Karma Mechanisms for Decentralised, Cooperative Multi Agent Path Finding

2026-04-09

AI Summary

This work addresses the scalability limitations of centralized path planning and the suboptimal solution quality and load imbalance often exhibited by decentralized approaches in large-scale multi-agent systems. To overcome these challenges, the authors propose a decentralized coordination framework grounded in a Karma mechanism—an inalienable, artificial credit system that records agents’ historical cooperation. Through bilateral negotiations under limited communication, agents resolve conflicts while promoting long-term fairness and efficiency. The approach integrates bilateral replanning, kinodynamic modeling with orientation constraints, and a distributed negotiation protocol, achieving a balance between system-wide efficiency and equitable workload distribution without relying on global priority structures. Evaluated in a continuous warehouse pick-and-delivery scenario, the method significantly reduces service time disparity among robots while maintaining high overall operational throughput.

Abstract

Multi-Agent Path Finding (MAPF) is a fundamental coordination problem in large-scale robotic and cyber-physical systems, where multiple agents must compute conflict-free trajectories with limited computational and communication resources. While centralised optimal solvers provide guarantees on solution optimality, their exponential computational complexity limits scalability to large-scale systems and real-time applicability. Existing decentralised heuristics are faster, but result in suboptimal outcomes and high cost disparities. This paper proposes a decentralised coordination framework for cooperative MAPF based on Karma mechanisms - artificial, non-tradeable credits that account for agents' past cooperative behaviour and regulate future conflict resolution decisions. The approach formulates conflict resolution as a bilateral negotiation process that enables agents to resolve conflicts through pairwise replanning while promoting long-term fairness under limited communication and without global priority structures. The mechanism is evaluated in a lifelong robotic warehouse multi-agent pickup-and-delivery scenario with kinematic orientation constraints. The results highlight that the Karma mechanism balances replanning effort across agents, reducing disparity in service times without sacrificing overall efficiency. Code: https://github.com/DerKevinRiehl/karma_dmapf

Problem

• Multi-Agent Path Finding
• Decentralised Coordination
• Scalability
• Fairness
• Conflict Resolution

Innovation

• Karma mechanisms
• Decentralised MAPF
• Cooperative multi-agent systems
• Bilateral negotiation
• Fairness in path planning