Gang Wang

Under complex and uneven road conditions, variations in vehicle load can readily induce stochastic disturbances in key parameters such as the center of mass and moments of inertia. These numerous and unpredictable fluctuations introduce stochastic structural uncertainties into the full-vehicle dynamic model, potentially destabilizing conventional suspension control strategies and degrading their overall performance. To address these challenges, this paper investigates a data-driven reinforcement learning-based H ₂ / H _∞ vibration control strategy, aiming to achieve robust suspension control in the presence of stochastic structural uncertainties. Firstly, a control-oriented model of the full-vehicle active suspension system is constructed. The control problem is reformulated as a stochastic zero-sum game problem, and an iterative algorithm is given to solve the associated stochastic game algebraic Riccati equation. To alleviate the complexity and high cost of parameter tuning and model identification, a data-driven off-policy reinforcement learning algorithm is introduced. This approach effectively compensates for the adverse effects introduced by stochastic structural uncertainties and provides a robust control solution without requiring explicit knowledge of system parameters. Finally, numerical simulations are conducted to validate the robustness of the proposed approach. The results show that the proposed algorithm can converge to the optimal solution within 10 steps.

The integration of high-precision vehicle-mounted visual sensors has revolutionized the field of intelligent suspension systems, enabling the implementation of preview control strategies that significantly enhance ride comfort and safety performance in vehicle platoons. Building on this advancement, this study presents a data-driven preview control framework for active suspension systems tailored to heterogeneous vehicle platoons. The proposed framework leverages a vehicle-to-vehicle (V2V) communication topology to incorporate comprehensive road preview information. Initially, a dynamic model of the active suspension system is formulated, encompassing inter-vehicle-distance preview, wheelbase preview, and look-ahead preview mechanisms. Subsequently, the preview H _∞ control problem is recast as a multi-player zero-sum game to address the interactions within the platoon. To reduce the calibration and design costs associated with the control scheme, a model-free Q-learning algorithm is presented. This approach determines the optimal control policy for the system without requiring prior knowledge of the output matrices, while simultaneously ensuring operational safety and ride comfort. Finally, numerical simulations and comparative analyses are conducted to validate the proposed control scheme. The results indicate that, under various road excitation scenarios, the proposed platoon preview control framework consistently outperforms traditional passive suspension systems and non-preview-based independent control schemes. Furthermore, the performance exhibits a strong correlation with the amount of available preview information. Notably, the scheme achieves optimal control performance in constant-speed platooning scenarios.

This paper addresses the constrained H∞ optimal control problem for nonlinear active vehicle suspension systems, with a focus on deriving an approximate solution through data-driven reinforcement learning in the context of differential games. A dynamic model of the half-car active suspension system with constraints is first established, where the constrained control forces and external road disturbances are formulated as a zero-sum game between two players. This leads to the Hamilton–Jacobi–Isaacs (HJI) equation, with a Nash equilibrium as the desired solution. To efficiently solve the HJI equation and mitigate the impact of model parameter uncertainties, an actor-critic neural network framework is employed to approximate both the control policy and the value function of the system. A reinforcement learning algorithm based on the input-output data of the suspension system is subsequently derived. Numerical examples are provided to demonstrate the effectiveness of the proposed approach for time-invariant suspension systems. Under varying control force constraints, the active suspension system consistently exhibits excellent vibration reduction performance.

To decrease the complexity and increase the efficiency of wireless power transfer (WPT) systems, this paper proposes a novel self-excited invert rectification method for the design of the invert rectifier of the receiver (Rx). The self-excited invert rectifier can perform the self-driving and soft-switching of the MOSFETs as well as the frequency-tracking function without a microcontroller. This allows us to greatly simplify the structure of the invert rectifier and increase the transfer efficiency (TE) of the WPT system. Firstly, a self-excited invert rectifier circuit is designed, and a self-excited invert rectification method is studied. Additionally, the power loss of the self-excited invert rectifier is analyzed. Finally, the self-excited invert rectifier of the WPT experimental system is designed. The self-excited invert rectification method is then verified. The key component parameters of the self-excited invert rectifier are provided and optimized. The TE of the WPT system that includes the self-excited invert rectifier is improved by more than 5% without a microcontroller. The self-excited invert rectifier of the Rx provides a practical solution for decreasing the complexity and increasing the TE of the WPT system.

