Research On Trajectory Planning And Control Method Of Emergency Steering And Collision Avoidance For Smart Cars Under Urban Conditions

The number of motor vehicles in our country is growing rapidly and the traffic is busy.Various urban traffic problems need to be solved urgently.Carrying out research on active safety system related technologies is of great significance for improving the safety situation of urban traffic.At this stage,researchers usually develop autonomous emergency braking systems around longitudinal braking.When facing driving conditions with high relative speeds,the distance between the vehicle and the interfering vehicles and pedestrians is less than the longitudinal braking distance required by the emergency braking system.In view of the similar situation of insufficient longitudinal braking distance,this paper proposes a smart vehicle emergency steering collision avoidance strategy suitable for urban working conditions,and conducts research on emergency steering collision avoidance trajectory planning and control methods.The smart car realizes active collision avoidance based on the emergency steering operation mode.The main process includes: the system obtains road environment information and driving status information in real time according to the multi-sensor fusion module.When the operating conditions meet the emergency steering collision avoidance function triggering conditions,the controller plans to comply with The emergency steering collision avoidance trajectory required by safety and adaptability,the tracking controller sends the tracking control instruction of the collision avoidance trajectory,so that the smart car reaches the target position to complete the autonomous emergency collision avoidance.In order to realize the emergency steering and collision avoidance of smart cars,the research contents completed in this paper include:(1)The establishment of steering safety distance model and the design of autonomous emergency steering collision avoidance decision.Firstly,the urban road conditions are determined as research scenarios;Secondly,the brake collision avoidance limit distance,steering collision avoidance limit distance,and safety distance for collision prevention under the interference of vehicles in adjacent lanes are analyzed;finally,based on the brake collision avoidance limit distance,Steering collision avoidance limit distance,collision avoidance safety distance under the interference of vehicles in adjacent lanes,and driver's subjective intention,design a complete trigger mechanism for autonomous emergency steering collision avoidancefunction.(2)Two-stage emergency steering collision avoidance trajectory planning for smart cars.Aiming at the problem of emergency steering collision avoidance trajectory planning for smart cars,in order to improve the safety and adaptability of intelligent vehicle emergency steering collision avoidance trajectory planning,the emergency steering collision avoidance process is divided into an initial stage and a planning stage.Firstly,an excellent driver steering avoidance experiment was carried out;secondly,based on K-means clustering analysis method,an excellent driver's emergency steering avoidance behavior characteristic analysis was completed,and the steering wheel angle following curves at different speed intervals were obtained and used as an emergency steering Direct input in the initial stage of collision avoidance;Finally,the global horizontal coordinate of the center point of the feasible region of the vehicle detected by the smart camera is used as the expected lateral distance in the planning stage,and the trajectory planning in the emergency lane change planning stage is completed based on the Sigmoid function,which improves the adaptability of the trajectory planning algorithm.(3)The establishment of a four-degree-of-freedom vehicle dynamics model.Based on the desired trajectory of vehicle steering and collision avoidance and the state parameters of the own vehicle,the front wheel deflection angle required for vehicle lateral control is calculated.Considering the lateral,longitudinal,yaw and roll motion of the vehicle,the above-mentioned motion model is verified by establishing a four-degree-of-freedom vehicle dynamics model.(4)Design of trajectory tracker based on MPC.In order to drive the smart car to the target position according to the planned path of steering and collision avoidance,and realize the tracking control of the emergency steering and collision avoidance trajectory,an MPC trajectory tracker controller is designed.First,a four-degree-of-freedom vehicle dynamics model was established and validated,and used as a predictive model for model predictive control;then,three major links related to the trajectory tracker were designed,including: linear time-varying prediction model,rolling optimization,and feedback Effectiveness.(5)Simulation and real vehicle test verification.Car related parameters and various simulation scenarios are set in Carsim simulation software,MPC control model is built in Simulink,joint simulation is carried out using Carsim/Simulink,a real vehicle verification platform is built and real vehicle verification is completed in the actual road environment.The simulation and real vehicle test results show that the test vehicle successfully avoids obstacles and the tracking error is within a reasonable deviation range.The indexes such as the center of mass deflection angle,body roll angle,and lateral acceleration are all within the scope of safety constraints.The results show that the smart car emergency steering collision avoidance function decision,emergency steering collision avoidance trajectory planning and trajectory tracking control method proposed are feasible and effective.