Xiaoqiang Sun

To address the nonlinear instability of vehicles under extreme operating conditions such as high speed, low road adhesion, and large steering angles, this paper establishes a lateral–yaw dynamics model considering tire nonlinear characteristics. The effects of longitudinal vehicle speed, front-wheel steering angle, and road adhesion coefficient on vehicle chaotic bifurcation behavior and phase-plane stability boundaries are analyzed. Based on a phase-plane point-set identification method, an adaptive stability region varying with operating conditions is constructed and transformed into a state constraint for model predictive control. Accordingly, a boundary-constrained model predictive control method incorporating nonlinear stability boundaries, referred to as BMPC, is proposed. In addition, the local stability of the system near equilibrium points is analyzed using the Lyapunov method. Co-simulation and hardware-in-the-loop test results demonstrate that the proposed BMPC method can effectively prevent the vehicle states from entering unstable regions. Under complex road conditions and extreme handling scenarios, the method achieves improved path-tracking accuracy, lateral stability, and robustness. The results provide theoretical support and engineering guidance for stability control of intelligent vehicles under complex driving conditions.

In this paper, a deep operator network (DeepONet)-based feedforward–feedback control framework for autonomous vehicles’ path tracking controller is introduced. In the proposed framework, DeepONet is utilized for data-driven vehicle dynamics modeling within the feedforward control component. To improve the accuracy of lateral dynamics, the physical model is integrated into the DeepONet architecture as prior knowledge. By incorporating experimentally validated physical models, the proposed approach captures the causal relationship between vehicle parameters (e.g. mass, tire-road friction) and vehicle dynamics, thereby enhancing the interpretability of the model. In the CarSim/Simulink simulation environment, the designed DeepONet-based feedforward–feedback controller is evaluated in two scenarios: double-lane change and oval track. The validation results demonstrate that the proposed approach achieves smaller lateral errors than other methods in both linear and nonlinear regions. Notably, when facing unknown roads and varying road friction conditions on the oval track, the DeepONet-based approach shows improvement in tracking accuracy compared to purely data-driven methods without physical model integration.

Accurate estimation of the tire longitudinal slip ratio is essential for improving vehicle stability and traction control under both driving and braking conditions. To enhance estimation robustness and reduce the dependence on empirical sensor placement in intelligent tire applications, this paper proposes a strain-based longitudinal slip ratio estimation framework with optimized sensor configuration and data-driven modeling. A finite element model of a strain-type intelligent tire is first developed and experimentally validated to ensure the reliability of strain responses. A Pearson correlation coefficient-based method is then proposed to determine the optimal sensor locations to improve sensitivity to slip variations. Key strain features are further selected using the variable importance in projection scores of partial least squares regression. Based on these features, four estimation models including a long short-term memory network with attention mechanism (LSTM-ATT), LSTM, support vector machine, and LSTM-ATT2 are constructed. The LSTM-ATT2 model uses strain data from uniformly distributed sensors as an empirical baseline to verify the effectiveness of the proposed optimal sensor placement strategy. Results from 10-fold cross-validation show that LSTM-ATT achieves the best performance with an average root mean square error of 0.0313. The proposed method also demonstrates strong robustness under complex conditions including varying speeds and step changes in road friction.

1. With the rapid development of intelligent electric vehicle technology, the 4-wheel independent steering (4WIS) vehicles have emerged as a novel chassis architecture that highly integrates drivin...

In recent years, end-to-end autonomous driving has achieved remarkable progress, yet challenges remain in achieving efficient deployment. Existing methods often rely on multi-modal sensor fusion, which incurs high costs and computational demands, limiting their large-scale adoption in low-cost scenarios. Reducing system cost while maintaining performance has consistently been a key focus of the industry. Moreover, the opaque nature of end-to-end models, due to the lack of intermediate representations, limits interpretability and hampers root-cause analysis when misjudgments occur in complex environments, thereby posing safety risks. To address these challenges, this paper proposes a lightweight and interpretable end-to-end autonomous driving architecture that utilizes only monocular camera images as input. The aim is to enhance decision-making efficiency and trajectory planning accuracy while ensuring system safety. The proposed architecture comprises two core components. First, in the scene perception part, a depth information supervision module is introduced to enable hierarchical geometric perception, and an online vector map construction module is designed to build structured environmental representations at low cost. This replaces high-definition maps with explicit constraint expressions, thereby improving model interpretability and decision-making reliability. Second, in the planning part, a dual-layer prediction framework is developed to decouple speed and waypoint prediction. It independently forecasts vehicle velocity and waypoints, while an optimized decoder structure produces fine-grained spatiotemporal trajectories, enhancing planning capabilities in complex traffic scenarios. Experimental results on multiple benchmarks from the CARLA Leaderboard 1.0 demonstrate that the proposed method significantly improves system efficiency while maintaining advanced decision-making performance. It outperforms existing multi-sensor fusion approaches, showcasing strong potential for practical deployment.

Driver’s attention area recognition (DAAR) is crucial for autonomous vehicle takeover and detecting distracted driving. Currently, deep learning-based models for DAAR have achieved promising results. However, due to substantial variations in driving environments and individual driver characteristics, end-to-end deep learning models struggle to maintain stable generalization performance across different driving scenarios. To address these challenges, this paper proposes a driver attention area recognition framework based on improved Transformer and Kalman Filtering (DAARF-TK). DAARF-TK comprises two key components: a facial landmarks extraction module and a Transformer modeling module. The face landmarks extraction module includes an Adaptive Unscented Kalman Face Landmark Filtering (AUKFLF) algorithm, calibrating face landmarks in scenarios with large face occlusions, significant head rotations, and blurriness. An efficient attention mechanism and a Tokens Dimension Reduction Module are embedded into the Transformer to enhance the adaptability of the Transformer architecture to the real-time requirements of the DAAR task. Experimental results reveal that when trained on the Recorded dataset, DAARF-TK attains mAcc values of 94.13%, 91.67%, and 75.38% on the Recorded, 100-Driver, and DGW test sets, respectively. In addition, a real-time validation conducted on a driving simulator indicates that DAARF-TK operates at 37.68 frames per second (FPS), confirming its real-time capability in DAAR. These findings underscore the superiority of DAARF-TK and its potential to offer technical support for autonomous vehicle takeover.

Indoor parking lots represent a critical scenario for the commercialization of intelligent vehicles. The absence of GNSS signals, coupled with complex environmental structures and frequent dynamic interference, makes high-precision localization a core and challenging technology for achieving autonomous parking in such settings. Existing research methods based on multi-sensor fusion often lack strong constraints from high-level behavioral decisions when addressing these challenges, leading to insufficient localization accuracy and robustness in complex dynamic environments. To address this, this paper proposes an innovative approach that deeply integrates the planned path for autonomous parking with multi-sensor data to achieve high-precision localization in underground parking lots. Firstly, the A* and DWA path planning algorithms are improved to generate smoother trajectories that better adapt to dynamic environments. Secondly, in a novel contribution, the real-time optimized velocity and steering angle information fed back from path planning is utilized as key observations. This information is deeply fused with LiDAR and IMU data and embedded into a Bundle Adjustment (BA) optimization framework. This integration compensates for the accuracy degradation associated with velocity estimation relying solely on IMU pre-integration in traditional methods. Experimental results demonstrate the excellent performance of the proposed method across multiple real indoor parking lots. The average localization accuracy stably reaches around 0.2 meters, and notably, achieves 0.2315 meters even in complex commercial underground parking environments, confirming its stability and reliability in complex and dynamic scenarios.