Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
PARTICLE SWARM OPTIMIZATION AND REINFORCEMENT LEARNING ALGORITHM-BASED DYNAMIC WALKING CONTROL SYSTEM FOR BIOMIMETIC BIPED ROBOT
Document Type and Number:
WIPO Patent Application WO/2017/181319
Kind Code:
A1
Abstract:
The invention relates to the technical field of biomimetic humanoid robot walking control. Provided is a particle swarm optimization and reinforcement learning algorithm-based dynamic walking control system for a biomimetic biped robot. The dynamic walking control system comprises: a motion mode control center; a central mode control center; an action stage control center; a postural reflex control center; and a joint unit. A modulation signal output terminal of the motion mode control center is connected to a modulation signal input terminal of the joint unit. A stimulus signal output terminal of the motion mode control center is connected to a stimulus signal input terminal of the central mode control center. An excitation signal output terminal of the central mode control center is respectively connected to an excitation signal input terminal of the action stage control center and an excitation signal input terminal of the joint unit. An action signal output terminal of the action stage control center is connected to an action signal input terminal of the postural reflex control center. The control system has a high stability, high efficiency and high adaptivity.

Inventors:
WU WEI (CN)
LIU QI (CN)
Application Number:
PCT/CN2016/079543
Publication Date:
October 26, 2017
Filing Date:
April 18, 2016
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV JIANGNAN (CN)
International Classes:
B62D57/032
Foreign References:
CN201208992Y2009-03-18
CN102202613A2011-09-28
CN105446345A2016-03-30
CN1290590A2001-04-11
EP2574527A22013-04-03
CN101751037A2010-06-23
Attorney, Agent or Firm:
HARBIN SHINEIP INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: