A5033009482- Traffic Prediction and Management Techniques
- Video Surveillance and Tracking Methods
- Simulation and Modeling Applications
- Autonomous Vehicle Technology and Safety
- VLSI and Analog Circuit Testing
- Integrated Circuits and Semiconductor Failure Analysis
- Advanced Sensor and Control Systems
- Remote Sensing and LiDAR Applications
- Vehicle License Plate Recognition
- Machine Fault Diagnosis Techniques
- Infrared Target Detection Methodologies
- Infrastructure Maintenance and Monitoring
- Traffic control and management
- Advanced Measurement and Detection Methods
- Human Pose and Action Recognition
- Industrial Technology and Control Systems
- Robotic Path Planning Algorithms
- Railway Engineering and Dynamics
Jilin University
2023-2025
State Key Laboratory of Automotive Simulation and Control
2023
Predicting future trajectories is crucial for autonomous vehicles, as accurate predictions enhance safety and inform subsequent decision-making planning modules. This however a challenging task due to the complex interactions between surrounding vehicles. Existing methods struggled extract deep representations often overlook spatial dependence. To address this problem, paper introduces GIVA, an interaction-aware trajectory prediction method based on Gated Recurrent Unit (GRU)-Improved Visual...
The presence of obstacles on railway tracks poses a significant threat to the safety train operations. In order accurately obtain information about obstacles, we propose novel obstacle detection method that combines camera and lidar technology. Initially, improved YOLOv5 network is employed extract 2D positional enhancing feature representation capabilities through dual-feature fusion channels. A two-stage serial structure then utilized progressively refine layer information. Subsequently,...
In unstructured environments, obstacles are diverse and lack lane markings, making trajectory planning for intelligent vehicles a challenging task. Traditional methods typically involve multiple stages, including path planning, speed optimization. These require the manual design of numerous parameters each stage, resulting in significant workload computational burden. While end-to-end simple efficient, they often fail to ensure that meets vehicle dynamics obstacle avoidance constraints...
Railway switches are vital for ensuring the safety of railroad transportation by managing transitions between different tracks. We propose a skip feature enhanced multi-source fusion network with an attention mechanism effectively identifying switch states. To further improve performance, we data augmentation method region horizontal splice 16 images and novel intersection over union loss function that takes into account area aspect ratio both predicted ground truth boxes. Finally, state...
In autonomous driving, accurately predicting the trajectories of surrounding vehicles is essential, particularly in dense and heterogeneous urban traffic. We propose a graph-structured model with category layer to efficiently forecast target vehicle's trajectory. The enables flexible selection interacting objects based on environmental interactions extracts spatial-temporal features using graph convolutional network. A categorical introduced account for different influences dynamic agents,...