Search Results

Self-piercing rivets (SPR) are efficient and economical joining methods used in the manufacturing of lightweight automotive bodies. The finite element method (FEM) is a potentially effective way to assess the joining process of SPRs. However, uncertain parameters could lead to significant mismatches between the FEM predictions and physical tests. Thus, a sensitivity study on critical model parameters is important to guide the high-fidelity modeling of the SPR insertion process. In this paper, an axisymmetric FEM model is constructed to simulate the insertion process of the SPR using LS-DYNA/explicit. Then, several surrogate models are evaluated and trained using machine learning methods to represent the relations between selected inputs (e.g., material properties, interfacial frictions, and clamping force) and outputs (cross-section dimensions).

Driving safety is an eternal theme of the transportation industry. In recent years, with the rapid growth of car ownership, traffic accidents have become more frequent, and the harm it brings to human society has become increasingly serious. In this context, car safety assisted driving technology has received widespread attention. As an effective means to reduce traffic accidents and reduce accident losses, it has become the research frontier in the field of traffic engineering and represents the trend of future vehicle development. However, there are still many technical problems that need to be solved. With the continuous development of computer vision technology, face detection technology has become more and more mature, and applications have become more and more extensive. This article will use the face detection technology to detect the driver's face, and then analyze the changes in driver's driving focus.

This paper presents a design process with data mining technique and advanced optimization strategy. The proposed design method provides insights in three aspects. First, data mining technique is employed for analysis to identify key factors of design variables. Second, relationship between multiple types of size and shape design variables and performance responses can be analyzed. Last but not least, design preference can be initialized based on data analysis to provide priori guidance for the starting design points of optimization algorithm. An exhaust system design problem which largely contributes to the improvement of vehicular Noise, Vibration and Harshness (NVH) performance is employed for the illustration of the process. Two types of design parameters, structural variable (gauge of component) and layout variable (hanger location), are considered in the studied case.

With the development of science and technology, breakthroughs have been made in the fields of intelligent algorithms, environmental perception, chip embedding, scene analysis, and multi-information fusion, which together prompted the wide attention of society, manufacturers and owners of autonomous vehicles. As one of the key issues in the research of autonomous vehicles, the research of vehicle lane change algorithm is of great significance to the safety of vehicle driving. This paper focuses on the conflict of interest between the lane-changing vehicle and the target lane vehicle in the fully autonomous driving environment, and proposes the method of coupling kinematics and game theory and reinforcement learning based optimization, so that when the vehicle is in the process of lane changing game, the lane-changing vehicle and the target lane vehicle can make decisions that are beneficial to the balance of interests of both sides.

Intelligent driving is an important research direction in the field of artificial intelligence. The fourth industrial revolution represented by the Internet of things provides more prospects for the development of intelligent vehicles. Trajectory planning and tracking control is one of the key technologies of intelligent driving vehicle. This paper takes intelligent driving vehicle as the starting point and establishes a research method of intelligent vehicle trajectory planning based on particle swarm optimization, based on the vehicle kinematics and dynamics model, a model predictive control algorithm is built for trajectory tracking control, the simulation scene is built by Prescan, the vehicle dynamics parameters are set in Carsim, and then the joint simulation is carried out with Simulink.

In recent years, game theory and reinforcement learning have become very popular research fields in today's society. As the most strategic analysis and optimization research method, they can be used in the study of subsidy strategy of China's new energy automobile industry to solve the problems caused by the government's subsidy of new energy vehicles. This paper studies the evaluation methods and strategy optimization methods of government subsidy strategies in different situations, and applies them to the subsidy strategies and other strategy optimization problems of new energy vehicles in China. Firstly, based on game theory, this paper studies the evaluation method of government subsidy strategy in the case of “double reciprocity” and “one strong and one weak” by constructing the game process of “double reciprocity” enterprises and “one strong and one weak” enterprises.

As a crucial part of the intelligent transportation system, traffic signal control will realize the boundary control of the traffic area, it will also lead to delays and excessive fuel consumption when the vehicle is driving at the intersection. To tackle this challenge, this research provides an optimized control framework based on reinforcement learning method and speed guidance strategy for the connected vehicle network. Prior to entering an intersection, vehicles are focused on in a specific speed guidance area, and important factors like uniform speed, acceleration, deceleration, and parking are optimized. Conclusion, derived from deep reinforcement learning algorithm, the summation of the length of the vehicle’s queue in front of the signal light and the sum of the number of brakes are used as the reward function, and the vehicle information at the intersection is collected in real time through the road detector on the road network.

Compared with traditional welding, self-piercing riveting technology has unique advantages and is widely used in automobile lightweight technology. The riveting quality of self-piercing riveting is closely related to the safety and durability of automobiles. The detection of riveting quality has gradually become an important part of the automobile manufacturing process. The generation of surface cracks under self-piercing riveting will affect the riveting strength, which in turn affects the riveting quality. Therefore, the detection of riveting external quality is transformed into the detection of riveting surface cracks. The existing artificial vision-based riveting lower surface crack recognition technology is inefficient, subjective and cannot be applied on a large scale. Therefore, this paper will propose a local-overall strategy based on image processing and computer vision.

