Title :
Repetitively structured cascade neural network model which generates an optimal arm trajectory
Author :
Uno, Yoji ; Kawato, Mitsuo ; Maeda, Yoshiharu ; Suzuki, Ryoji
Author_Institution :
Dept. of Math. Eng. & Inf. Phys., Tokyo Univ., Japan
Abstract :
The trajectory along which a human moves his arms tends to minimize the time integral of the square of the rate of change of torque. A neural network model for trajectory formation, based on the minimum torque change criterion is proposed. The neural network first acquires a forward dynamics model of a controlled object by training, and then it calculates the motor command by a relaxation computation, utilizing the learned forward dynamics model. The model can be applied to many types of ill-posed motor control problems, for example, inverse kinematics and inverse dynamics for redundant controlled objects
Keywords :
cascade control; dynamics; neural nets; optimal control; position control; cascade neural network model; forward dynamics model; ill-posed motor control problems; inverse dynamics; inverse kinematics; minimum torque change criterion; optimal arm trajectory; redundant controlled objects; repetitively structured model; time integral minimization; torque change rate; Arm; Computer networks; Humans; Laboratories; Muscles; Neural networks; Neurons; Physics; Torque; Visual perception;
Conference_Titel :
Decision and Control, 1989., Proceedings of the 28th IEEE Conference on
Conference_Location :
Tampa, FL
DOI :
10.1109/CDC.1989.70453