A dynamic tire model based on HPSO-SVM

The automatic parking system based on vision is greatly affected by uneven lighting, which is difficult to make an accurate judgment on parking spaces in the case of complex image information. To solve this problem, this paper proposes a parking space visual detection and image processing method based on deep learning. Firstly, a 360-degree panoramic system was designed to photograph the vehicle environment. The image has been processed to obtain a panoramic aerial view, which was input as the original image of the parking space detection system. Secondly, the Faster R-CNN (Region-Convolutional Neural Network) parking detection model was established based on deep learning. It was aimed to detect and extract the parking space from the input image. Thirdly, the problems of uneven illumination and complex background were solved effectively by removing the background light from the image. Finally, a parking space extraction method based on the connected region has been designed, which further simplified the parking space extraction and image processing. The experiment results show that the mAP (mean Average Precision) value of the Faster R-CNN model using 101-Floor ResNet as the feature extraction network is 89.30%, which is 2.28% higher than that of the Faster R-CNN model using 50-Floor ResNet as the feature extraction network. The model built in this paper can detect most parking spaces well. The position of the output target box is accurate. In some test scenarios, the confidence of parking space recognition can even reach 100%. In summary, the proposed method can realize the effective identification and accurate positioning of parking spaces.

The current vehicle stability judgment methods in the stability control of distributed drive electric vehicles focuses on using the phase plane method, but the existing phase plane stability boundary delineation methods ignore the influence of the front wheel rotation angle. Therefore, a new phase plane partitioning method (NPPPM) is proposed in this paper, and a new phase plane stability domain library is established. Based on this, the direct yaw moment control of distributed drive electric vehicle through phase plane analysis is established. Firstly, the influence of the front wheel turning angle on the phase plane is considered in the design of the phase plane stability boundary, and a new stability boundary function is fitted by the least squares method. This approach avoids the lateral instability caused by too large a front-wheel turning angle and improves the accuracy of vehicle stability judgment. Next, a hierarchical direct transverse moment controller is constructed using Matlab/Simulink. Among them, the upper layer controller adopts sliding mode control with the sideslip angle error of the center of mass as the tracking target. The lower controller adopts the optimal configuration algorithm with the optimal road adhesion utilization as the objective function. Finally, the effectiveness and superiority of the proposed phase plane partitioning method (NPPPM) for distributed drive electric vehicles with sinusoidal and angular order inputs for the front wheel rotation angle are verified by simulation.

Sensor fusion in intelligent suspension systems constitutes a fundamental technology for optimizing vehicle dynamic stability, ride comfort, and occupant safety. By integrating data from multiple sensor modalities, this study proposes a hierarchical multi-sensor fusion framework for active suspension control, aiming to enhance control precision. Initially, a binocular vision system is employed for target detection, enabling the identification of lane curvature initiation points and speed bumps, with real-time distance measurements. Subsequently, the integration of Global Positioning System (GPS) and inertial measurement unit (IMU) data facilitates the extraction of road elevation profiles ahead of the vehicle. A BP-PID control strategy is implemented to formulate mode-switching rules for the active suspension under three distinct road conditions: flat road, curved road, and obstacle road. Additionally, an ant colony optimization algorithm is utilized to fine-tune four suspension parameters. Utilizing the hardware-in-the-loop (HIL) simulation platform, the observed reductions in vertical, pitch, and roll accelerations were 5.37%, 9.63%, and 11.58%, respectively, thereby substantiating the efficacy and robustness of this approach.

As the upgrade of people’s requirements for automotive driving comfort, conventional passive suspensions for cars have fallen short of existing demands due to their nonadjustable damping and stiffness, so semiactive suspensions and active suspensions have gained growing acceptance. Compared with active suspensions, semiactive suspensions offer the advantages of a low manufacturing cost and reliable structure, and thus have become the preferred choice for most vehicles. To optimize the control effect of semiactive suspensions under different working conditions, this paper completed the modeling of magnetorheological semiactive suspension system dynamics and road inputs; then, the design of binocular camera sensing algorithms was performed to obtain the real-time distance of the target using the point cloud ranging function, and the parameters required for suspension control were also obtained. This was followed by the completion of the control-mode-switching rules and the design of the suspension controller. According to the different control objectives, the mode could be divided into the obstacle-road mode, straight-road mode, and curved-road mode. The suspension controller included the BP-PID (neural network PID controller) controller and the force distributor. Finally, the effectiveness of the mode-switching rules and the control method was verified through system simulation and the hardware-in-the-loop test.

Semi-active charging suspension has been the highlight in the research of ride comfort, handling, and safety of road vehicles in real time. Adjustable damping shock absorber is the key part of semi-active suspension. Many studies are focused on the control and impacts of automotive ride comfort. However, few of them are about the relationship among the damping of adjustable damping shock absorber, handling stability, and safety. In this article, a full car model based on multi-body dynamics was built, including the steering system, front and rear suspensions, tire, driving controller, and road. And the model was verified by tests. Based on the co-simulation, a controller was built based on hybrid sensor network control. The hybrid network control principle was switched among comfort controller, stability controller, and safety controller, in accordance with working conditions. The design effectively improved ride comfort, handling stability, and driving safety. Finally, a rapid control prototype was built based on dSPACE to conduct a real vehicle test. By comparison of the time response diagram, the results on pulse input and S-shaped road indicate that handling stability and driving safety enter into the stable domain and negative effects are successfully suppressed.