Research On Path Planning Of Intelligent Vehicle Based On Potential Field Ant Colony Algorithm

With the rapid development of Internet technology,intelligent vehicle is gradually coming into public view.The emergence of intelligent vehicle not only improves the safety of driving,but also improves traffic efficiency,reduces resource consumption and provides a good solution for congested roads.Among them,path planning and trajectory tracking is an important part to ensure that intelligent vehicle can run normally on the road,and is the basis for intelligent vehicle to plan non-collision safe paths and ensure accurate tracking,which has extremely important research significance.Firstly,the path planning of intelligent vehicle is studied in this thesis.As the intelligent vehicle is greatly affected by the constraints of the vehicle itself,the kinematics of the vehicle should be considered in the path planning and the kinematic model should be established.Then,in order to verify that the potential ant colony algorithm can plan a safe and feasible path regardless of the number of obstacles,a random grid environment model is established.In order to simplify calculation,the intelligent vehicle is regarded as a particle,and the obstacle is expanded considering the safety of vehicle driving.On this basis,a potential field ant colony algorithm is proposed for path planning.The algorithm combines the improved artificial potential field algorithm with ant colony algorithm,gives full play to the advantages of the two algorithms,and forms a local add global integrated path planning algorithm.In order to ensure the smoothness of the planned path and reduce redundant paths,cubic B-spline curve is added for optimization.Finally,the feasibility and effectiveness of the proposed algorithm are verified by experimental simulation.Secondly,the tracking control of intelligent vehicle based on model predictive control is studied.To ensure the tracking accuracy of the MPC controller,the three-degree-of-freedom vehicle dynamics model was established,the nonlinear dynamics model was transformed into a linear model,and the objective function and constraint conditions satisfying the driving conditions of the intelligent vehicle were set up.In order to simplify the calculation,the objective function was processed by quadratic programming,and the tracking effect of the controller was verified in the linear driving simulation condition.The simulation results show that the MPC controller is simple,practical,real-time,and can track the target trajectory quickly and stably.Finally,in order to verify the real-time performance of intelligent vehicle path planning and the accuracy of trajectory tracking,a co-simulation platform of Matlab and Carsim was built.The potential field ant colony algorithm was used to plan a safe and feasible path in the driving environment of intelligent vehicle.The path was taken as the expected path of intelligent vehicle tracking control.Then,the tracking control effect of MPC trajectory tracking controller was verified by taking the vehicle’s traveling speed as the variable.