Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

FACA: Fair and Agile Multi-Robot Collision Avoidance in Constrained Environments with Dynamic Priorities

Jaskirat Singh, Rohan Chandra

TL;DR

FACA addresses the challenge of safe, scalable navigation for heterogeneous multi-robot systems operating in cluttered, constrained environments with dynamic priorities. It combines an LLM-based priority negotiation framework with a novel roundabout enhancement of artificial potential fields to achieve agile yet safe collision avoidance around both dynamic agents and static obstacles. The key contributions are the LLM-driven priority negotiation module, the Roundabout Effect that integrates non-linear attractive and tangential repulsive forces, and extensive simulations showing substantial improvements in Time to Goal and flow through tight gaps compared with classical APF, MPC, and RIPNA baselines. The approach demonstrates that language-mediated coordination, coupled with a tailored APF, can dramatically improve mission efficiency while maintaining robust safety in demanding real-world scenarios, with practical implications for crisis response, logistics, and emergency services.

Abstract

Multi-robot systems are increasingly being used for critical applications such as rescuing injured people, delivering food and medicines, and monitoring key areas. These applications usually involve navigating at high speeds through constrained spaces such as small gaps. Navigating such constrained spaces becomes particularly challenging when the space is crowded with multiple heterogeneous agents all of which have urgent priorities. What makes the problem even harder is that during an active response situation, roles and priorities can quickly change on a dime without informing the other agents. In order to complete missions in such environments, robots must not only be safe, but also agile, able to dodge and change course at a moment's notice. In this paper, we propose FACA, a fair and agile collision avoidance approach where robots coordinate their tasks by talking to each other via natural language (just as people do). In FACA, robots balance safety with agility via a novel artificial potential field algorithm that creates an automatic ``roundabout'' effect whenever a conflict arises. Our experiments show that FACA achieves a improvement in efficiency, completing missions more than 3.5X faster than baselines with a time reduction of over 70% while maintaining robust safety margins.

FACA: Fair and Agile Multi-Robot Collision Avoidance in Constrained Environments with Dynamic Priorities

TL;DR

FACA addresses the challenge of safe, scalable navigation for heterogeneous multi-robot systems operating in cluttered, constrained environments with dynamic priorities. It combines an LLM-based priority negotiation framework with a novel roundabout enhancement of artificial potential fields to achieve agile yet safe collision avoidance around both dynamic agents and static obstacles. The key contributions are the LLM-driven priority negotiation module, the Roundabout Effect that integrates non-linear attractive and tangential repulsive forces, and extensive simulations showing substantial improvements in Time to Goal and flow through tight gaps compared with classical APF, MPC, and RIPNA baselines. The approach demonstrates that language-mediated coordination, coupled with a tailored APF, can dramatically improve mission efficiency while maintaining robust safety in demanding real-world scenarios, with practical implications for crisis response, logistics, and emergency services.

Abstract

Multi-robot systems are increasingly being used for critical applications such as rescuing injured people, delivering food and medicines, and monitoring key areas. These applications usually involve navigating at high speeds through constrained spaces such as small gaps. Navigating such constrained spaces becomes particularly challenging when the space is crowded with multiple heterogeneous agents all of which have urgent priorities. What makes the problem even harder is that during an active response situation, roles and priorities can quickly change on a dime without informing the other agents. In order to complete missions in such environments, robots must not only be safe, but also agile, able to dodge and change course at a moment's notice. In this paper, we propose FACA, a fair and agile collision avoidance approach where robots coordinate their tasks by talking to each other via natural language (just as people do). In FACA, robots balance safety with agility via a novel artificial potential field algorithm that creates an automatic ``roundabout'' effect whenever a conflict arises. Our experiments show that FACA achieves a improvement in efficiency, completing missions more than 3.5X faster than baselines with a time reduction of over 70% while maintaining robust safety margins.

Paper Structure

This paper contains 22 sections, 16 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Model-Based Approaches
  4. Learning-based Approaches
  5. Multi-Robot Communication through LLMs
  6. Problem Formulation and Background
  7. FACA Framework
  8. Technical Approach
  9. Collision Avoidance Around Dynamic Agents: The "Roundabout Effect" (Solving Constraint \ref{['eq:constraint1']})
  10. LLMs for Priority Determination and Negotiation (Solving Constraint \ref{['eq:constraint3']})
  11. Non-linear Exponential Attractive Force
  12. Tangential Exponential Repulsive Force
  13. Collision Avoidance Around Static Obstacles (Solving Constraint \ref{['eq:constraint2']})
  14. Final Velocity Computation using $\theta_i$
  15. Experiments and Results
  16. ...and 7 more sections

Figures (5)

  • Figure 1: An overview of the proposed framework for autonomous robot collision avoidance using Large Language Model (LLM) based negotiation in heterogenous robots. The system operates in a loop, beginning with agent initialization. Upon predicting a potential collision, an LLM is used to dynamically negotiate priorities based on mission context. As illustrated in the LLM-based Priority Negotiation box where the "Medical robot" is granted higher priority over a "Logistic robot" with inputs as mission, and urgency to output new, dynamically-assigned priorities. The framework then applies scenario-specific adjustments for challenges like obstacle or small-gap passages before executing the final collision avoidance maneuvers.
  • Figure 2: FACA
  • Figure 3: Classical APF
  • Figure 4: MPC
  • Figure 6: Visually demonstrating the Roundabout effect of FACA in both complex and simple scenarios.