In this article, based on a novel electro-hydraulic braking system for full self-driving vehicles, a model predictive control approach is designed for precisely tracking of braking pressure. Concerning electric motor, reduction mechanism, and hydraulic system, a new braking system model is formulated. Due to the nonlinear hydraulic characteristic of the master cylinder piston, a quadratic polynomial is utilized to fit a relationship between piston position and cylinder pressure, by which a desired pressure of master cylinder is transformed into a desired position of the piston. A recursive least square with forgetting factor is developed to find the optimal parameters of pressure–position relationship. A system friction model is established to calculate the equivalent friction torque, which is regarded as the compensation of the desired torque. A vehicle test bench of full self-driving is established to validate the designed approach. According to the accurate tracking and the quick response of braking pressure from experimental results, it can be concluded that the model predictive control approach is a good candidate for the pressure demand control of electro-hydraulic braking by wire.

In this article, based on the estimation of optimal slip ratio, a model predictive control approach is proposed for the antilock braking system with motor of a battery electric vehicle. A wheel dynamics model with a virtual control quantity is established. A method for estimating the optimal slip ratio of different roads is introduced, which consists of a sliding mode observer for longitudinal braking force and a recognizer of slip ratio. A ground slip map that describes the relationship between slip ratio and ground adhesion coefficient is proposed, which is utilized to identify the optimal slip ratio. Based on the model predictive control algorithm, a braking torque controller is designed to track the optimal slip ratio for good performance of braking stability and security. The vehicle control unit embedded with the estimator and model predictive control law is tested in a battery electric vehicle on the dry and wet roads. The experimental results indicate that the proposed approach has a good performance for the estimation and control of slip ratio during the antilock braking with electric motor, especially suitable for the low adhesion road conditions.

In this article, a novel steer-by-wire system architecture called iSteer is designed to realize the high performance of steering angle tracking for self-driving vehicles (especially, self-driving trucks), where the vehicle steers in response to the virtual driver’s command instead of the driver’s hand wheel. Based on a newly steering dynamical model of the iSteer system,an adaptive proportional-integral controller with parameters tuning by particle swarm optimization is developed for real-time or embedded application. Hence, the fast and efficient computation plays an essential role in our control law design. The developed steering controller not only meets this computational demand but also takes into account the self-aligning torque and ground resistance torque (as the disturbances to the iSteer control system). According to the system transfer function, the one-step prediction of steering angle is utilized to find the best particle of the swarm. Through a graph analysis using Nyquist stability criterion, we prove that the steering control system is stable and the error of steering angle tracking is convergent. Some experiments are tested on the physical iSteer system with the developed steering controller, the results of which further verify the effectiveness of the designed architecture and the high performance (e.g., high precision in tracking and fast response time) of the developed control law for steering angle tracking in self-driving vehicle without hand wheel.

In order to ensure the safety and service life of lithium-ion batteries for automotive applications, accurate state of charge (SOC) is required for system management in the process of driving. It is particularly challenging to estimate the SOC by using the online approach, for which the battery dynamics model is nonlinear. Many researchers have focused on model-based or data-driven algorithms alone, but comparatively few of them use a joint approach with the two types of algorithms. The data-driven algorithm is self-learning and has better adaptability, while the model-based algorithm is more stable and has stronger robustness. If these advantages can be combined, a better SOC estimation approach will be developed. In this article, based on battery charge dynamics, a complex fractional-order model of battery is simplified into a discrete fraction model for engineering application of control algorithm. A Bayesian belief network (BBN) is used to estimate the battery model parameters, and the adaptive extended Kalman particle filter (aEKPF) is used to estimate the SOC. In order to obtain accurate parameters of battery model for training, linear programming is used to identify the parameters online. Collectively, this article designs a joint approach of how the aEKPF with BBN estimates the SOC precisely. A developed approach has been downloaded into a battery control unit and tested in real-world conditions using a battery test bench to realize practical applications of the joint approach.

State of charge (SOC) plays a crucial role in battery management systems as it significantly impacts battery lifespan and energy efficiency. However, accurately estimating SOC is challenging due to the highly nonlinear electrochemical characteristics of batteries. Traditional machine learning algorithms struggle with substantial SOC estimation errors and limited convergence capability, particularly when faced with significant current fluctuations. Recursive algorithms heavily depend on battery models, which can introduce uncertainties during computation, potentially leading to system instability or divergence. Therefore, achieving accurate SOC monitoring has become a significant technical obstacle. To address these challenges, this study presents a novel SOC estimation algorithm named XGBoost-ACKF, which combines the strengths of extreme gradient boosting (XGBoost) and adaptive cubature Kalman filter (ACKF). XGBoost establishes a nonlinear mapping model between input and output characteristics, while ACKF filters and estimates the SOC approximation obtained from XGBoost. This integration yields highly accurate SOC estimation. Experimental results demonstrate that the proposed algorithm surpasses existing approaches in terms of estimation accuracy, generalization ability, and error convergence. Across various temperatures and testing conditions, the algorithm achieves a mean absolute error (MAE) and root mean square error (RMSE) of less than 1.06% and 1.25%, respectively. Furthermore, both RMSE and MAE are reduced by over 20% compared to the extended Kalman filter (EKF) and gradient-boosted decision tree (GBDT) methods. These findings establish the proposed algorithm as a promising solution for accurate SOC estimation in battery management system applications.