LIVEPOINT: Fully Decentralized, Safe, Deadlock-Free Multi-Robot Control in Cluttered Environments with High-Dimensional Inputs

Jeffrey Chen, Rohan Chandra
University of Virginia
Paper arXiv Code

ORCA Doorway

MPC-CBF Doorway

MP-Net Doorway

Single Agent Doorway

LivePoint w/o Liveness Doorway

ORCA Intersection

MPC-CBF Intersection

MP-Net Intersection

Single Agent Intersection

LivePoint w/o Liveness Intersection

LivePoint Doorway

LivePoint Intersection



Abstract

Fully decentralized, safe, and deadlock-free multirobot navigation in dynamic, cluttered environments is a critical challenge in robotics. Current methods require exact state measurements in order to enforce safety and liveness e.g. via control barrier functions (CBFs), which is challenging to achieve directly from onboard sensors like lidars and cameras. This work introduces LIVEPOINT, a decentralized control framework that synthesizes universal CBFs over point clouds to enable safe, deadlock-free real-time multi-robot navigation in dynamic, cluttered environments. Further, LIVEPOINT ensures minimally invasive deadlock avoidance behavior by dynamically adjusting agents’ speeds based on a novel symmetric interaction metric. We validate our approach in simulation experiments across highly constrained multi-robot scenarios like doorways and intersections. Results demonstrate that LIVEPOINT achieves zero collisions or deadlocks and a 100% success rate in challenging settings compared to optimization-based baselines such as MPC and ORCA and neural methods such as MPNet, which fail in such environments. Despite prioritizing safety and liveness, LIVEPOINT is 35% smoother than baselines in the doorway environment, and maintains agility in constrained environments while still being safe and deadlock-free..

Image 1

Distances between each agent and the closest static obstacle, as well as the inter-agent distance, plotted over time. The CBF boundary is included to indicate the safety threshold for maintaining safe navigation.


Image 2

Liveness values computed at each time step, with liveness threshold and CBF boundary overlaid. Velocity perturbation is performed when liveness value falls below the liveness threshold.


Image 3

Velocity profiles of both robots using LIVEPOINT frame- work, shown before and after navigating through a doorway and an intersection, highlighting the velocity adjustments made to ensure safe and deadlock-free passage.




BibTeX

@article{chen2025livepoint,
  title={LiveNet: Robust, Minimally Invasive Multi-Robot Control for Safe and Live Navigation in Constrained Environments},
  author={Chen, Jeffrey and Chandra, Rohan},
  journal={arXiv preprint arXiv:2412.04659},
  year={2025}
}