Vehicle Cooperative Control Basd On Information Exchange And Motion Coupling

Vehicle cooperative control (VCC) system can achieve multi-vehicle cooperationand multi-objective coordination, satisfying the requirements of vehicle safety incomplicated traffic scenario, thus becoming an important development direction in thefield of vehicle motion control. To address the weakness of traditional autonomousvehicle in environment perception, target identification and control timing, this thesisproposed a VCC system based on information exchange and motion coupling control.Corresponding key technologies and fundamental issues are studied to extend theapplication field of cooperative control.By analyzing the information requirement of VCC system, information fusionbased vehicle motion state estimation is utilized to observe the necessary dynamicparameters for cooperative control, to make the provided information accurate, real timeand structured.Based on information exchange technology, the kinematic inter-vehiclerelationship in complicated traffic environment is analyzed, and approaching index (AI)is proposed to describe the adjacency between self-vehicle (SV) and neighbor vehicle(NV). Taking the time to trajectory cross point (TTCP) of NV, the time to headway(THW) of SV and the inverse of time to collision (TTCi) of SV as inputs, the fuzzymathematic model of AI is built, which improves the performance of multi-vehiclecooperative motion control.To shorten the latency of controller intervention on curved road, the model of VCCsystem considering car following safety and lane keeping performance is established, bytaking the advantage of information exchange. Through zero-dynamic design method, aVCC comprehensive control strategy with state feedback and feedforwardcompensation is developed. State feedback control keeps system asymptotically stableand outer performance optimal, while feedforward compensation control responses andcompensates outer disturbance rapidly.To take into account the influence to longitudinal tire force control accuracy fromthe side slip angle, a novel longitudinal slip ratio estimation method considering sideslip angle is proposed. The increment PID method is adopted to enhance the precision of longitudinal force control, lower the danger caused by accumulated longitudinal forcecontrol error, and decrease the overshoot of tire slip ratio control.Finally, five different experiment scenes are designed to compare the control effectof VCC controller, ACC controller/preview controller, driver by car dynamic simulatorand hardware/driver in-the-loop test. Results show that the VCC controller can detectpotential preceding target in advance, and hold the dual functions of crash avoidanceand car-following in multi-vehicle cooperation. On curved roads, the VCC controllercan coordinate longitudinal and lateral control objectives ensuring following safetywithout sacrifice lane keeping capability. The experiment results have verified theeffectiveness of the proposed VCC controller.