Research On Improved Path Planning Algorithm Based On A-star

The location service application is an intelligent technology that integrates informationization,intelligence and industrialization,and represents an important benchmark for intelligent algorithm technology innovation.The core technology in location service application lies in the path planning algorithm in the underlying framework.As the core technology in the navigation system,path planning technology has been a research hotspot at home and abroad.Path planning algorithms include traditional algorithms such as Dijkstra algorithm,BFS algorithm,simulated annealing algorithm,fuzzy logic,and new algorithms represented by A~* algorithm based on heuristic information function.The A~* algorithm introduces the heuristic function as the auxiliary decision basis,which effectively improves the controllability and accuracy of the algorithm.However,the introduction of heuristic information functions in the A~* algorithm may lead to premature deletion of path nodes in the road network environment,and thus cannot ensure that each planned path is an optimal solution.In addition,as far as the node mesh in the A~* algorithm is concerned,the researcher performs a multi-level refinement operation on the node mesh to improve the accuracy of the algorithm,and the node mesh is divided too small,which causes the path length to be an exponential level increase.Furthermore,the A~* algorithm is a lossy algorithm and is not suitable for dynamic road network environment search.When the road density is uneven or there are multiple minimum values in the solution network,the optimal path solution cannot be guaranteed.In this paper,the following aspects are studied for the defects of the A~* algorithm:(1)The road network environment is dynamically modeled based on the grid environment,and the first pre-processing operation is performed on the road network environment.The obstacles in the environment are initialized by introducing an array of neighbor nodes and neighboring nodes,and stored in the form of key-value pairs.In the array,the subsequent traversal operation is optimized under the condition that the road network environment is unchanged,and the time loss is reduced.(2)The problem of classical path algorithm and A~* fusion operation is analyzed,and the binary idea is introduced as the basic idea of complete sorting.The open list and close list of the tree structure matching original algorithm suitable for A~* algorithm are designed to improve.Traversing efficiency.(3)In order to further optimize the operation efficiency of A~* algorithm,this paper defines the search range by adding distance and direction factors.The definition of distance and directionbased on the original valuation function is intended to reduce the map nodes with excessive deviation of search direction,and greatly reduce the number of search nodes.Free up computer resources.Simultaneous matching of the imported weighting factors can ensure that the weight coefficients of g(n)and h(n)interact with each other,and ensure that the variation of h(n)is positively correlated with itself,which is beneficial to improve node convergence during dynamic adjustment routing speed.(4)Through the research on the two-way A~* processing measures of early scholars,the parallelization processing method in the limited area is referenced and introduced,that is,whether the two open lists have the same effective node at the same time makes a reasonable judgment.Simulation experiments show that the method is feasible..