In order to drive safely in a dynamic environment, autonomous vehicles should be able to predict the future states of traffic participants nearby, especially surrounding vehicles, similar to the capability of predictive driving of human drivers. That is why researchers are devoted to the field of trajectory prediction and propose different methods. This paper is to provide a comprehensive and comparative review of trajectory-prediction methods proposed over the last two decades for autonomous driving. It starts with the problem formulation and algorithm classification. Then, the popular methods based on physics, classic machine learning, deep learning, and reinforcement learning are elaborately introduced and analyzed. Finally, this paper evaluates the performance of each kind of method and outlines potential research directions to guide readers.

On low-friction surfaces, vehicle stability is significantly influenced by nonlinear vehicle dynamics. To realize stability improvement, a real-time nonlinear model predictive control (NMPC) strategy that combines torque vectoring control (TVC) and rear-wheel steering (RWS) is proposed. A multitarget optimization objective function that considers system state tracking, stability, control constraints, and actuation energy consumption is established. Then, a fast-solving algorithm based on Pontryagin’s minimum principle (PMP) is proposed to reduce the heavy computational burden of NMPC and satisfy the real-time computation requirement of vehicular applications. The effectiveness of the proposed fast-solving algorithm is verified by comparing its closed-loop performance with interior point optimization (IPOPT). A series of hardware-in-the-loop (HIL) experiments with the real human driver is conducted to verify the effectiveness of the proposed fast-solving algorithm and combined control strategy. Furthermore, the HIL results show that the proposed combined control strategy can improve a vehicle’s handling stability on a low-friction surface and has a better performance and lower actuation energy consumption than the TVC-only method.

Active pedestrian collision avoidance (APCA) systems can significantly reduce road injuries and thus have attracted considerable attention from both the automobile and the transportation industries. Numerous studies have focused on APCA; however, it remains challenging to model variable and complex scenarios in a safe, efficient and low-cost way. For this purpose, this paper proposes a novel multipedestrian collision risk assessment framework comprising a motion prediction module, a collision checking module and a collision risk assessment module. First, the motion of the ego vehicle (EV) is predicted through a constant-acceleration (V-CA) model and a constant-turn-rate and constant-velocity (V-CTRV) model in different scenarios. Next, the pedestrian motion is estimated using Kalman filter (KF) approaches based on a constant-velocity (PCV) model and a constant-acceleration (P-CA) model. Then, the potential collision area (PCA) is defined according to the predicted motions of the EV and pedestrian, and the time-to collision (TTC) is selected to conduct collision checking. Finally, the most dangerous pedestrian is identified as the target with the minimum TTC in a multipedestrian scenario, and field tests are conducted on an autonomous vehicle platform. Comparative simulations indicate that the P-CA-based KF is more accurate and robust in variable-velocity scenarios than the P-CV-based KF. Furthermore, multipedestrian scenario simulations validate the effectiveness of the proposed framework.

Vehicle stability control under extreme conditions is influenced by the coupled nonlinear characteristics of vehicle dynamics, corresponding safety constraints, and rapid response requirements. To address these problems, this brief proposes a real-time nonlinear predictive controller for a distributed drive electric vehicle. First, nonlinear lateral dynamics of the vehicle are applied to develop the stability controller on low friction coefficient surfaces. Second, the requirement for suppressing the sideslip angle is integrated into the objective function to prevent the vehicle from destabilizing due to excessive sideslip angles. Finally, a fast solution algorithm is proposed by solving the transformed two-point boundary value problem, making it possible to apply nonlinear predictive controller to experimental road tests. The experiments with a production vehicle are conducted on the snow-covered dynamic roads of the DongFeng's winter test center. The test results on low friction coefficient roads show that the overall passing speed can be improved from 50-55 to 60-70 km/h with the proposed controller.