Papers
47
Total Citations
6,426
H-Index
34
About
Jur van den Berg is a prominent robotics researcher whose work spans multi-agent motion planning, collision avoidance, planning under uncertainty, and robotic manipulation. He is perhaps best known for his foundational contributions to reciprocal collision avoidance, most notably the Reciprocal Velocity Obstacle (RVO) framework and its extension to n-body scenarios, which has become a cornerstone of multi-robot and crowd simulation research, accumulating over 1,800 citations. His Hybrid Reciprocal Velocity Obstacle work further refined these methods to eliminate oscillatory behavior in decentralized robot navigation. Beyond collision avoidance, van den Berg has made significant contributions to probabilistic motion planning, introducing LQG-MP for planning under sensor and motion uncertainty, and developing belief-space optimization techniques for partially observable environments. His Kinodynamic RRT* extended asymptotically optimal planning to robots with complex dynamics. Remarkably, his research breadth extends to robotic manipulation — including a widely cited geometric approach to autonomous laundry folding — and surgical robotics, where he demonstrated superhuman performance of robotic surgical subtasks through iterative learning. With multiple papers exceeding 200 citations, van den Berg's work has had lasting, cross-disciplinary impact on autonomous systems research.
Research Focus
Key Achievements
Top Papers
- 1Reciprocal n-Body Collision Avoidance1,811 citations · 2011
- 2The Hybrid Reciprocal Velocity Obstacle450 citations · 2011
- 3
- 4
- 5
- 6Reciprocal collision avoidance with acceleration-velocity obstacles286 citations · 2011
- 7A geometric approach to robotic laundry folding252 citations · 2011
- 8Generalized velocity obstacles244 citations · 2009
- 9
- 10Anytime path planning and replanning in dynamic environments237 citations · 2006