On the comparison of niching strategies for finding the solution of multiniodal robot inverse kinematics functions

Abstract

Two niching strategies for the tournament selection, along with the simulated binary crossover operator and a parameter-based mutation operator are used to solve the inverse kinematics problem of a robotic manipulator. The niching strategies are compared on the basis of their ability to evaluate the multiple inverse kinematics solutions and maintain an acceptable distribution of individuals around these solutions. The approach requires the use of robot kinematics equations and the limits of joint angles only. It does not require the use of geometric heuristics as suggested by earlier researchers. The proposed approach provides the correct number of inverse kinematics solutions even when some of them may be unachievable due to the limits of the joint angles. In other approaches, all the inverse kinematics solutions are calculated and the unachievable ones are discarded. The total joint displacements associated with the multiple inverse kinematics solutions provide a means for multiplicity resolution.