A Genetic Algorithm-Inspired UUV Path Planner Based on Dynamic Programming

Path planning can be viewed as an optimization process in which an optimum path between two points is to be found under some predefined constraints. Some typical constraints are path length, fuel consumption, and path safety factor. Exact algorithms such as linear programming (LP) and dynamic programming (DP) are widely adopted in vehicle maneuvering systems. However, as the problem domain scales up, exact algorithms suffer from high computational complexity. In contrast, metaheuristic algorithms such as evolutionary algorithms (EA) and genetic algorithms (GA) can provide suboptimum solutions without the full understanding of the problem domain. Metaheuristic algorithms are capable of providing decent solutions within a finite period of time, even for large-scaled problems. In this paper, a GA-inspired unmanned underwater vehicle (UUV) path planner based on DP is proposed. Simulation results show that the proposed algorithm can outperform a GA-based UUV path planner in terms of speed and solution quality.