Title :
Motion planning of truss-type robot: acquisition and representation of criterion by modular neural network
Author :
Tanaka, Masao ; Hanahara, Kazuyuki
Author_Institution :
Dept. of Mech. Eng., Osaka Univ., Japan
Abstract :
Discusses a neural network approach to the acquisition of the criterion for the motion planning of the truss-type robot. The modular formation is employed for the motion planning neural network in order to represent the acquired criterion as well as the kinematical constraint. The feasibility and the effectiveness of the proposed approach is demonstrated through the simulation studies for the truss-type robot of helical mast
Keywords :
neural nets; planning (artificial intelligence); position control; robots; helical mast; kinematical constraint; modular neural network; motion planning; truss-type robot; Control theory; Mathematical programming; Mechanical engineering; Mechanical systems; Motion planning; Multi-layer neural network; Neural networks; Robot motion; Robot sensing systems; Topology;
Conference_Titel :
Intelligent Robots and Systems '91. 'Intelligence for Mechanical Systems, Proceedings IROS '91. IEEE/RSJ International Workshop on
Conference_Location :
Osaka
Print_ISBN :
0-7803-0067-X
DOI :
10.1109/IROS.1991.174587
