Research On Underwater Vehicle Route Planning Based On Bat Algorithm

Route Planning is one of the key technologies for the autonomous underwater vehicle’s self-control and motion.According to certain optimization criteria,the main task is to search the optimal or sub-optimal route from the starting point to the ending point under the environment information is known or partially known.This route must avoid the obstacle or dangerous object in the planning area.Route planning actually is one of the optimization problems.Bat algorithm is one of the meta-heuristic optimization method proposed in recent years.At present,it has been widely used in scientific computing and engineering optimization problems.Practical results show bat algorithm is good at solve the non-linear,high-dimensional and complex multi-constrain optimization problems.Under the real three-dimensional seabed terrain environment,it carries out the research on the route planning method of underwater vehicle based on bat algorithm in this paper.The main research work is as follows.1.The principle,mathematical model and implementation process of the original bat algorithm is introduced.For the different update way of bat’s speed and position,the model I and model II bat algorithm is defined.Convergence of the bat algorithm is theoretically analyzed by using eigenvalue method.It proved that model I can’t converge,model II can converge under certain conditions.For model II bat algorithm,the scope of the parameter selection is given.Finally,further validate the theoretical analysis by simulation comparing experiments.2.For population size,inertia weight,pulse loudness factor,pulse rate factor and other parameters,its effect on the performance of bat algorithm is analyzed by numerical simulation.The optimum value of each parameter is given.3.For the convergence of the original bat algorithm can’t ensure,the size of the search area isn’t considered in the local search strategy,and lack the mutation strategy.Bat speed formula is redefined by introducing the inertia weight.The local search strategy is redefined by absorbing the mutation principle in the differential evolution algorithm.And a new disturb mutation strategy is introduced.A new improved bat algorithm is proposed in this paper.The simulation results show higher precision and faster convergence for the improved bat algorithm.4.By grid and interpolated the water depth data from electronic sea chart,the true three-dimensional seabed topography is build.Marine environment’s impact on the underwater vehicle is analyzed,and the evaluation function of the route is designed.Account for the marine environment,a new route planning method based on improved bat algorithm for underwater vehicle is proposed in this paper.The simulation is implemented under different planning environment.Simulation results show stronger environmental adaptability for the proposed method,which can plan a route correctly and effectively.Finally,the simulation comparing experiment is conducted under original and improved bat algorithm.The results show the improved bat algorithm has higher precision,faster convergence and better route planning.