Research On Multi-objective Coordination And Steering System Fault-tolerant Control Of Distributed Drive Unmanned Vehicle Based On State Estimation

Unmanned vehicle technology can effectively improve road utilization and traffic security,and reduce the operation cost of traffic management,in recent years,its related intelligent driving technology has become the research hotspot of automobile industry and related interdisciplinary,which has good development prospects and practical application value.The dynamic coordinated control of unmanned vehicle can ensure the stability and control mobility of vehicle while achieving the purpose of trajectory tracking control,and at the same time,the steering control ability of unmanned vehicle directly affects the trajectory tracking and dynamic coordinated control effect of unmanned vehicle.However,in the intelligent driving scenario,if the vehicle steering system fails,the reliability of unmanned vehicle coordinated control system will be greatly reduced,which will affect the trajectory tracking effect and even lead to instability.Therefore,it is necessary to design corresponding multi-objective coordination and fault-tolerant control method,so as to improve the fault-tolerant ability of unmanned vehicles in the face of sudden failure.In order to solve this problem,this paper proposes a multi-objective coordination and steering system fault-tolerant control method based on state estimation,which aims to ensure that the unmanned vehicle can maintain good trajectory tracking ability and coordination control effect in the face of sudden failure,so as to provide new research basis and practical experience for fault-tolerant control and reliability research of distributed drive unmanned vehicles.Firstly,the model of distributed drived unmanned vehicle is constructed,and the fault-tolerant control problem is analyzed.Combined with the electric-driving characteristics and intelligent driving scene of the distributed drive unmanned vehicle,the dynamic vehicle model,electric-driving-wheel model,tire model,trajectory tracking model and autonomous steering system model are built to characterize the trajectory tracking model relationship and its own dynamic control characteristics of the unmanned vehicle.On this basis,in order to further realize the fault-tolerant control of the unmanned vehicle,the model analysis and control problem description are carried out respectively considering the situation of incomplete failure of vehicle steering motor and complete failure of vehicle steering motor.Secondly,the running state estimation method of distributed drive unmanned vehicle considering estimation accuracy and reliability is designed.Combined with the characteristics of vehicle dynamics and the advantages of different estimation algorithms,a vehicle driving state estimation method based on weighted fusion of multi-model observers is designed,in which the weighted fusion method between observers in closed-loop estimation system is used to improve the overall estimation accuracy of vehicle state.In addition,a vehicle driving state estimation method based on error iteration and information fusion is designed,which makes full use of the relationship between different vehicle models and the fusion results of redundant state information,and the reliability and adaptive adjustment ability of the estimation system are further improved while ensuring the estimation accuracy.Thirdly,the multi-objective coordinated hierarchical control system of distributed drived unmanned vehicle is constructed.In order to realize the trajectory-tracking and yaw-stability multi-objective control of distributed drived unmanned vehicle,the coordinated control strategy of trajectory tracking and yaw stability is designed by using hierarchical control method.In the upper controller,a trajectory tracking controller based on model predictive control algorithm and a yaw stability controller based on sliding mode control algorithm are included to obtain the required front wheel angle and additional yaw moment respectively.In the lower controller,there are front wheel angle tracking controller and four-wheel tire force optimal distribution controller,which are respectively used to complete the front wheel angle demand and additional yaw moment demand instructed by the upper controller.Through the coordination of trajectory tracking control and yaw stability control,the yaw stability of unmanned vehicles can be ensured while tracking the reference trajectory,which lays a research foundation for the fault-tolerant control of unmanned vehicles.Fourthly,a fault-tolerant control method of distributed drive unmanned vehicle considering the incomplete failure of vehicle steering motor is proposed.In order to maintain the original control performance in the face of incomplete failure of steering motor,the corresponding fault-tolerant control method is designed.Through the description and mathematical modeling of steering motor fault,the representation of incomplete failure of steering motor and the problem of fault-tolerant control are further realized.On this basis,the fault-tolerant hierarchical control strategy of the whole vehicle is designed.Among them,Kalman filter is used to estimate the vehicle sideslip angle and steering motor fault,and a passive fault-tolerant controller is designed based on the state estimation results.Then a new optimal allocation method of tire force with feedback compensation mechanism is proposed in the lower controller,so as to realize the closed-loop control of distributed drive unmanned vehicle in the case of steering motor fault.Fifthly,the fault-tolerant control method of distributed drive unmanned vehicle considering the full fault mode of vehicle steering motor is further studied.In order to ensure that the unmanned vehicle can realize the trajectory tracking through the corresponding steering control when steering motor fails completely,the autonomous steering driving mechanism of unmanned vehicle is fully analyzed,and the feasibility of using differential steering control to replace the steering control when the steering motor fails completely is proposed,and the method of front-wheel angle estimation and tracking control in this case is designed,thus,the fault-tolerant control is realized under the condition of complete failure of steering motor.On this basis,an active fault-tolerant control strategy considering all fault modes of the steering motor is designed,and the corresponding fault-tolerant control mode is selected through the fault identification results of steering motor,so as to ensure that the unmanned vehicle can maintain good control performance under different steering motor faults.Finally,the hardware-in-the-loop simulation test and the real vehicle test are carried out to verify the actual effect of the proposed vehicle state estimation method,the multi-objective coordinated control strategy,and the fault-tolerant control method.The hardware-in-the-loop simulation platform and test vehicle are built and debugged.The chassis dynamometer bench test and real vehicle road test are used to verify the actual effect of proposed state estimation method,the hardware-in-the-loop platform simulation test and vehicle road test are used to verify the effect of the multi-objective coordinated control strategy in trajectory tracking and yaw stability control,and the hardware-in-the-loop platform simulation test is used to verify the actual effect of fault-tolerant control method under different fault modes of steering motor.The results show that: a)The vehicle driving state estimation method proposed in this paper can real-time observe the longitudinal tire force,lateral vehicle speed,yaw rate,vehicle sideslip angle and other driving states of distributed drive unmanned vehicle,the overall estimation accuracy and reliability are improved by 8.73% and11.69% respectively,and the estimation ability meets the actual control requirements;b)The proposed multi-objective coordinated control strategy of unmanned vehicle can meet the requirements of trajectory tracking control and yaw stability control,and in the road test,the trajectory tracking accuracy and yaw stability control accuracy under coordinated control are improved by 13.76% and 25.11% respectively compared with the conventional MPC algorithm;c)The fault-tolerant control method of distributed drive unmanned vehicle can improve the fault-tolerant ability of distributed drive unmanned vehicle and basically maintain the original trajectory tracking and yaw stability control effect under different steering motor failure conditions,and the accuracy of trajectory tracking and yaw stability control of vehicles after the fault is 93.07% and 91.66% respectively.The research work of this paper enriches the existing unmanned vehicle motion control technology,improves the reliability of unmanned vehicle system,and accumulates valuable research experience for the intelligent research of unmanned vehicles.