An improved path planner based on adaptive genetic algorithm for autonomous underwater vehicle

In allusion to the problem of global path planning for autonomous underwater vehicle (AUV) on an environment of large-scale chart data, an improved adaptive genetic algorithm (AGA) is proposed in this paper. In the improved AGA, it is adopted the grid-based approach for environment model and variable length codes method, and designs five kinds of genetic operators, an adaptive probability algorithm of crossover related with evolution generations, an adaptive probability algorithm of mutation related with evolution generations and fitness, and furthermore, an adaptive emigration algorithm based on prematurity estimation. The simulation result shows that all these methods can help to enhance the capability of the AGA and make it had the excellent character of good stability and high speed global convergence, the path described simply and clearly, and it can more efficiently and effectively solve the problem of path planning for AUV. This improved AGA planner can satisfies the demand of real-time for system.