Trajectory planning of a robot using learning algorithms

The authors consider the problem of a robot manipulator operating in a noisy workspace. The manipulator is required to move from an initial position P_i to a final position P _f. P_i is assumed to be completely defined. However, P_f is obtained by a sensing operation and is assumed to be fixed but unknown. The authors approach to this problem involves the use of three learning algorithms, the discretized linear reward-penalty (DL_R-P) automaton, the linear reward-penalty (L_R-P) automaton and a nonlinear reinforcement scheme. An automaton is placed at each joint of the robot and by acting as a decision maker, plans the trajectory based on noisy measurements of P_f