Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Strategic Pseudo-Goal Perturbation for Deadlock-Free Multi-Agent Navigation in Social Mini-Games

Abhishek Jha, Tanishq Gupta, Sumit Singh Rawat, Girish Kumar

TL;DR

The paper tackles deadlock in multi-agent navigation within constrained social mini-games and proposes Strategic Pseudo-Goal Perturbation (SPGP). SPGP integrates Safety Barrier Certificates with a strategic pseudo-goal perturbation to steer agents away from deadlock while preserving safety and then resume progress toward the original goals. Key contributions include the SPGP framework, a pseudo-goal selection rule with radius $\delta$, a quadratic programming formulation with SBC constraints, and extensive simulations showing improved success rates and reduced makespan across four social navigation scenarios. The work offers practical impact for autonomous systems operating in crowded environments and opens avenues for ML-assisted congestion management and coordination of heterogeneous agents.

Abstract

This work introduces a Strategic Pseudo-Goal Perturbation (SPGP) technique, a novel approach to resolve deadlock situations in multi-agent navigation scenarios. Leveraging the robust framework of Safety Barrier Certificates, our method integrates a strategic perturbation mechanism that guides agents through social mini-games where deadlock and collision occur frequently. The method adopts a strategic calculation process where agents, upon encountering a deadlock select a pseudo goal within a predefined radius around the current position to resolve the deadlock among agents. The calculation is based on controlled strategic algorithm, ensuring that deviation towards pseudo-goal is both purposeful and effective in resolution of deadlock. Once the agent reaches the pseudo goal, it resumes the path towards the original goal, thereby enhancing navigational efficiency and safety. Experimental results demonstrates SPGP's efficacy in reducing deadlock instances and improving overall system throughput in variety of multi-agent navigation scenarios.

Strategic Pseudo-Goal Perturbation for Deadlock-Free Multi-Agent Navigation in Social Mini-Games

TL;DR

The paper tackles deadlock in multi-agent navigation within constrained social mini-games and proposes Strategic Pseudo-Goal Perturbation (SPGP). SPGP integrates Safety Barrier Certificates with a strategic pseudo-goal perturbation to steer agents away from deadlock while preserving safety and then resume progress toward the original goals. Key contributions include the SPGP framework, a pseudo-goal selection rule with radius , a quadratic programming formulation with SBC constraints, and extensive simulations showing improved success rates and reduced makespan across four social navigation scenarios. The work offers practical impact for autonomous systems operating in crowded environments and opens avenues for ML-assisted congestion management and coordination of heterogeneous agents.

Abstract

This work introduces a Strategic Pseudo-Goal Perturbation (SPGP) technique, a novel approach to resolve deadlock situations in multi-agent navigation scenarios. Leveraging the robust framework of Safety Barrier Certificates, our method integrates a strategic perturbation mechanism that guides agents through social mini-games where deadlock and collision occur frequently. The method adopts a strategic calculation process where agents, upon encountering a deadlock select a pseudo goal within a predefined radius around the current position to resolve the deadlock among agents. The calculation is based on controlled strategic algorithm, ensuring that deviation towards pseudo-goal is both purposeful and effective in resolution of deadlock. Once the agent reaches the pseudo goal, it resumes the path towards the original goal, thereby enhancing navigational efficiency and safety. Experimental results demonstrates SPGP's efficacy in reducing deadlock instances and improving overall system throughput in variety of multi-agent navigation scenarios.
Paper Structure (16 sections, 11 equations, 5 figures, 1 table, 1 algorithm)

This paper contains 16 sections, 11 equations, 5 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Safety Based Methods
  4. Learning Based Methods
  5. Deadlock Resolution Methods
  6. Preliminaries
  7. Control Barrier Functions
  8. Safety Barrier Certificates
  9. Deadlock Condition
  10. Proposed Methodology
  11. Mathematical Formulation
  12. Safety Barrier Certificates
  13. Pseudo-Goal Selection Strategy
  14. Optimization Problem Formulation
  15. Results and Discussions
  16. ...and 1 more sections

Figures (5)

  • Figure 1: Overview of Strategic Pseudo-Goal Perturbation
  • Figure 2: Condition for deadlock among agents to occur
  • Figure 3: Pseudo goal selection to resolve deadlock and enforce separation between agents
  • Figure 4: Planned trajectories of two agent in social navigation settings using strategic perturbation
  • Figure 5: Effect of different perturbation radius on the makespan of agent.