A5024997323- Autonomous Vehicle Technology and Safety
- Advanced Neural Network Applications
- Traffic and Road Safety
- Infrastructure Maintenance and Monitoring
- Physics of Superconductivity and Magnetism
- Traffic control and management
- Sparse and Compressive Sensing Techniques
- Advanced Measurement and Detection Methods
- Traffic Prediction and Management Techniques
- Video Surveillance and Tracking Methods
- Vehicle emissions and performance
- Industrial Vision Systems and Defect Detection
- Elevator Systems and Control
- Cooperative Communication and Network Coding
- Superconductivity in MgB2 and Alloys
- Human-Automation Interaction and Safety
- Advanced Decision-Making Techniques
- Magnetic properties of thin films
- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
- Advanced SAR Imaging Techniques
- Network Security and Intrusion Detection
- Research Data Management Practices
- Remote Sensing and LiDAR Applications
- Histone Deacetylase Inhibitors Research
Beijing Institute of Technology
2010-2024
Shanxi University
2024
Beijing Jiaotong University
2024
Ruijin Hospital
2024
Shanghai Jiao Tong University
2024
Shandong University
2024
Inspur (China)
2023
Shanghai Normal University
2023
Nankai University
2019
Beijing Transportation Research Center
2015
Due to the high mileage and heavy load capabilities of hybrid electric vehicles (HEVs), energy management becomes crucial in improving efficiency. To avoid over-dependence on hard-crafted models, deep reinforcement learning (DRL) is utilized learn more precise strategies (EMSs), but cannot generalize well different driving situations most cases. When cycles are changed, neural network needs be retrained, which a time-consuming laborious task. A efficient transferable way combine DRL...
Although autonomous vehicles have introduced a promising potential for improving traffic safety and efficiency, ensuring the of in complex road environments is still huge challenge to be tackled. To quickly quantify risk factors environments, this paper focuses mainly on influence depth breadth environment elements driving system, uses field theory establish model impact environmental combines AHP equivalent virtual electric quantity each element, so as realize quantitative evaluation...
Maternal distress increased during the COVID-19 pandemic, significantly impacting children's media use. The purpose of this study was to explore influence mechanism maternal on preschoolers' problematic us through a moderated mediation model; specifically, we examined possible mediating roles parenting stress and negative instrumental use in moderating role supportive co-parenting.An online survey conducted sample 1357 children (Mage = 4.01, SD 1.06; 47.4% boys) their parents from six public...
The primary goal of the paper is to explore human-vehicle-road interaction mechanism in traffic environment and evaluate complexity for unmanned vehicles urban roads.In particular, we propose quantitative evaluation models roads paper.Specifically, structure system complex dissected from aspect human-vehicle-road, laying basis proposing influencing factors complexity.We divide into static dynamic light relative characteristics various environmental elements.For environment, model established...
To improve the technology of unmanned ground vehicles, it is necessary to conduct a proper evaluation on various technologies. Previous methods are mainly based completion task; this may mislead most teams vehicles using conservative strategy during evaluation. In paper, new method proposed. Based typical working conditions including intersection, car-following, and obstacle-avoiding, indicator system established, entropy-cost function applied comprehensive vehicles. As reported in numerical...
In this paper, we propose a new adaptive pre-handover algorithm according to the characteristics of heterogeneous networks and users' demands for seamless connection. Firstly, movement user is predicted based on Gauss-Markov mobility model. Then, handover triggering time decided by value reference signal receiving quality (RSRQ) in position. Finally, reasonable decision speed service type using made reduce unnecessary handovers (UHO) lower failure (HOF) rate. Simulation results show that...
Gray correlation analysis combined with improved AHP (Analytic Hierarchy Process) method is applied to evaluate intelligent U-turn behavior of unmanned vehicles. On the basis establishing evaluation index system behavior, used determine weight every index, and then gray comprehensive model constructed, which can be directly converted score easily compared or ordered by results. The results applying scheme "Future Challenge 2012" show that scores four racing vehicles are 94.6, 60.5, 80.9,...
