Less is More: Robust Zero-Communication 3D Pursuit-Evasion via Representational Parsimony

2026-03-09

AI Summary

This work addresses the significant challenge of achieving robust communication-free 3D pursuit-evasion under communication constraints, partial observability, and nonholonomic dynamics. The authors propose a representational parsimony principle that explicitly eliminates redundant inter-agent channels to suppress error propagation and enhance deployment robustness. They design a decoupled, compact observation interface (50-dimensional) and a contribution-gated credit assignment (CGCA) mechanism, forming a path-guided decentralized pursuit framework. Evaluated on a 4-vs-1 pursuit-evasion task, the approach achieves a success rate of 0.753 with only a 0.223 collision rate—outperforming a full-observation baseline—and demonstrates strong generalization in stress tests and zero-shot transfer scenarios.

Technology Category

• Multiagent Systems: Adversarial Agents
• Reasoning under Uncertainty: Sequential Decision Making
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Asymmetric 3D pursuit-evasion in cluttered voxel environments is difficult under communication latency, partial observability, and nonholonomic maneuver limits. While many MARL methods rely on richer inter-agent coupling or centralized signals, these dependencies can become fragility sources when communication is delayed or noisy. Building on an inherited path-guided decentralized pursuit scaffold, we study a robustness-oriented question: can representational parsimony improve communication-free coordination? We instantiate this principle with (i) a parsimonious actor observation interface that removes team-coupled channels (83-D to 50-D), and (ii) Contribution-Gated Credit Assignment (CGCA), a locality-aware credit structure for communication-denied cooperation. In Stage-5 evaluation (4 pursuers vs. 1 evader), our configuration reaches 0.753 +/- 0.091 success and 0.223 +/- 0.066 collision, outperforming the 83-D FULL OBS counterpart (0.721 +/- 0.071, 0.253 +/- 0.089). It further shows graceful degradation under speed/yaw/noise/delay stress tests and resilient zero-shot transfer on urban-canyon maps (about 61% success at density 0.24). These results support a practical paradigm shift: explicitly severing redundant cross-agent channels can suppress compounding error cascades and improve robustness in latency-prone deployment.

Problem

• zero-communication
• 3D pursuit-evasion
• partial observability
• nonholonomic constraints
• multi-agent robustness

Innovation

• representational parsimony
• zero-communication coordination
• Contribution-Gated Credit Assignment
• decentralized MARL
• robust pursuit-evasion