Research On Collision Avoidance Planning For Unmanned Surface Vessel Based On Improved Bacterial Foraging Algorithm

Entering the 21 st century,countries around the world attach more and more importance to maritime rights and interests,China has clearly proposed the national strategy: “safeguarding maritime rights and interests,building a maritime power”.Unmanned Surface Vessel(USV)is a new type of intelligent unmanned surface naval vessel capable of autonomous planning and navigation,its characteristics are small hull,fast speed,strong stability,high safety and strong concealment.USV can execute many tasks that are not suitable for manned vessels in the complex surface environment,and play a significant role in the exploration and exploitation of national marine resources,maritime military conflicts and the construction of maritime power strategy.Collision avoidance planning technology is the core technology of USV,including global path planning and local collision avoidance planning.Collision avoidance planning algorithm is the key to realize the global path planning and local collision avoidance planning of USV,and also the key to improve the intelligent level of USV.Bacterial Foraging Optimization(BFO)algorithm is a novel swarm intelligence optimization algorithm.In order to improve the intelligence,real-time performance and stability of collision avoidance planning algorithm,in this paper,the BFO algorithm is improved,and an Improved Bacterial Foraging Optimization(IBFO)algorithm is proposed,then the collision avoidance planning of USV based on the Improved Bacterial Foraging algorithm is studied.Firstly,the improvement strategies of BFO algorithm are studied in this paper.Through analysing the steps of basic BFO algorithm,proposing the shortcomings existing in the steps,then improving the chemotactic step,swimming,replication and migration of BFO algorithm according to the shortcomings,proposing IBFO algorithm,finally the algorithm performance is tested and compared in benchmark functions.The simulation results show that IBFO algorithm has higher convergence rate,better optimization accuracy,better global optimization capability and stronger stability.Secondly,realizing the global path planning of USV by simulating the bacterial chemotaxis.The potential field environment model of USV is established by analogy with the bacteria foraging environment,establishing the perception model and motion strategy of USV by simulating bacterial chemotaxis,improving the moving step of USV and designing the target function based on chemotaxis and obstacle avoidance,then searching for the optimal forward direction of USV through IBFO algorithm,realizing the global path planning of USV,at last,the simulation tests show that the method is effective.Finally,studying local collision avoidance planning of USV based on IBFO algorithm.Obtaining the information of dynamic ship obstacles through Automatic Identification System(AIS),according to adopting the collision avoidance strategy of adjusting speed and heading at the same time,designing the target function of collision avoidance optimization combined with local collision avoidance model,then finding the optimal strategy by applying IBFO algorithm,in the end,carrying out the simulation of local collision avoidance,the results show that USV can avoid multiple dynamic ship obstacles successfully and realize local collision avoidance quickly and safely by adopting this method.