Power-law fitness function scaling in the evolutionary method of path planning

Abstract

The main goal of the present paper is to discuss the application of the power-law fitness function scaling in the evolutionary method of path planning and its impact on the quality of final solution and the computation time necessary to solve the problem with the described scaling applied. In this paper we use the navigation problem of avoiding collision at sea for testing ground. A wide range of marine collision situations was provided to reveal advantages of power-law scaling application. The performed investigations have shown that the use of power-law scaling in the evolutionary path planning method preserves the diversity of solutions by a larger number of generations in the exploration phase. Results were also compared with a previous research that was performed using linear scaling.