Research On Unmanned Driving Dynamic Path Planning Method Based On Vehicle Ad Hoc Network

Path planning is the top module of the driverless control plan,guiding the vehicle to travel at a macro level.The existing path planning two-way A~* algorithm does not consider the traffic flow state factor,and the planned path space distance is the shortest,but may pass through the congestion and accident sections,resulting in a long travel time.In the vehicle-borne network environment,the traffic flow state information is collected,and the inadequacies of the dual A~* algorithm are improved,effectively guiding the vehicle to avoid congestion and accident sections,reducing travel time,and improving path planning efficiency.Based on the research and analysis of the dynamic path planning method,the two-way A~* algorithm with high search efficiency and good performance is selected as the basic model.The traffic flow state factor is introduced and a dynamic two-way A~* algorithm is established.The vehicle is dynamically planned according to traffic conditions.path.In order to verify the theoretical feasibility of the improved algorithm,theoretical simulation analysis was carried out under three traffic flow states of low density,medium density and high density.Theoretical simulation experiments show that the dynamic bidirectional A~* algorithm is significantly better than the traditional bidirectional A~* algorithm.In the low-density,medium-density and high-density traffic flow state,the travel path is reduced by 7.14%,23.52%,and 10%,respectively.After the theoretical verification of the improved algorithm is feasible,it is proposed to transform from the theoretical simulation environment to the vehicular self-organizing network environment closer to the real complex traffic conditions,and further verify the performance advantages of the improved algorithm.In the vehicle-borne network environment,the travel time is used as the performance index of the evaluation dynamic algorithm,and the travel time weight is used as the primary key value of the link attribute.The travel time evaluation function is designed as the dynamic path planning search rule.In order to carry out simulation verification under the vehicle self-organizing network environment,a two-way coupling joint simulation experiment platform was established by using the network simulation tool OMNeT++and the microscopic traffic simulation tool SUMO.The simulation results based onthe Venet environment show that the dynamic bidirectional A~* algorithm is significantly better than the traditional bidirectional A~* algorithm.In the low-density and medium-density traffic flow state,the travel time is reduced by 4.90% and 7.24%,respectively.The dynamic bidirectional A~* path planning algorithm and the coupled Vanet simulation experimental platform established in this paper provide theoretical basis and practical significance for the path planning of intelligent vehicle self-organizing network.