A novel vehicle dual-motor steer-by-wire (SBW) system is proposed to improve the reliability and safety of traditional SBW system. However, the biggest challenge of dual-motor SBW is its tracking and synchronization control issues. In order to solve the problems of poor tracking and synchronization of the vehicle dual-motor SBW system caused by load disturbance, parameter perturbation, and model mismatch, a tracking and synchronization control strategy of the vehicle dual-motor SBW system via the active disturbance rejection control is proposed. First, the driving mechanism of dual-motor SBW system is analyzed, and the system dynamics model is established. Then, considering the system disturbance and parameter uncertainty, a second-order active disturbance rejection controller is designed for the position loop of the steering motor, which is composed of a third-order extended state observer and a state error feedback control law. Next, in order to simplify the adjustment difficulty of the controller, the bandwidth method is introduced to greatly reduce the number of controller parameters and the difficulty of parameter adjustment. Based on this, the disturbance rejection performance and stability of the designed active disturbance rejection controller are proved. Furthermore, the cross-coupling control structure is employed to enhance the synchronization performance of the dual-motor system. Finally, the performance of the controller to suppress the external disturbance and internal parameter perturbation is analyzed in step condition and double lane change condition, respectively, and the effectiveness of the proposed control strategy is verified on the dual-motor experimental platform.

The existing vehicle obstacle avoidance path planning methods generally aim at obtaining the collision-free path, ignoring the impact of the planned path on the vehicle stability in the obstacle avoidance process, so that the controlled vehicle has the risk of rollover in the obstacle avoidance process. To solve the above problems, a two-layer obstacle avoidance path planning algorithm considering path pre-planning and re-planning is proposed in this paper. In the path pre-planning layer, an improved APF algorithm with road boundary function constraints is proposed. By introducing the repulsion field adjustment factor, the shortcomings of GNRON and local optimization existing in the existing artificial potential field method are effectively solved. In the path re-planning layer, taking the rollover stability index as the constraint, a pre-planning result optimization method based on particle swarm optimization algorithm is proposed. The simulation results show that the obstacle avoidance path planning algorithm proposed in this paper can not only generate the obstacle avoidance path in real-time, but also reduce the yaw rate and yaw angle of the main vehicle in the process of obstacle avoidance, and effectively improve the rollover stability of the vehicle in the process of obstacle avoidance.

To improve the agility and efficiency of the highway decision-making system and overcome the local optimal dilemma of the existing safety field, this paper builds an improved safety field to reflect the advantage of the reachable states and the learning process is further employed to make the decision long-term optimal. Firstly, the improved safety field is prepared by the kinematic model-based prediction of surrounding vehicles and the boundary is determined elaborately to ensure real-time performance. Then, the field is constructed by three individual fields. One is the kinematic field, which is built based the safe-distance model to measure the colliding risk of both moving or no-moving objects accurately. Another is the road field that reflects the lane-marker constraint. The last is the efficiency field, which is introduced creatively to improve efficiency. Furthermore, the learning algorithm is adopted to learn the long-term optimal state-action sequence in the safety field. Finally, the simulations are conducted in Prescan platform to validate the feasibility of the improved safety field in complex scenarios. The results show that the proposed decision algorithm can always drive autonomous vehicle to the state with a long-term optimal payoff and can improve the overall performance compared to the existing pure safety field and the interaction-aware method.

In order to improve the fault-tolerant performance of wheel angle sensor of steer-by-wire (SBW) vehicle, a fault diagnosis and fault-tolerant compensation (FDFTC) strategy for wheel angle sensor of SBW vehicle via extended Kalman filter (EKF) is proposed in this paper. Firstly, the steering motor model, rack and pinion model, vehicle dynamics model and variable transmission ratio model are established. Then, based on the analytical redundancy method, the EKF algorithm is adopted to obtain the wheel angle estimation signal. On this basis, a sensor FDFTC strategy with improved majority voting scheme is proposed, which is composed of fault diagnosis module and fault isolation compensation module. Finally, combined with Matlab/Simulink and Carsim, the estimation results are analyzed under sinusoidal condition and double lane change condition, and the performance of the proposed FDFTC strategy is further verified under the sensor stuck fault condition, deviation fault condition and noise interference condition. The results indicate that the proposed FDFTC strategy can not only accurately diagnose the fault signal, but also reconstruct the correct wheel angle signal, which is of great significance to improve the safety and reliability of SBW vehicle.

In order to improve the accuracy of fault detection results, this paper proposes a novel fault detection strategy of vehicle wheel angle signal via long short-term memory network (LSTM) and improved sequential probability ratio test (SPRT). Firstly, a signal estimation method based on data-driven modeling is presented, which fuses the vehicle current status information and adopts the LSTM based on deep learning to estimate the vehicle wheel angle signal. Then, the signal residual sequence is obtained by comparing the estimated wheel angle signal with the measured wheel angle signal. Based on this, the improved SPRT method based on mathematical statistics is used to analyze the signal residual sequence, so as to detect the fault signal timely and accurately. Finally, the accuracy of the estimation results is analyzed under sinusoidal condition, double-lane change condition and sinusoidal sweep frequency condition, and the effectiveness of the fault detection strategy proposed in this paper is further verified under the stuck fault condition and drift fault condition. The results indicate the effectiveness of the proposed fault detection strategy, which is of great significance to improve the safety and reliability of the vehicle.

Autonomous Vehicles (AVs) research has attracted much attention in recent years. In fact, some functionality of AV, such as cruise control, lane keeping and automatic parking have already been employed in commercial vehicles. In a fully AV system, the vehicle drives like a human being and might even outperform humans in some aspects, e.g., shorter reaction time when facing dangerous situations. To have AVs drive like human beings, discovering how to transfer human driving skills to smart vehicles has become a key issue. In this study, we have developed a human-to-vehicle skills transfer system that is able to control the smart vehicle to drive safely in the longitudinal direction under a virtual environment constructed within the Carla simulator. A deep Q-learning algorithm has been employed to build the learning network, which will obtain its control policy during the interaction between the agent and the environment. To achieve human-to-vehicle skills transfer, human data collected with the simulator is employed as the learning target for the learning network. Various experimental results show that longitudinal human driving skills can successfully transfer to a smart vehicle through the developed learning network.