on training data. The meta parameter indicates a probability distribution in a learning model in which a value of a parameter follows the probability distribution. The training data represents input and output in the learning model. The learning device calculates an evaluation value indicating an evaluation of a generalization error of the learning model. The learning device determines, based on the evaluation value, whether or not it is necessary to continue the learning of the value of the meta parameter.

公开(公告)号：US20250164944A1

公开(公告)日：2025-05-22

申请号：US18840962

申请日：2022-03-01

Applicant: NEC Corporation

Inventor： Hiroyuki OYAMA ,  Rin TAKANO

IPC: G05B13/04 ,  G05B13/02

Abstract: A learning device selects, from among search points indicating an operation of a control target, a search point to be subjected to training data acquisition for learning of a control of the control target. The learning device calculates information indicating an evaluation of whether or not an operation indicated by the selected search point is executable, and an output value for the operation indicated by the selected search point to be output by a controller for controlling the control target. The learning device acquires, based on the selected search point, the information indicating the evaluation of whether or not the operation indicated by the selected search point is executable, and the output value for the operation indicated by the selected search point to be output by the controller, training data for learning a control of the control target that is performed by the controller.

公开(公告)号：US20250130562A1

公开(公告)日：2025-04-24

申请号：US18688892

申请日：2021-09-22

Applicant: NEC Corporation

Inventor： Hiroyuki OYAMA ,  Rin TAKANO

IPC: G05B23/02

Abstract: A constraint condition acquisition device acquires success time time-series data, which is time-series data pertaining to control over a control target when a predetermined task carried out through the control is successful, and failure time time-series data, which is the time-series data in the case of task failure; acquires a constraint condition template, which is a constraint condition that includes a parameter; and determines the value of the parameter such that the constraint condition holds in the success time-series data and the constraint condition does not hold in the failure time-series data.

公开(公告)号：US20230364792A1

公开(公告)日：2023-11-16

申请号：US18029578

申请日：2020-10-09

Applicant: NEC Corporation

Inventor： Rin TAKANO ,  Hiroyuki Oyama

IPC: B25J9/16

CPC classification number: B25J9/1664

Abstract: The operation command generation device 1Y mainly includes a skill information acquisition means 341Y, a skill tuple generation means 342Y, and a skill use operation command generation means 343Y. The skill information acquisition means 341Y is configured to acquire skill information relating to a skill to be used in a motion planning of a robot. The skill tuple generation means 342Y is configured to generate, based on the skill information, a skill tuple which is a set of variables in a system model, the variables being associated with the skill, the system model being set in the motion planning. The skill use operation command generation means 343Y is configured to generate a skill use operation command that is a temporal logic command representing an operation corresponding to the skill tuple.

公开(公告)号：US20230364791A1

公开(公告)日：2023-11-16

申请号：US18029261

申请日：2020-10-09

Applicant: NEC Corporation

Inventor： Rin TAKANO ,  Hiroyuki OYAMA

IPC: B25J9/16 ,  B25J13/08

CPC classification number: B25J9/1664 ,  B25J13/089

Abstract: The temporal logic formula generation device 1X mainly includes a target relation logical formula generation means 331X and a target relation logical formula integration means 332X. The target relation logical formula generation means 331X is configured to generate, based on object-to-object relation information representing a relation between objects in a target state relating to a task of the robot, one or more target relation logical formulas that are temporal logic formulas representing relations, in the target state, between respective pair(s) of objects between which the relation is defined. The target relation logical formula integration means 332X is configured to generate a temporal logic formula into which the target relation logical formulas are integrated.