Manipulator Redundancy Reduction as a Tool for Reinforcing Motion Planning Using Genetic Algorithms

A novel approach to plan an optimum motion of redundant robot manipulators for a predefined end-effector trajectory using genetic algorithms (GA) is presented. The efficiency of the proposed approach, without loss of generality, is demonstrated through a simulation carried out on a planar 6-DOF robot manipulator. The approach benefits from two key features. First, the method of data representation which guarantees the satisfaction of joints angle limits, and second the conversion of considered model’s 6-DOF construction to 4-DOF construction along with an additional binary value which guarantees the exact placement of the end-effector on the predefined trajectory. Comparison with three other approaches shows that the result of the presented solution is substantially better. In addition the difference of two kinds of Random Number Generator (RNG) is addressed. It is shown that using RNG with normal distribution leads to faster convergence of the proposed algorithm than RNG with uniform distribution.