Research On Emergency Collision Avoidance Strategy Of Distributed Drive Electric Vehicle Based On Vehicle-in-the-loop Test

The non-standard operation of drivers is the main cause of traffic accidents.As an important part of ADAS(advanced driving assistance system),the emergency collision avoidance technology used to avoid risks has become a research hotspot.In the process of emergency collision avoidance,it is necessary to supplement vehicle stability control to avoid vehicle instability.Distributed drive electric vehicle has a simple and reliable structure,and the torque of each wheel is independently controllable.It can adjust the body posture by distributing the torque of each wheel,so as to restrain the vehicle instability.So it is the natural carrier of intelligent driving function of vehicles in the future.Therefore,this paper selects distributed drive electric vehicle as the research object and emergency collision avoidance strategy as the research content.The real vehicle test of emergency collision avoidance function has many difficulties,high risks and high requirements for the test site.As a result,it is urgent to develop a test system that breaks through the limitations of real vehicle test.Based on the investigation of the research status of emergency collision avoidance,distributed drive electric vehicle stability control and vehicle in the loop test,this paper designs the emergency collision avoidance strategy of distributed drive electric vehicle and the vehicle in the loop test system for algorithm verification.According to the collision time,the threshold value of forward collision warning coefficient and the limit steering / braking distance,the collision risk is judged,the hierarchical emergency collision avoidance decision-making logical structure is built,the collision avoidance path is generated by the quintic polynomial method,and the path cluster is offset along the Y-axis of the vehicle coordinate system under the constraint conditions such as the maximum lateral acceleration of the vehicle and the road boundary,Then,the cost function with weight is established to evaluate the safety,efficiency and ride comfort of different paths,and the path with the lowest generation value is selected as the optimal collision avoidance path.The model predictive control method is used to build the path tracking controller.Based on the five degree of freedom vehicle model and linear tire model,the prediction model is derived,and the quadratic optimization objective function aiming at the path lateral deviation and yaw angle error is constructed.The steering wheel / wheel angle is obtained by rolling under the constraints of control quantity,control increment and output.Then the yaw stability fuzzy controller is designed with the sideslip angle of mass center and yaw rate as the control objectives,the expected additional yaw moment is calculated,and the yaw stability is controlled by distributing the driving torque of left and right rear wheels.The whole vehicle in the loop test system is built based on the NI-PXI framework.The NI-PXI cabinet runs the real-time model and Car Sim-RT vehicle dynamics model,and serves as a transfer station for software and hardware data interaction.After transformation,the real vehicle runs on the drum dynamometer,realizes signal command sending and receiving through CAN communication,and realizes upper computer management by using NI Veritand software.The single obstacle condition and multi obstacle condition are designed.The algorithm is verified by the joint simulation of MATLAB / Simulink,Car Sim and the real vehicle test of the whole vehicle in the loop test system.The joint simulation results show that the emergency collision avoidance strategy is reasonable and effective,the emergency collision avoidance enable can be triggered normally,track the expected collision avoidance path accurately,and avoid obstacles smoothly on the premise of ensuring stability.The consistency between the vehicle in the loop test results and the joint simulation results is high,which proves the practical utility of the algorithm and the real-time and reliability of the vehicle verification function of the vehicle in the loop test platform.