An efficient hybridization of Genetic Algorithms and Particle Swarm Optimization for inverse kinematics

This paper presents a novel biologically-inspired approach to solving the inverse kinematics problem efficiently on arbitrary joint chains. It provides high accuracy, convincing success rates and is capable of finding suitable solutions for full pose objectives in real-time while incorporating joint constraints. The algorithm tackles the problem by evolutionary optimization and merges the benefits of genetic algorithms with those of swarm intelligence which results in a hybridization that is inspired by individual social behaviour. A multi-objective fitness function is designed which follows the principle of natural evolution within continually changing environments. A further simultaneous exploitation of local extrema then allows obtaining more accurate solutions where dead-end paths can be detected by a simple heuristic. Experimental results show that the presented solution performs significantly more robustly and adaptively than traditional or various related methods and might also be applied to other problems that can be solved by optimization techniques.