- Autonomous Vehicle Technology and Safety
- Traffic control and management
- Vehicle Dynamics and Control Systems
- Electric and Hybrid Vehicle Technologies
- Traffic Prediction and Management Techniques
- Vehicle emissions and performance
- Human-Automation Interaction and Safety
- Earthquake Detection and Analysis
- Real-time simulation and control systems
- Ionosphere and magnetosphere dynamics
- Advanced Battery Technologies Research
- Electric Vehicles and Infrastructure
- Traffic and Road Safety
- Transportation and Mobility Innovations
- Inhalation and Respiratory Drug Delivery
- Intravenous Infusion Technology and Safety
- Advanced Battery Materials and Technologies
- Advanced battery technologies research
- Rocket and propulsion systems research
- Pharmaceutical studies and practices
- Geoscience and Mining Technology
- Advanced Thermoelectric Materials and Devices
- Underwater Vehicles and Communication Systems
- Simulation and Modeling Applications
- Mechanical Engineering and Vibrations Research
Guilin University of Technology
2025
Chongqing University
2011-2024
China Jiliang University
2024
China Automotive Engineering Research Institute
2018-2020
Hong Kong Polytechnic University
2012
Wuhan University
2012
Knowledge transfer is a promising concept to achieve real-time decision-making for autonomous vehicles. This paper constructs deep reinforcement learning (RL) framework transform the driving tasks in intersection environments. The missions at unsignalized are cast into left turn, right and running straight automated goal of ego vehicle (AEV) drive through situation efficiently safely. objective promotes studied increase its speed avoid crashing other policy learned from one task transferred...
The formulation and device collectively constitute an inhalation drug product. Development of inhaled drugs must consider the compatibility between in order to achieve intended pharmaceutical performance usability product improve patient compliance with treatment instruction. From points formulation, use, this article summarizes drugs, including pressurized metered dose inhaler (pMDI), dry powder (DPI), nebulizer that are currently available US UK markets. It also discusses practical...
The ionosphere is characterized by a high concentration of ionized particles, plays critical role in the propagation characteristics radio signals upper atmosphere. During major geological events, such as earthquakes and volcanic eruptions, often exhibits anomalous disturbances. In event an earthquake, energy accumulated within Earth's crust released propagates along fault lines, generating waves—including Rayleigh waves, acoustic gravity waves—that ascend into induce fluctuations Total...
Typhoons, originating from tropical ocean surfaces, are among the most severe natural disasters worldwide, often leading to significant loss of life and property. Research indicates that Total Electron Content (TEC) ionosphere experiences various disturbances before after a typhoon event. This study utilizes global ionospheric data provided by Chinese Academy Sciences (CAS) analyze anomalies in TEC during 15-day period surrounding Super Typhoon "Meranti" (No. 14) 2016, employing Singular...
During magnetic storms, scintillation is instigated by significant disturbances in the ionosphere that deviate from mean level. Utilizing global ionospheric TEC and ROTI data, this study analyzes impact during substantial storm of December 2015. The analysis reveals storm, southern hemisphere was primarily characterized positive-phase whereas northern exhibited brief storms succeeded prolonged intense negative-phase storms. response to more pronounced at low mid-latitudes compared high...
This study proposes a novel mixed motion planning and tracking (MPT) control framework for autonomous vehicles (AVs) based on model predictive (MPC), which is made up of an MPC-based longitudinal module, feed-forward integrated lateral module. First, given the global reference path surroundings information obtained from onboard devices V2X network, vehicle kinematics applied to determine local target path, desired acceleration, velocity considering safety priority. Then, planned velocity,...
Testing in the closed field is one of important means to verify features and performance automated vehicles. Due complicated changeable traffic conditions, how design testing scenarios with relatively few number but significant value a problem that deserves research. We analyze Level 2 automatic driving production models market. Based on applicable main functions such as adaptive cruise control, active lane changing control keeping for vehicles, relative location among ego-vehicle its...
