Path Planning Algorithm With Security Constraints For Intelligent Vehicle

As everyone knows,path planning is one of the core technology of an automatic driving system.A safe and reliable path is the basic premise for the safe driving of intelligent vehicles.Because the path planning is based on environment and the vehicle’s own posture information,and the environment is complex and changeable,it has a great impact on the safety and driving efficiency of vehicles.Therefore,it is of great significance for the automatic driving of vehicles to study a path planning method that can not only ensure vehicle safe but also make the vehicle travel with high efficiency.In order to plan a safe and reliable path,the representation method and model of intelligent vehicle’s safety should be studied.To solve this problem,a threat assessment model basing on potential field is constructed to evaluate the impact of external environment factors on the safety of intelligent vehicles in this thesis.Firstly,environmental factors closely related to vehicle safety are extracted,including object characteristics,road environment and kinematic factors.Secondly,the fuzzy reasoning method is used to calculate the threat coefficient.And the static threat potential field model is constructed.Since vehicle kinematic factors have an important influence on the threat assessment model,kinematic factors is added to construct a dynamic threat potential model based on the static threat potential field to improve the accuracy of the model.Finally,experiments prove that the threat assessment model in this thesis is effective.In order to balance the high safety and high driving efficiency of the vehicle,a path planning method with security constraints based on the threat assessment model is proposed in this thesis.Firstly,a safety-constrained path planning model is established according to the requirements.And it ensures that the vehicle travels in a safe area by establishing safety constraints.Then,the objective function with the shortest time is constructed.For the solution of the model,it combines the threat potential field and the grid map to construct a multi-value matrix grid map.At the same time,it improves the direction priority sequence selection algorithm and uses the improved algorithm to search in the map to plan the optimal path.Finally,the path is optimized to meet the requirements of vehicle control smoothness.The validity of this path planning method is verified by simulation experiments in this thesis.According to the research on the path planning method with security constraints,a path planning system is developed,including threat assessment module,map building module,path planning module and path smoothing module.The actual vehicle experiments verify that the research method and the system is effective.