A nonlinearity-based genetic algorithm for ship path planning

In this thesis, a nonlinearity-based genetic algorithm for ship path planning was proposed. This novel algorithm applied the nonlinear programming to compensate the inherent deficiency of genetic algorithm in local optimization. Meanwhile, problem-specific knowledge and heuristic knowledge were incorporated into encoding, evaluation and genetic operators of genetic algorithm. Furthermore, the application of minimum generation kept of the optimal fitness as the criterion of the termination condition reflected the knowledge accumulation in the optimal process which perfectly suited this nonlinearity-based genetic algorithm. Finally, the feasibility and effectiveness of this algorithm were evaluated with a set of test cases simulating various traffic scenarios.