Research On Intelligent Vehicle Local Path Planning Method Considering Vehicle Dynamics

Intelligent vehicles have gradually become the focus technology of industry and academia.As an important part of the intelligent vehicle system,the local path planning system evaluates the current local surrounding environment of the vehicle,considering factors such as driving safety,driving efficiency and riding comfort.In the end,the local path planning system will plan the host vehicle’s feasible driving path and speed based on the global planning path information.Due to the complexity and variability of vehicle driving scenarios,the local path planning system of intelligent vehicles needs to complete the planning task in a short time,which requires the calculation speed of the local path planning module to be fast enough.In addition,for intelligent vehicles,the planned path needs to meet the constraints of the vehicle system,so vehicle dynamics need to be considered in the planning process.At the same time,the trade-off between the performance indicators of local path planning is also very important.Model predictive control algorithm is one of the most common numerical optimization methods for local path planning of intelligent vehicles.The optimal path can be planned according to the evaluation indexes under the conditions of constraints and vehicle dynamic model constraints.This paper uses model predictive control algorithms to realize the local path planning based on vehicle dynamics of intelligent vehicles.The main research contents of this paper are as follows:1.According to the different coordinate systems used in the local path planning method,the coordinate systems are introduced and analyzed.The article contains the mathematical derivations for the different coordinate systems conversion and vehicle dynamic model.Knowing the operating intention of the surrounding vehicles over a period of time,the operating intention of surrounding vehicles and the vehicle dynamic model are combined to form the trajectory prediction module.The predicted trajectory is used by the behavior planning system and the motion planning system to avoid obstacles.2.The designed local path planning module includes a behavior planning module and a motion planning module.Through the behavior planning module,intelligent vehicles can achieve basic obstacle avoidance and lane keeping operations while driving on structured roads,which provides a reference path for subsequent motion planning.After behavior planning,a motion planning module based on model predictive control is used.The motion planning module put vehicle dynamics characteristics,the future driving trajectory of the surrounding vehicle,safety,efficiency,and comfort into consideration.C/GMRES method is used for fast nonlinear model predictive control problem solving.3.Simulation is done based on MATLAB / Simulink software.A variety of simulation experiments are finished including obstacle avoidance on straight roads,curved roads,and obstacle lane-changing problem to verify the reasonableness of the designed local path planning system in different driving scenarios.