Search

Motion control of lane-change maneuvers under high-speed scenarios creates a great challenge for autonomous vehicles. However, a novel framework of lane-change trajectory planning and control based on the dynamic characteristics of distributed drive electric vehicles (DDEVs) is proposed in this paper to increase the possibility of safe driving in complex traffic conditions. First, a dilatational stability region determined by limits of stable yaw rate and saturated slip angles of each tire is employed to reveal dynamic instantaneous characteristics. On the basis of dilatational stability regions, the expanded envelopes by additional direct yaw moment can be identified by the expressions of irregular hexagons. Then, a series of unconstrained lane-change trajectory clusters can be generated to minimize longitudinal and lateral jerk, which is used to divide the global feasible region determined by the dilatational stability region and three-dimensional collision-free tunnel. Then, the technique for order preference for similarity to the ideal solution (TOPSIS) is utilized to achieve optimization of trajectory candidates from the global feasible region. The tracking controller aims at following the optimal lane-change trajectory, which is designed by the expanded stability boundaries of DDEV. Finally, simulations and experiments are conducted to validate the performance of the proposed planner and controller under high-speed, full-road and multi-vehicle traffic conditions.

With the popularity and increasing demand of new energy automotive, four-wheel independent drive electric vehicles (4WID-EVs) and its distributed control mode is promising to be a better chassis architecture for its flexibility and maneuverability. However, due to the uncertainty of the vehicle internal time-varying parameters and imprecise fault detection method, a novel robust fault estimator-based stochastic model predictive control (SMPC) is proposed in this article to ensure the vehicle lateral motion control and track the desired longitudinal velocity considering the estimation error of motor failure degree. Firstly, a robust H-infinity fault estimator with a linear parameter-varying (LPV) model is established to estimate the real-time fault signal considering the external disturbance and parameter perturbation (i.g. longitudinal and lateral vehicle speed). Then, a SMPC controller framework is developed to alleviate the effect of inevitable evaluated estimation error while the vehicle lateral safety and vehicle speed tracking can be ensured. Finally, the hardware-in-loop (HIL) tests are conducted to prove the feasibility of SMPC controller. The experiment results indicate that the proposed controller framework can effectively preserve vehicle stability and keep vehicle in a healthy state with faster response compared with the traditional fault-tolerant controller.

This paper studies the distributed dynamic event-triggered model predictive control (MPC) of vehicle platoon systems subject to denial-of-service (DoS) attacks and external disturbances. We develop a novel DoS-attack-aware event-triggered mechanism for each vehicle to determine whether the MPC problem should be solved or not at each sampling time instant. If the MPC problem is solved, the constructed vehicular beacon information will be transmitted to its neighbouring vehicles through the vehicular ad-hoc network (VANET). In the DoS-attack-aware event-triggered condition, the dynamic threshold adjusts adaptively according to DoS attack durations, local vehicular data, and beacon information from its neighbours. Then, an exponential-type robustness constraint is introduced in the MPC problem to deal with external disturbances. Subsequently, sufficient conditions are given to guarantee the recursive feasibility of the MPC algorithm and closed-loop platoon system stability. Finally, extensive simulation examples are provided to illustrate the effectiveness of the proposed algorithm in terms of communication efficiency and platoon resilience control performance.