The use of lightweight materials is one of the important means to reduce the quality of the vehicle, which involves the connection of dissimilar materials, such as the combination of lightweight materials and traditional steel materials. The riveting quality of self-piercing riveting (SPR) technology will directly affect the safety and durability of automobiles. Therefore, in the initial joint development process, the quality of self-piercing riveting should be inspected and classified to meet safety standards. Based on this, this paper divides the self-piercing riveting quality into riveting appearance quality and riveting section quality. Aiming at the appearance quality of riveting, the generation of cracks on the lower surface of riveting will seriously affect the riveting strength. The existing method of identifying cracks on the lower surface of riveting based on artificial vision has strong subjectivity, low efficiency and cannot be applied on a large scale.

Forward Collision Warning System is an important part of vehicle active safety system, it can reduce the occurrence of rear-end collision accidents with high fatality rate and improve the safety of driving. At present, there are still some outstanding issues to be addressed among the existing forward collision warning systems, such as the high cost of information acquisition based on LiDAR and other high-definition sensors, and the poor real-time performance of target detection based on vision. In view of the aforementioned issues and in order to improve the detection accuracy and real-time requirements of the target detection function of the early warning system, this paper proposes an enhanced deep learning model-based vehicle target detection method, and improves the key techniques of target detection, ranging and speed measurement and early warning strategy in the warning system.

Nowadays, the automobile industry is booming and the number of vehicles is proliferating while the road traffic environment is also deteriorating. Therefore, attention should be paid to the protection of vulnerable road users in traffic accidents, such as pedestrians. In order to reduce the pedestrians’ head injury in collision accidents, in this study, the vehicle engine hood which responds significantly to head injuries was taken as the design object, so as to put forward a new optimization design process. The parameters of the hood’s main components, manufacturing materials and structural scheme were considered to carry out simultaneous optimization from various aspects such as pedestrian protection and hood stiffness.

In recent years, aimed to promote the improvement of China’s new energy vehicle market, a series of incentive policies issued by the Chinese government: including the new energy vehicle subsidy policy, the double credit policy, and the charging pile infrastructure subsidy.Relevant research on new energy vehicle industry is mainly ground on multi-stage game, this paper employs multi-agent games theory, and summarizes the multi-agent decision-making optimization method in differential game based on dynamic programming and reinforcement learning. Then, in the context of new energy vehicles, research and improve the industrial policy of new energy vehicles through this method.A multi-agent differential game decision-making optimization framework is proposed. Complex multi-agent differential game decisions can be solved using the dynamic programming solver or deep reinforcement learning solver in this framework. Case studies and some observations will be given in the end.

Self-piercing rivet (SPR) joints are a key joining technology for lightweight materials, and they have been widely used in automobile manufacturing. Manual visual crack inspection of SPR joints could be time-consuming and relies on high-level training for engineers to distinguish features subjectively. This paper presents a novel machine learning-based crack detection method for SPR joint button images. Firstly, sub-images are cropped from the button images and preprocessed into three categories (i.e., cracks, edges and smooth regions) as training samples. Then, the Artificial Neural Network (ANN) is chosen as the classification algorithm for sub-images. In the training of ANN, three pattern descriptors are proposed as feature extractors of sub-images, and compared with validation samples. Lastly, a search algorithm is developed to extend the application of the learned model from sub-images into the original button images.

The control strategy of switched reluctance motor (SRM) in-wheel motor is investigated in order to reduce the influence of torque ripple of SRM on the ride comfort. The nonlinear model of switched reluctance motor (SRM) is established and the variable angle control strategy with optimal switch angles is applied to control SRM. However, the variable angle control strategy can not reduce the torque ripple of SRM significantly. Therefore, some advanced control strategies are developed to improve the ride comfort in electric vehicle. In this paper, the fuzzy proportional integration differential (PID) is developed to improve the torque ripple of SRM in which the fuzzy control idea is utilized to adjust the parameters of proportional integration differential (PID) control online and ensure the adaptive capabilities of the fuzzy proportional integration differential (PID) control to motor driving system.

This paper proposes a detection algorithm based on deep learning for Vulnerable Road Users such as pedestrians and cyclists, which is improved on the basis of YOLOv5 network model. (1) Aiming at the problems of low resolution and insufficient information for small targets, a multi-scale feature fusion method is adopted to integrate shallow features with deep features.

Most traffic accidents are caused by the driving behavior of drivers, so the study of driving behavior plays an important role in traffic safety. In this paper, we propose a comprehensive system for evaluating driver driving behavior safety, including classification and prediction of driver driving styles, as well as comprehensive scoring of driving behaviors. The driving behavior dataset needs to be pre-processed first. After that, we extracted driving behaviors and indicators, and performed cluster analysis on the reconstructed data to obtain three styles: conservative, aggressive, and impatient. We introduced a BP neural network to predict driving styles and obtained a relatively high prediction accuracy. Finally, we calculated the combined subjective and objective weights of the five behaviors and combined them with the developed scoring function to evaluate the driving safety of all drivers with a comprehensive score.

With economic development and the increasing number of vehicles in cities, urban transport systems have become an important issue in urban development. Efficient traffic signal control is a key part of achieving intelligent transport. Reinforcement learning methods show great potential in solving complex traffic signal control problems with multidimensional states and actions. Most of the existing work has applied reinforcement learning algorithms to intelligently control traffic signals. In this paper, we investigate the agent-based reinforcement learning approach for the intelligent control of ramp entrances and exits of urban arterial roads, and propose the Proximal Policy Optimization (PPO) algorithm for traffic signal control. We compare the method controlled by the improved PPO algorithm with the no-control method.