This paper presents a Takagi-Sugeno (T-S) fuzzy based robust H∞ integrated lane keeping (LKS) and direct yaw moment control (DYC) system to improve the automatic driving performance of unmanned electric vehicles. Firstly, a T-S fuzzy lateral model of unmanned electric vehicles is constructed, in which the cornering stiffness uncertainties, external disturbance, time varying look-ahead distance and longitudinal velocity are considered. Secondly, an integrated LKS and DYC control system of unmanned electric vehicles to handle with the time-varying features, parametric uncertainties and over-actuated actuators of unmanned electric vehicles, in which a T-S fuzzy based robust H∞ output feedback control law is designed to produce the desired front wheel steering angle and external yaw moment, and a dynamic quadratic programming (QP) distribution law is presented to allocate the desired external yaw moment of unmanned electric vehicles. Finally, the simulation and experimental results are implemented to verify the availability of the proposed control scheme.

This paper presents a systematic method on how to design the coordinated lateral and longitudinal motion control system of autonomous four wheel drive (4WD) electric vehicles for platooning and trajectory tracking. First, mathematical models that perfectly describe the behaviors of autonomous 4WD vehicles are built-up, and the coupled effects in vehicle dynamic systems are given. Second, owing to the fact that autonomous vehicles are large-scale systems with strong coupling, nonlinearities, and uncertainties, a novel multi-objective hierarchical architecture used for coordinated lateral and longitudinal motion control is constructed, which is composed of a global cooperative control layer, a control allocation layer, and an action execution layer. A robust backstepping sliding mode controller (RBSMC) is presented in the cooperative control layer to provide the resultant forces/moment. The control allocation layer is designed using interior-point (IP) algorithm to determine the tire lateral and longitudinal forces, which result in the desired resultant forces/moment. The action execution layer consists of an inverse tire model, a slip ratio regulator for each wheel, and a slip angle regulator. Finally, simulation experiments are carried out under adverse driving conditions, and the results show that the proposed control architecture not only possesses excellent tracking performance but also enhances the riding comfort, stability, and safety of autonomous 4WD electric vehicles.

Since autonomous four‐wheel independently drive electric vehicles have the characteristics of parameter uncertainties, non‐linearities and redundant actuators, trajectory tracking control for lane change of autonomous electric vehicles is regarded as a challenging task. A novel non‐linear trajectory tracking control strategy is designed for lane changing manoeuvre. First, a dynamic trajectory planning strategy is proposed to update the desired trajectory according to the real‐time information acquired through vehicle‐to‐vehicle communications. Second, a robust adaptive non‐linear fuzzy backstepping controller is presented to produce the generalised forces/moment of autonomous electric vehicles, and the stability of this proposed adaptive controller is proven by the Lyapunov theory. Then, the quadratic optimisation goal function of tire energy dissipated power is constructed, and the optimal control allocation method is proposed to produce the desired longitudinal and lateral tire forces of autonomous electric vehicles. Finally, simulation results manifest that the proposed adaptive control strategy has the distinguished tracking performance.

The automatic parking system based on vision is greatly affected by uneven lighting, which is difficult to make an accurate judgment on parking spaces in the case of complex image information. To solve this problem, this paper proposes a parking space visual detection and image processing method based on deep learning. Firstly, a 360-degree panoramic system was designed to photograph the vehicle environment. The image has been processed to obtain a panoramic aerial view, which was input as the original image of the parking space detection system. Secondly, the Faster R-CNN (Region-Convolutional Neural Network) parking detection model was established based on deep learning. It was aimed to detect and extract the parking space from the input image. Thirdly, the problems of uneven illumination and complex background were solved effectively by removing the background light from the image. Finally, a parking space extraction method based on the connected region has been designed, which further simplified the parking space extraction and image processing. The experiment results show that the mAP (mean Average Precision) value of the Faster R-CNN model using 101-Floor ResNet as the feature extraction network is 89.30%, which is 2.28% higher than that of the Faster R-CNN model using 50-Floor ResNet as the feature extraction network. The model built in this paper can detect most parking spaces well. The position of the output target box is accurate. In some test scenarios, the confidence of parking space recognition can even reach 100%. In summary, the proposed method can realize the effective identification and accurate positioning of parking spaces.