In wireless power transfers (WPTs), it is challenging to obtain a constant output of power (COP) and constant transmission efficiency (CTE) in large coupling variation ranges. In this study, the eight-coil WPT system achieves a uniform magnetic field (UMF) in the transmitter and receiver sides using two transmitting (Tx) coils and two receiving (Rx) coils, respectively. COP and CTE are then achieved with large coupling variation ranges. The circuit model and equations of the transmission characteristics are first obtained based on the structure and working principle of the Helmholtz coil. The model of the mutual inductance and equation of the impedance coupled factor are then developed. The laws of the transmission characteristic are also determined by adopting a simulation tool and equations of the transmission characteristics. Finally, the eight-coil WPT experimental system is designed. In a fixed-frequency mode, the COP and CTE are achieved when the coupling and misalignment distances are changed within a quarter or one-fifth of the relay coil diameter, respectively. This topology provides an efficient solution for problems faced in practical applications, such as wireless chargers of kitchen appliances and automatic mobile robots of small size.

This paper investigates a parameter-free H_∞ differential game approach for nonlinear active vehicle suspensions. The study accounts for the geometric nonlinearity of the half-car active suspension and the cubic nonlinearity of the damping elements. The nonlinear H_∞ control problem is reformulated as a zero-sum game between two players, leading to the establishment of the Hamilton–Jacobi–Isaacs (HJI) equation with a Nash equilibrium solution. To minimize reliance on model parameters during the solution process, an actor–critic framework employing neural networks is utilized to approximate the control policy and value function. An off-policy reinforcement learning method is implemented to iteratively solve the HJI equation. In this approach, the disturbance policy is derived directly from the value function, requiring only a limited amount of driving data to approximate the HJI equation’s solution. The primary innovation of this method lies in its capacity to effectively address system nonlinearities without the need for model parameters, making it particularly advantageous for practical engineering applications. Numerical simulations confirm the method’s effectiveness and applicable range. The off-policy reinforcement learning approach ensures the safety of the design process. For low-frequency road disturbances, the designed H_∞ control policy enhances both ride comfort and stability.

Electromagnetic actuators, characterized by their lack of pneumatic or hydraulic circuits, rapid response, and ease of control, have the potential to significantly enhance the dynamic performance of automotive active suspensions. However, the complexity associated with their models and the calibration of control parameters hampers the efficiency of control design. To address this issue, this paper proposes a reinforcement learning vibration control strategy for electromagnetic active suspension. Firstly, a half-car dynamic model with electromagnetic active suspension is established. Considering the unknown dynamics of the actuator and its preset convergence performance, an optimal control method based on reinforcement learning is investigated. Secondly, a heuristic PI adaptive dynamic programming algorithm is presented. This method can update to the optimal control solution without requiring model parameters or initial design parameters. Finally, the energy consumption and dynamic performance of this method are analyzed through rapid prototyping control simulation. The results show that the ride comfort of the vehicle suspension can be improved with the given preset convergence rate.

An active suspension system ensures the controllability of a vehicle in the vertical direction, which greatly enhances the control redundancy and safety of an intelligent driven vehicle. However, many calibrated model parameters are not conducive to the application of optimal control. To reduce the control cost of active suspension, a model-free H_∞ output feedback control method is studied in this research. First, the optimal governing equation of the active suspension is transformed into a zero-sum game problem of two players, and an off-policy reinforcement learning algorithm is established to solve the game algebraic Riccati equation. This method could overcome the disadvantage of constant interactions between Q-learning and the environment. Secondly, with the consideration that some state variables are difficult to measure, a data-driven H_∞ output feedback controller is designed using road sensing information and historical measurement data, and the Bellman equation of the system is solved using the least squares method to obtain the optimal control solution of the active suspension. The simulation and rapid prototype experimental results show that the proposed method could produce the optimal control strategy of the system without model parameters, overcome the strong dependence and sensitivity of traditional design methods to model parameters, and improve the robust control effect of the active suspension.