Unmanned Surface Vehicle Path Planning Based On Bacterial Foraging Optimization Algorithm

Unmanned surface vehicle(USV)is a highly integrated surface intelligence platform,its working scene covers oceans,rivers,lakes,freshwater reservoirs,and other waters.The USV has been successfully applied in a large number of military and civilian fields with a high degree of autonomy and versatility.The working environment of the USV is complex and changeable.A high precision global path planning method and a fast response local path planning method are designed for USV to cope with the complex and changeable working environment.The high-quality path planning not only ensures the security of the USV,but also further enhances its intelligence level.The challenge faced by traditional methods for solving path planning problems is that as the scale of feasible solutions increases,the search space increases significantly and the search efficiency decreases continuously.The bacterial foraging optimization(BFO)algorithm has fine search characteristics and strong global search ability.Focusing on the research problem of the USV path planning,this dissertation has done some work,including the fusion method of BFO algorithm and environment map,the efficient method of global path planning,and the method of local path planning considering multiple constraints.Several novel BFO algorithms are proposed to obtain high-quality optimization results for the problem.The main work and innovations are as follows:(1)Aiming at the fusion of the BFO algorithm and the environment map,an ASBFO algorithm(A Star-Bacterial Foraging Optimization Algorithm,AS-BFO)is proposed,introducing the cost function of the A* algorithm.The patch notes are generated around the discontinuity points to repair discontinuous paths generated when the bacteria perform chemotaxis,and the repaired path is guaranteed to be the optimal one.The Morris method is used to analyze the global sensitivity of the main parameters of the AS-BFO algorithm,and the influence degree of each parameter on the model output is determined.The simulation experiments are carried out on environment maps of various scales and obstacles.The results indicate that the AS-BFO algorithm has good compatibility with the environment map.(2)Aiming at the global path planning problem of the USV,a multi-subpopulation bacterial foraging optimization algorithm(MS-BFO)is proposed.The initial population is divided regularly to improve the uniform distribution of the initial solution.The deletion and the immigration strategies are introduced into the MS-BFO algorithm to realize communication between sub-populations and the co-evolution of multiple groups.The internal counter is implanted in the traditional BFO algorithm to further improve search efficiency.The simulation results show that the MS-BFO algorithm has good stability and accuracy in various environments.(3)Aiming at the local path planning problem of the USV,an SA-BFO algorithm(Simulated Annealing-Bacterial Foraging Optimization Algorithm,SA-BFO)is proposed.The simulated annealing mechanism is introduced in the migration operation of the traditional BFO algorithm so that it can better jump out of the local extremum and converge to the global optimum.The COLREGS and the dynamic obstacle types are considered when the proposed algorithm plans the local paths.The simulation results show that the SA-BFO algorithm can plan an efficient and safe local path.(4)Aiming at the testing problems of the USV path planning method before the practical application,a visual test platform with strong operability and high simulation degree is established,and a collision avoidance process based on the collision risk assessment and the improved BFO algorithms is designed.The inland port waters,the lake waters,and the coastal port waters were selected as the test environments.The global path planning based on the MS-BFO algorithm,the local path planning based on the SA-BFO algorithm with multiple constraints,and the test experiment of the USV encountering multiple ships were realized.The experimental results show that the USV can successfully avoid the various obstacles in the above environments,and efficiently plan safe and collision-free paths.