Path planning of mobile robot based on compound shape and simulated annealing hybrid algorithm

Abstract

To deal with the problem that the convergence rate of simulated annealing algorithm is very slow, a new hybrid optimal algorithm, based on compound shape combined with simulated annealing algorithm, is proposed, and is successfully applied to the path planning of the neural network of mobile robot. The advantages of this algorithm lie in the fact that it can not only achieve the global optimal solution which characterizes simulated annealing algorithm, but also search for an ideal descending direction, thus, the convergence rate being improved. The simulation experiments demonstrate that the computation of this algorithm is simple, the convergence rate is fast and the robustness of initial value is good also