公开(公告)号：US20250021107A1

公开(公告)日：2025-01-16

申请号：US18266643

申请日：2022-10-18

Applicant: SHANDONG UNIVERSITY

Inventor： Wei ZHANG ,  Yanyun CHEN ,  Jiapeng SHENG ,  Xing FANG ,  Wenhao TAN ,  Ran SONG ,  Xiaolei LI ,  Jiyu CHENG

IPC: G05D1/495 ,  G05D1/60 ,  G05D101/15 ,  G05D109/12

Abstract: A method and a system for rhythmic motion control of a robot based on a neural oscillator, including: acquiring a current state of the robot, and a phase and a frequency generated by the neural oscillator; and obtaining a control instruction according to the acquired current state, phase and frequency and a preset reinforcement learning network so as to control the robot. The preset reinforcement learning network includes an action space, a pattern formation network and the neural oscillator. A control structure designed by the present disclosure, which is composed of the neural oscillator and the pattern formation network, can ensure formation of an expected rhythmic motion behavior; and meanwhile, a designed action space for joint position increment can effectively accelerate the training process of rhythmic motion reinforcement learning, and solve a problem that design of the reward function is time-consuming and difficult in learning with existing model-free reinforcement learning.

公开(公告)号：US20250021109A1

公开(公告)日：2025-01-16

申请号：US18266694

申请日：2022-10-18

Applicant: SHANDONG UNIVERSITY

Inventor： Wei ZHANG ,  Jiapeng SHENG ,  Yanyun CHEN ,  Xing FANG ,  Wenhao TAN

IPC: G05D1/646 ,  B62D57/032

Abstract: A method for controlling motions of a quadruped robot based on reinforcement learning and position increment, including: acquiring motion environment information, quadruped robot attitude information, and foot sole position information; based on the acquired information, generating foot sole positions of the quadruped robot during motions within all preset time steps, and calculating a change of the foot sole positions in all the time steps; taking a maximum moving distance within a single time step as a constraint, and accumulating the time steps at the same time to obtain a foot sole position trajectory; and controlling the quadruped robot to perform corresponding actions based on the foot sole position track combined with a preset reward function, so as to keep motion balance of the quadruped robot.