Research On Collision Avoidance Path Planning Technology Of USV Based On Improved Beetle Antenna Search Algorithm

As an important member of intelligent ship field,surface unmanned ship plays an irreplaceable role in many fields such as Marine patrol,military reconnaissance and resource exploitation The intelligent navigation control system is the key technology for unmanned ship to realize unmanned navigation,and the path planning,as the decision of unmanned ship,plans the feasible route for the autonomous navigation of unmanned ship,which is an important technical component of the unmanned ship intelligent navigation system Path planning algorithm is the core of the unmanned ship route automatically generated technology,combining with the longicorn must search algorithm for path planning of unmanned ship technology research,according to the characteristics of the unmanned ship sailing on the waters of the global path planning and local path planning algorithm is designed,the algorithm is verified by simulation experiment contrast and analysis in the surface of the water there is no boat on the feasibility and superiority of path planning.First,in this paper,the longicorn must search algorithm analyzes the features of the global path planning,a direct result of gradient algorithm using single intelligent search and there is no operation while the characteristics of the difficulty of algorithm is reduced but also lead to strong randomicity and poor stability of the algorithm,based on the above questions will be monomer longicorn search mechanism to improve the search mechanism for the colony and dynamic search step length Instead of fixed step length,and enhances the stability of the algorithm can improve the local search area at the same time.According to the characteristics of unmanned ship is not easy to substantially in the fitness function of algorithm design the maximum steering Angle of the constraint function,considering the path is generated line path and the inflection point out the discontinuous differentiable,using bezier path for the smooth processing.Finally,the simulation results show that the improved longhorn must search algorithm is more suitable for unmanned ship path planning than the original algorithm.Secondly,set up the simulation chart environment,under the environment of the chart will be improved algorithm with the current applied to the surface of the water path planning of unmanned ship artificial potential field method and fast search of the randomized tree algorithm for path planning simulation comparison,according to the path length maximum steering Angle and accumulated in the evaluation index,such as Angle of the superiority of the improved algorithm is verified The fuzzy PID controller is used to simulate the path generated by the three algorithms.The advantages of the improved algorithm are verified by the maximum tracking error and average tracking error of the tracking time.Finally search algorithm to improve the longicorn unmanned ship local path planning study,has established the unmanned ship and dynamic he dynamic collision model and extract the collision parameters DCPA and requirement,based on the DCPA and requirement of collision risk fuzzy mathematical model of combination of collision risk model and Marine collision convention algorithm of fitness function State when encounter with a local path planning Collision avoidance to resume after the stage design an unmanned ship to the original path function can make resume planning stage unmanned ship Courier back to the original path At last,through a variety of encounter situation of local path planning simulation and mixed path planning simulation verified the longicorn must search algorithm can undertake the local path planning of unmanned ship on the water surface.