The traditional conflict resolution algorithm is designed for automated vehicles based on the premise of right way, but there an unclear road way at unsignalized intersections, which brings trouble to decision making vehicles. objective this work provide our system with evolving cooperative and non-cooperative decisions. We achieve by integrating game theory into making. When decides drive cooperatively, joint action planned optimize overall revenue multi-vehicles game, considering...
It has always been an important and arduous task to detect the complex dynamic traffic environment, especially for unmanned driving. Although existing advanced detection models have reached speed requirements detection, accuracy needs be further elevated improve driving's safety. How balance of detecting environment is still primary problem solved vehicles. Therefore, this article proposes a model vehicles by following framework idea YOLOv3. Firstly, we regard MobileNetv3 as backbone replace...
In the wake of development unmanned ground vehicle technologies, it is necessary to conduct a proper evaluation various technologies. Previous method saremainly based on completion task, this may mislead most teams using conservative strategy during evaluation. paper, new proposed. Based typical working conditions including intersection, car-following and obstacle-avoiding, indicator system established, information entropy applied comprehensive vehicles. Empirical results show that proposed...
１４２５ • 行更加安全可靠，考虑部件自身特点和部件间依赖关系的 动态维修决策优化研究具有重要意义。 多部件系统的部件间可能存在经济相关
The autonomous parking system is an intelligent which can park a car into spot. Recently, more and studies pay attention on this system. However, research evaluation of in these not enough. Because most methods vehicle are based task completion degree, the cannot show its real level. In order to ensure feasibility accuracy system, paper proposes effective method BP neural network. established index using network model. Using samples collected from vehicles, model trained by MATLAB software...
Nowadays, in the area of autonomous driving, computational power object detectors is limited by embedded devices and public datasets for driving are over-idealistic. In this paper, we propose a pipeline combining both block-wise pruning channel-wise to compress detection model iteratively. We enforce introduced factor residual blocks scale parameters Batch Normalization (BN) layers sparsity select less important channels. Moreover, modified loss function has been proposed remedy...
To promote the development of unmanned ground vehicle technologies, it is necessary to design a scientific and reasonable test method. Road an important part environmental elements, different road conditions can examine adaptability vehicles environment. Therefore, calculation complexity great importance. Previous studies on are mainly based concept roughness; however due unicity feature indicators, only be reflected certain extent. This paper proposes new road-feature-based multiparameter...
Staphylococcus aureus (S. aureus) pneumonia has become an increasingly important public health problem. Recent evidence suggests that epigenetic modifications are critical in the host immune defence against pathogen infection. In this study, we found S. infection induces expression of histone deacetylase 6 (HDAC6) a dose-dependent manner. Furthermore, by using mouse model, showed HDAC6 inhibitor, tubastatin A, demonstrates protective effect pneumonia, decreasing mortality and destruction...
Curriculum construction is an important part of improving the quality higher education. With rapid development artificial intelligence technology, generative AI having a profound impact on first-class courses in art majors colleges and universities. This study aims to explore how can enable majors, analyze its specific application effect curriculum through practical cases.
Abstract We consider a wavelet based image reconstruction model with the ℓ p (0 < 1) quasi-norm regularization, which is non-convex and non-Lipschitz minimization problem. For solving this model, Figueiredo et al (2007 IEEE Trans. Image Process. 16 2980–2991) utilized classical majorization-minimization framework proposed so-called Isoft algorithm. This algorithm computationally efficient, but whether it converges or not has been concluded yet. In paper, we propose new to accelerate...
Vehicle-infrastructure cooperative perception is becoming increasingly crucial for autonomous driving systems and involves leveraging infrastructure's broader spatial perspective computational resources. This paper introduces CoFormerNet, which a novel framework improving perception. CoFormerNet employs consistent structure both vehicle infrastructure branches, integrating the temporal aggregation module spatial-modulated cross-attention to fuse intermediate features at two distinct stages....
In complex urban scenarios like intersections without dedicated left-turn signals, the construction of planning systems that maximize efficiency while guarantee safety has been a significant challenge. this paper, we propose reinforcement learning approach based on curriculum using real world dataset, and develop partial end-to-end control model capable adapting to variable temporal spatial dimensional state inputs, applying it autonomous driving task. Our is compared with mainstream...