Parallel Evolutionary Programming Optimisation for Evader’s Trajectory in Pursuit-Evasion Game

The paper discusses the use of parallel evolutionary programming optimisation algorithm to help pilot or to guide an UCAV to avoid interception against a very agile and fast pursuer such as a missile in full three dimensions. The use of evolutionary optimisation algorithm opens the possibility to do optimisation in a complete three dimension, in which established methods such as differential games and non-linear programming suffers from singularity, surface discontinuity, discrete and time consuming. The idea is that it is not necessary to fully search for global optimal solution (which may take forever) for n! problems, but rather a sub-optimal solution would suffice. This algorithm has been successfully tested on a single CPU computer but it takes longer time to search for optimal solutions. The use of multiple CPUs in parallel is found to improve the execution time as well as the quality of the solutions. In this paper we study the effect parallel computation, the population size and the number of generations on the quality of the solution and the processing time. The results shows the use of multiple CPUs in parallel is able to improve both in the quality of the solution and the execution time