In order to enhance the trajectory tracking accuracy of distributed-driven intelligent vehicles, this paper formulates the tasks of torque output control for longitudinal dynamics and steering angle output control for lateral dynamics as Markov decision processes. To dissect the requirements of action output continuity for longitudinal and lateral control, this paper adopts the deep deterministic policy gradient algorithm (DDPG) for longitudinal velocity control and the deep Q-network algorithm (DQN) for lateral motion control. Multi-agent reinforcement learning methods are applied to the task of trajectory tracking in distributed-driven vehicle autonomous driving. By contrasting with two classical trajectory tracking control methods, the proposed approach in this paper is validated to exhibit superior trajectory tracking performance, ensuring that both longitudinal velocity deviation and lateral position deviation of the vehicle remain at lower levels. Compared with classical control methods, the maximum lateral position deviation is improved by up to 90.5% and the maximum longitudinal velocity deviation is improved by up to 97%. Furthermore, it demonstrates excellent generalization and high computational efficiency, and the running time can be reduced by up to 93.7%.

In the context of autonomous driving, sensing systems play a crucial role, and their accuracy and reliability can significantly impact the overall safety of autonomous vehicles. Despite this, fault diagnosis for sensing systems has not received widespread attention, and existing research has limitations. This paper focuses on the unique characteristics of autonomous driving sensing systems and proposes a fault diagnosis method that combines hardware redundancy and analytical redundancy. Firstly, to ensure the authenticity of the study, we define 12 common real-world faults and inject them into the nuScenes dataset, creating an extended dataset. Then, employing heterogeneous hardware redundancy, we fuse MMW radar, LiDAR, and camera data, projecting them into pixel space. We utilize the “ground truth” obtained from the MMW radar to detect faults on the LiDAR and camera data. Finally, we use multidimensional temporal entropy to assess the information complexity fluctuations of LiDAR and the camera during faults. Simultaneously, we construct a CNN-based time-series data multi-classification model to identify fault types. Through experiments, our proposed method achieves 95.33% accuracy in detecting faults and 82.89% accuracy in fault diagnosis on real vehicles. The average response times for fault detection and diagnosis are 0.87 s and 1.36 s, respectively. The results demonstrate that the proposed method can effectively detect and diagnose faults in sensing systems and respond rapidly, providing enhanced reliability for autonomous driving systems.

To reduce the low-frequency noise inside automobiles, a lightweight plate-type locally resonant acoustic metamaterial (LRAM) is proposed. The design method for the low-frequency bending wave bandgap of the LRAM panel was derived. Prototype LRAM panels were fabricated and tested, and the effectiveness of the bandgap design was verified by measuring the vibration transmission characteristics of the steel panels with the installed LRAM. Based on the bandgap design method, the influence of geometric and material parameters on the bandgap of the LRAM panel was investigated. The LRAM panel was installed on the inner side of the tailgate of a traditional SUV, which effectively reduced the low-frequency noise (around 34 Hz) during acceleration and constant-speed driving, improving the subjective perception of the low-frequency noise from “very unsatisfactory” to “basically satisfactory”. Furthermore, the noise reduction performance of the LRAM panel was compared with that of traditional damping panels. It was found that, with a similar installation area and lighter weight than the traditional damping panels, the LRAM panel still achieved significantly better low-frequency noise reduction, exhibiting the advantages of lightweight, superior low-frequency performance, designable bandgap and shape, and high environmental reliability, which suggests its great potential for low-frequency noise reduction in vehicles.

To improve the multidirectional motion control accuracy and driving stability of Distributed Drive Autonomous Vehicles (DDAVs) under extreme conditions, the extreme speed estimation method based on dynamic boundary and the multidirectional motion coupling control law design method based on multi degrees of freedom vehicle dynamic model are proposed. The extreme speed estimation method identifies the stable state of DDAVs by the dynamic boundary composed of yaw rate, sideslip angle and roll angle, and then estimates the extreme speed of the vehicle. The design method of multidirectional motion coupling control law adopts eight-degrees-of-freedom (8-DOF) vehicle dynamic model to design path tracking control law, speed tracking control law, yaw stability control law, and active suspension control law at the same time, so as to realize multidirectional motion coupling control. Based on the above method, the Multidirectional Motion Coupling Control System (MMCCS) of DDAVs is designed. The effectiveness of the proposed method is verified by double-line-shifting and serpentine driving simulations under different road adhesion conditions. The superiority of the method is proved by comparing the existing integrated control method.

In this study, a state feedback robust control strategy with time delay dependence for the uncertainty, fault, and input time delay of a vehicle’s controllable suspension system is designed. Firstly, based on the working principle of the adjustable damping suspension system, the total damping force of the suspension is divided into “foundation damping force” and “controllable damping force”, taking the adjustable damping interval as the constraint boundary, which varies with the actuating velocity. Furthermore, based on the sensitivity analysis of the input time delay’s relationship with damping force disturbance, it exerts influence on the controllable damping force part. The associated uncertainty and fault are defined through linear fraction transformation and proportional gain, respectively, and the two are decoupled from the control mechanism. Then, based on the Lyapunov stability theory, a robust control law for the time-delay-dependent H∞ state feedback is designed and transformed into a linear matrix inequality convex optimization problem to solve. Finally, the feasibility and effectiveness of the proposed method are verified by designing different combinations of uncertainty and fault tests and by comparing and analyzing these with the time-delay-independent H2/H∞ robust control strategy under random and pulse road excitation.

For the robust fault-tolerant control of the controllable suspension system, a control strategy driven by knowledge-data fusion is proposed. Firstly, the boundary fuzziness between perturbation type uncertainty and gain type fault is analyzed, and then a data-driven method is introduced to avoid the state estimation of system uncertainty and fault. The proximal policy optimization algorithm in reinforcement learning is selected to construct a “data control law”, to deal with uncertainty and fault. On the other hand, based on the classical sky-hook control, the “knowledge control law” for system performance optimization is designed, taking into account the nonlinear and non-stationary characteristics of the system. Furthermore, the dependency between robust fault tolerance and performance optimization control is revealed, and the two control laws are fused by numerical multiplication, to realize the performance matching optimization control of robust fault tolerance of controllable suspension system driven by knowledge-data fusion. Finally, the effectiveness and feasibility of the proposed method are verified by the simulation and real-time experiment of non-stationary excitation and near-stationary excitation under the combination of uncertainty and fault.