Minimally Invasive Social Navigation
Stefan H. Kiss, K. Y. Cadmus To, Chanyeol Yoo, Robert Fitch, Alen, Alempijevic

TL;DR
This paper introduces a flow field representation of crowds for robot navigation, aiming to minimize human disturbance, and presents an algorithmic framework with promising experimental results.
Contribution
It proposes a novel flow field approach for social navigation that reduces invasiveness, enhancing computational efficiency and safety in robot-human interactions.
Findings
Flow field representation effectively models crowd movement.
The proposed algorithm reduces invasiveness in navigation.
Experimental results demonstrate improved navigation performance.
Abstract
Integrating mobile robots into human society involves the fundamental problem of navigation in crowds. This problem has been studied by considering the behaviour of humans at the level of individuals, but this representation limits the computational efficiency of motion planning algorithms. We explore the idea of representing a crowd as a flow field, and propose a formal definition of path quality based on the concept of invasiveness; a robot should attempt to navigate in a way that is minimally invasive to humans in its environment. We develop an algorithmic framework for path planning based on this definition and present experimental results that indicate its effectiveness. These results open new algorithmic questions motivated by the flow field representation of crowds and are a necessary step on the path to end-to-end implementations.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsEvacuation and Crowd Dynamics · Video Surveillance and Tracking Methods · Robotic Path Planning Algorithms
