When distributed drive electric vehicles (DDEVs) encounter challenges such as congested parking, abrupt dead ends, and narrow roads, in situ steering provides an excellent solution. However, the operational mechanisms and control strategies of in situ steering are currently a gap in distributed drive technology. To achieve stable and precise control of in situ steering, this article proposes an adaptive control strategy for synchronizing yaw rate and wheel speed, based on the analysis of the kinematics and dynamics of in situ steering. The control strategy is structured in layers, comprising an observation layer, a decision-making layer, and upper layer control. Initially, the observation layer calculates the road adhesion coefficient and slip ratio, using a dynamics-based method to assess road conditions at the onset of in situ steering. Subsequently, the decision-making layer determines the nominal yaw rate and the desired wheel speed. Then, the upper layer control designs an active disturbance rejection control (ADRC) algorithm, based on the tire force characteristics of in situ steering, along with a multisegment linear dynamic learning rate adaptive control algorithm, to precisely control the yaw rate and wheel speed while minimizing steering center deviation. In addition, a collaborative control submodule coordinates torque and braking pressure based on the designed state determination criteria, enhancing the system’s robustness and safety. Ultimately, simulations and real-vehicle tests validate the effectiveness of the proposed in situ steering mechanisms and control strategy design, confirming that the designed control strategy achieves satisfactory performance.
16 January 2024 in Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering
Based on the dynamic coupling analysis of the front wheel shimmy and vehicle plane motion, a three degrees of freedom (DOF) nonlinear shimmy model considering driver steering input is established. Firstly, when the driver steering input is not considered, the modal properties and dynamic stability of the system are investigated by solving the Jacobian matrix. Then, the nonlinear shimmy behavior considering the driver steering input is discussed by means of numerical calculations. The slow flow equations of the vehicle system considering the driver steering input are derived based on the complexification-averaging (CA) method. According to the nonlinear dynamics theory, the saddle node (SN) and Hopf bifurcation characteristics of the system are analyzed, the analytical solution of the front wheel shimmy angle is also obtained. Finally, with the help of the understeer gradient, the influence of the linear cornering stiffness of the front wheel on the nonlinear shimmy motion is discussed. It is summarized that the SN and Hopf bifurcation are more likely to occur in the shimmy system for an oversteered vehicle. The relevant conclusions are useful for the early anti-shimmy design of vehicles.
The four-wheel independent drive and four-wheel independent steering (4WID-4WIS) configuration enhances distributed control in intelligent electric vehicles. As a typical over-actuated system, its application in distributed control can provide various fault-tolerant control solutions to ensure vehicular safety. This paper proposes a hierarchical tire force allocation strategy for the 4WID-4WIS configuration based on constraint optimization. Additionally, it introduces an adaptive fault-tolerant control specifically for steering system failures. When steering failures arise, the fault-tolerant control system discerns the fault and sustains trajectory tracking performance using fuzzy control and control reallocations. Simulation results demonstrate that our strategy maintains trajectory tracking control performance, even under complete steering failures. In contrast, the vehicle loses control without fault-tolerant control, resulting in significant trajectory tracking errors in identical fault scenarios and raising the risk of serious traffic accidents.
Electromechanical braking (EMB) is an important braking system. It is necessary to maintain the effect of anti-lock-braking (ABS) on a low adhesion road surface. After multiple clamping releases, the motor parameters degenerate, resulting in EMB braking decline. Therefore, a PI adaptive control (PIAC) method based on threshold trigger parameter estimation (TTPE) is proposed for a permanent magnet synchronous motor (PMSM) in the EMB system. First, the dynamic model of PMSM is established, and the inertia of PMSM is identified offline by the acceleration and deceleration method (ADM). Then, a parameter recognition method based on threshold trigger (TT) is constructed, which mainly includes neural network recognition and neural network identification (NNI). Finally, based on the identified parameters, the PIAC is dynamically adjusted to realize the effective adjustment of EMB. Simulation and experimental results show that this method can effectively improve the robustness of motor control and solve the problem of deviation of control effect after EMB braking for a long time.