Research On Multi-objective Optimal Control Of Intelligent Driving Vehicles On Highway

With the development of technology and the continuous improvement of living standards,people have higher and higher requirements for the safety,stability,economy and other performance of vehicles.Intelligent driving vehicles apply intelligent technology to electric vehicles,which can sense the surrounding environment through electronic devices such as sensors,and control the motion of the vehicle through intelligent control strategies to liberate drivers from complex and heavy driving behavior.Intelligent driving vehicle has significant advantages in enhancing the safety and stability of the vehicle driving process and improving economic performance,so it has become an important direction of automobile development in the future.Once the vehicle becomes unstable at high speed,it will cause very serious casualties and property losses.Therefore,under the condition of high-speed driving,how to control the intelligent vehicle to track the desired path and how to carry out multi-objective motion control,so that the vehicle can maintain safety and stability while taking into account the economic performance at the same time,has become the research direction of many scholars at home and abroad,and it is also the main research content of this paper.In the aspect of dynamic modeling,the vehicle dynamic model with 14 degrees of freedom is established,and the body model,tire model,suspension model and motor model are established respectively by using the idea of modularization.In order to further obtain the functional relationship between motor speed,torque and motor efficiency,the regression equation describing motor efficiency is obtained by response surface analysis.Finally,the validity of the vehicle model is verified based on the real vehicle test.In the aspect of tracking the desired path,a model predictive control(Model Predictive Control,MPC)strategy based on Kalman filter(Kalman Filter,KF)is designed.First of all,a three-degree-of-freedom vehicle dynamics model considering the lateral,longitudinal and yaw motion of the vehicle is established as the prediction model,and the prediction model is linearized and discretized,and then a Kalman filter is designed to reduce the noise of vehicle state information.Furthermore,the model predictive controller is designed,by setting the objective function and constraint conditions,the path tracking problem is transformed into a constrained quadratic programming problem,and the positive set algorithm is used to solve the steering wheel angle to realize lateral control.And continuous rolling optimization is carried out in the control process.Finally,the 14-degree-of-freedom vehicle model is used as the control object to design the simulation experiment,and the Logitech G29 driving simulation platform is used to collect the real driver driving data.The control effect of the skilled driver is compared with that of the KF-MPC path tracking control strategy proposed in this paper to verify the effectiveness of the control strategy proposed in this paper.In the aspect of multi-objective optimization control,a hierarchical control strategy is designed.The upper control strategy adopts sliding mode control to ensure the safety and stability of the vehicle.In order to track the expected vehicle speed,the expected yaw rate and the desired side slip angle,the sliding mode function is set as the tracking error of longitudinal speed,yaw rate and side slip angle.The expected longitudinal resultant force,lateral resultant force and yaw moment which affect the vehicle stability are obtained through calculation,and the saturation function is used instead of the symbol function to reduce chattering.The lower torque distribution control strategy adopts the optimal control allocation method,which tracks the expected vehicle state obtained by the upper control strategy while taking into account the economic performance of the motor.Finally,based on the error of the expected state and the system efficiency,the objective function is established,and the extreme value of the objective function is solved under the constraint condition.The optimal torque of each wheel is obtained,and the complete control process is realized by inputting the optimal torque into the vehicle model.In order to verify the effectiveness of the multiobjective optimization control strategy,a rule-based braking torque allocation method is established as a comparison,and simulation experiments are carried out under high adhesion road conditions and low adhesion road conditions respectively.In the real-time simulation experiment,the d SPACE real-time simulation platform is built based on the vehicle and control strategy model.The control strategy code is generated and applied to the real-time control.The vehicle state driven by the real driver is compared with the vehicle state under the control strategy proposed in this paper.The results show that the multi-objective optimization control strategy proposed in this paper can control the vehicle according to the driving situation,improve the system efficiency on the premise of keeping the vehicle running safely and stably,complete the desired path tracking,and meet the needs of multi-objective optimization.