Rongliang Sun
Papers
2
Total Citations
4
H-Index
1
About
Rongliang Sun is a rising researcher at the forefront of autonomous driving and intelligent transportation systems, with a focused expertise in 3D perception, multi-object tracking, and trajectory prediction. His work addresses two of the most critical challenges in autonomous navigation: reliably tracking objects in dynamic point cloud data and predicting the future motion of heterogeneous agents in complex, high-risk environments. In his 2024 paper on point cloud multi-object tracking, Sun introduced an innovative framework that leverages intra-frame graph structures and inter-frame bipartite graph matching, enhanced by re-identification-based occlusion resilience. This approach directly tackles the persistent problem of identity switches and track loss when objects are temporarily hidden, achieving 3 citations and setting a new standard for robust 3D tracking. Complementing this, Sun’s work on trajectory prediction presents a novel heterogeneous multi-agent risk-aware graph encoder paired with a continuous parameterized decoder. This model explicitly accounts for diverse road users—from vehicles to pedestrians—and their interactions at complex intersections, significantly improving collision risk assessment. With a combined citation count of 4 for these foundational contributions, Sun is establishing himself as a key innovator in creating safer, more perceptive autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2