Papers
12
Total Citations
405
H-Index
8
About
Panpan Cai is a robotics and autonomous systems researcher whose work sits at the intersection of motion planning, decision-making under uncertainty, and autonomous driving. Her research focuses on enabling robots and self-driving vehicles to navigate safely and efficiently in complex, dynamic environments populated by pedestrians and other agents. Cai's most influential contribution is the PORCA framework (2018, 207 citations), a planning system for autonomous driving among crowds that introduced a novel pedestrian motion prediction model integrating both global navigation intent and local collision-avoidance behavior. This work established her as a leading voice in human-aware autonomous navigation. She further advanced the field with HyP-DESPOT (2020, 45 citations), a hybrid parallel algorithm that dramatically improves computational efficiency for online planning under uncertainty — a critical challenge for real-world robotics deployment. Her LeTS-Drive system (2019) pushed boundaries further by combining tree search with learning-based methods to tackle crowded intersection scenarios modeled as partially observable decision problems. More recently, her work on closing the planning–learning loop (2022) reflects a growing interest in tightly integrating planning and machine learning for robust real-time performance. With nearly 400 cumulative citations across her key publications, Cai's research has meaningfully shaped modern approaches to safe, intelligent autonomous navigation.
Research Focus
Key Achievements
Top Papers
- 1PORCA: Modeling and Planning for Autonomous Driving Among Many Pedestrians207 citations · 2018
- 2
- 3Collision Detection Using Axis Aligned Bounding Boxes41 citations · 2013
- 4LeTS-Drive: Driving in a Crowd by Learning from Tree Search34 citations · 2019
- 5
- 6PORCA: Modeling and Planning for Autonomous Driving among Many Pedestrians13 citations · 2018
- 7Closing the Planning–Learning Loop With Application to Autonomous Driving12 citations · 2022
- 8
- 9
- 10LeTS-Drive: Driving in a Crowd by Learning from Tree Search8 citations · 2019