In order to improve vehicle driving safety in a low-cost manner, we used monocular camera study lane-changing warning algorithm for highway vehicles based on deep learning image processing technology. We improved the mask region-based convolutional neural network target detection. Suitable anchor frame ratios were obtained by means of K-means++ method clustering 66,389 targets with width/height ratio, which is one more set frames than original setting, so as ensure that generation accuracy...
Scenario-based testing is an important verification and certification measure to evaluate the safety of automated vehicles. In view existing test scenario composition methods, which may miss some critical problems that have low occurrence probability, we fully combined ego-vehicle with possible relative positions movement directions surrounding traffic participants based on a complex group. We applied scenario-screening rules obtain functional scenarios different environments driving task...
According to the requirements of heavy commercial compressed natural gas (CNG) vehicle satisfy China’s rapid economic development, authors develop and test a new high energy direct ignition system CNG engine. Hardware software are designed debugged. Based on tradition bench, an optimal bench for this engine is established. Ignition advance angle calibrated analyzed, which has economy power performances effects matched developed system. To emissions by system, control strategy simplified...
The traffic environment in the highway confluence area is complicated, and accidents are prone to occur when changing lanes, which a major difficulty vehicle driving. In view of low adaptability rule-based decision-making algorithms environmental changes simplification complexity existing learning-based algorithms, we had established Deep Q Network Double models for lane-changing decision an automated based on deep reinforcement learning, as well autonomous model vehicles, two new reward...
In order to study the dynamics coupling effect between human and vehicle on roll characteristics of miniature electric (MEV), firstly mathematical models for motion stable steering human-vehicle system were established, respectively. And then, virtual prototype driver MEVs built by ADAMS/Car. Finally, simulations conducted. The influences (HVDCE) MEV with different curb weight compared analysed. Results show that HVDCE exhibit characteristic lower equivalent angular stiffness damping under...
Electric automated vehicles are zero-emission, energy-saving, and environmentally friendly vehicles, testing verification is an important means to ensure their safety. Because of the scarcity dangerous scenarios in natural driving roads, it required conduct accelerated tests evaluations for electric vehicles. According scenario data road cut-in conditions, we used kernel density estimation method calculate probability distribution parameters. Additionally, Metropolis–Hastings algorithm...
In recent years, automated vehicles have de-veloped rapidly, and the safety of autonomous should be tested verified before they can enter mar-ket. The evaluation complexity driving tasks is one important foundations for design test scenarios tests vehicles. were divided into a series operation steps, five metrics number logic, traffic environment infor-mation, decision-making, motion planning track controlling de-termined through analysis By establishing behavior control charts or...
Ecological driving assistance technology is used to help driver improve operation behaviour, reduce the fuel consumption and emissions. The existing ecological system does not consider changing trends of vehicle state, real-time voice prompt time reaction led lag driver, resulting in economy deteriorated before complete adjust operation. In order predict correct driver`s dissipation behaviour timely, this paper proposed a predictive optimization model based on dynamic programming algorithm....
In addition to some test standards in the level 1 automated vehicles, it still lacks perfect evaluation procedures for 2 vehicles. The method of vehicle field L2 vehicles is studied, and multi-level index system preliminarily established from aspects safety, intelligence experiential. order relationship analytic hierarchy process are applied empower indicators at all levels. A comprehensive model was by using fuzzy grey method. Taking results three models on ACC mode as an example, a...
Automated vehicle testing and evaluation is an important guarantee for safety reliability. The current L3 automated procedures are not yet perfect. For the field test of vehicles, we proposed to establish a comprehensive index system from five dimensions safety, intelligence, experience, energy consumption, efficiency. A scientific method was designed select screen indicators in each dimension, preprocess behavior based on effect size. analytical hierarchy process entropy were used determine...
This paper introduces the structure of powertrain system based on hybrid urban bus. The scheme operation mode control is proposed according to analysis modes. Through vehicle tests, experimental results verify that shifting modes respond rapidly and correctly under controller.