Research On The Application Of Swarm Intelligence Optimization Algorithm In Unmanned Vehicle Path Planning In Complex Environmen

With the continuous progress of computer technology in recent years,unmanned vehicles have been widely applied to military and civilian fields.Path planning,as a key technology for controlling unmanned vehicles,has been a research hotspot for scholars at home and abroad.The swarm intelligence optimization algorithm optimizes the ground problem without having clear mathematical characteristics,which is suitable for solving such formally complex problems as path planning,but it still has the disadvantage of solving high-dimensional problems that are easily trapped in local optimality and the solution quality is not high.Therefore,an improved spider monkey optimization algorithm is proposed in this paper,and the algorithm is combined with an improved dynamic window method and applied to unmanned vehicle path planning.The specific research of this paper is as follows:(1)Analyzes the research background and significance of unmanned vehicle path planning as well as the current situation of unmanned driving technology and path planning algorithms at home and abroad,and points out the direction for the subject research of this paper.(2)A dynamic spider monkey optimization algorithm incorporating chaos and contrastive learning is proposed,introducing Tent chaotic sequences to initialize the population so that the initial population is more uniformly distributed in the solution space.The population diversity is evaluated after each iteration of the algorithm,and if the diversity is low,a contrastive learning strategy is randomly applied to some individuals in the population to avoid premature convergence of the algorithm.Non-linear dynamic adaptive weights are introduced in the local leader stage and local leader decision stage,which better balance the global exploration and local exploitation ability.And the elite contrastive learning strategy is applied in the local leader decision stage to enhance the population diversity.The improved spider monkey optimization algorithm in this paper is tested using nine standard test functions in CEC2017 and compare it with the same kind of algorithm,the experimental results show that the improved algorithm in this paper has good adaptive ability and stability,and effectively overcomes the problem that the standard spider monkey optimization algorithm does not optimize well in middle and high dimensions.(3)For the first time,the improved spider monkey optimization algorithm is applied to the short-and medium-range global path planning problem by fusing chaos and contrastive learning.In the two-dimensional environment model established using the raster method,the improved spider monkey optimization algorithm is used for path planning,and the optimal path is selected by considering the safety,smoothness and path length of the path,and the planned path is obtained and smoothed by using a cubic B spline curve.In the raster environment with different sizes and obstacle distribution,the simulation experiments compared with four intelligent optimization algorithms show that the improved spider monkey optimization algorithm plans better quality paths and improves the planning efficiency.(4)To improve the smoothness and purpose-orientedness of the path planning by the dynamic window method,an improved dynamic window method is proposed and implemented.Combining the global path planning of the spider monkey optimization algorithm and the local path planning of the improved dynamic window method,a hybrid path planning algorithm is proposed,which is guided by the global path information for local path planning and can find a safe and feasible route without collision in the complex environment with a combination of dynamic and static obstacles.Simulation experiments in two-dimensional raster environment and three-dimensional CARLA environment prove that the proposed algorithm is a practical and effective method to solve the path planning problem of unmanned vehicles.