With the rapid development of deep learning in recent years, the level of automatic driving perception has also increased substantially. However, automatic driving perception under adverse conditions, such as fog, remains a significant obstacle. The existing fog-oriented detection algorithms are unable to simultaneously address the detection accuracy and detection speed. Based on improved YOLOv5, this work provides a multiobject detection network for fog driving scenes. We construct a synthetic fog dataset by using the dataset of a virtual scene and the depth information of the image. Second, we present a detection network for driving in fog based on improved YOLOv5. The ResNeXt model, which has been modified by structural re-parameterization, serves as the model’s backbone. We build a new feature enhancement module (FEM) in response to the lack of features in fog scene images and use the attention mechanism to help the detection network pay more attention to the more useful features in the fog scenes. The test results show that the proposed fog multitarget detection network outperforms the original YOLOv5 in terms of detection accuracy and speed. The accuracy of the Real-world Task-driven Testing Set (RTTS) public dataset is 77.8%, and the detection speed is 31 frames/s, which is 14 frames faster as compared with the original YOLOv5.

In the condition of actual obstacle avoidance, where longitudinal deceleration is large and variable, it is difficult for the autonomous vehicle to simultaneously track longitudinal velocity and lateral motion. In addition, when the planned trajectory exceeds the adhesion of the road, the vehicle is also prone to losing stability. Most studies of the obstacle avoidance process only consider constant velocity. This paper presents a novel control architecture for four-wheel driving (4WD) and four-wheel steering (4WS) autonomous vehicles to track predefined trajectory and velocity. It is proposed to use preview technology to obtain the desired state of the vehicle and a controller based on multi-input multi-output nonlinear model predictive control (MIMO-NMPC) that considers the deceleration-steering combination condition. According to the sensitivity analysis of the influence of tyre slip rate on tyre lateral force, the optimal working boundary of the tyre is determined. The output of the controller is constrained by estimating tyre force to prevent tyre force saturation and maximise tyre adhesion limit utilisation. The simulation results demonstrate that the proposed controller can better track the trajectory and longitudinal velocity with a faster response on the high-adhesion road. The overshoot of lateral displacement and the maximum error of longitudinal velocity are less than 0.03m and 0.2m/s, respectively. In addition, when the planned trajectory exceeds the adhesion limit of the road, it can still follow the desired trajectory as closely as possible while maintaining stability on low-adhesion roads.

This paper is concerned with the problem of constraint control for an Antilock Braking System (ABS) with time‐varying asymmetric slip ratio constraints. A quarter vehicle braking model with system uncertainties and a Burckhardt’s tire model are considered. The Time‐varying Asymmetric Barrier Lyapunov Function (TABLF) is embedded into the controllers for handling the time‐varying asymmetric slip ratio constraint problems. Two adaptive nonlinear control methods (TABLF1 and TABLF2) based on TABLF are proposed not only to track the optimal slip ratio but also to guarantee no violation on the slip ratio constraints. Simulation results show that the proposed controllers can guarantee no violation on slip ratio constraints and avoid self‐locking. In the meantime, TABLF1 controller can achieve a faster convergence rate, shorter stopping time, and shorter distance, compared to TABLF2 controller with the same control parameters.

Controlling an autonomous vehicle's unprotected left turn at an intersection is a challenging task. Traditional rule‐based autonomous driving decision and control algorithms struggle to construct accurate and trustworthy mathematical models for such circumstances, owing to their considerable uncertainty and unpredictability. To overcome this problem, a rule‐constrained reinforcement learning (RCRL) control method is proposed in this work for autonomous driving. To train a reinforcement learning controller with rule constraints, outcomes of the path planning module are used as a goal condition in the reinforcement learning framework. Since they include vehicle dynamics, the proposed approach is safer and more reliable compared to end‐to‐end learning, thereby ensuring that the generated trajectories are locally optimal while adjusting to unpredictable situations. In the experiments, a highly randomized two‐way four‐lane intersection is established based on the CARLA simulator to verify the effectiveness of the proposed RCRL control method. Accordingly, the results show that the proposed method can provide real‐time safe planning and ensure high passing efficiency for autonomous vehicles in the unprotected left turn task.