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

Jialin Ying; Zhihao Li; Zicheng Dong; Guohua Wu; Yihuan Liao

cs.ROMar 9, 2026

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

Authors: Jialin Ying, Zhihao Li, Zicheng Dong, Guohua Wu, Yihuan Liao

Organizations: Department of Automation, Central South University, Changsha, China

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.

Figures & tables

Item	Value
World/grid	$311.0\times 311.0\times 110.5$ m, $52\times 52\times 18$ voxels
Team setup	4 pursuers vs. 1 evader
Capture radius (Stage 5)	8.0 m
Visibility range (Stage 5)	60.0 m
Pursuer bounds	$\|v_{x}\|\leq 8$ , $\|v_{y}\|\leq 4$ , $\|v_{z}\|\leq 3$ m/s
Pursuer yaw cap	$\|\dot{\psi}\|\leq 0.8$ rad/s (swept down to 0.2)

TABLE I: Stage-5 environment and asymmetry constants.

Fig. 1: Training reward curves under Stage 5. CTDE MAPPO shows marked instability after the visibility-gated transition, whereas OURS-LITE converges to the highest stable return regime.

Fig. 2: Inherited path-guided decentralized pipeline [ 1 ] . Our change is at the actor interface: 83-D full observation is masked to 50-D by removing teammate/slot/encirclement channels for zero-communication robustness.

Block	83-D	50-D
LiDAR + target + self + IMU	41	41
Teammate top- $k$	24	0
Guidance + ID + mode + delay	9	9
Slot + encirclement	9	0
Total	83	50

TABLE II: Observation-profile reduction for representational parsimony.

Fig. 3: Contribution-Gated Credit Assignment. Direction reward is fully active within 40 m, decays over 40–80 m, and vanishes beyond 80 m; capture-share is hard-gated beyond 60 m. This local credit geometry suppresses free-rider equilibria in zero-communication team pursuit.

Method	Success	Clean	Collision	Avg Steps
OURS-LITE	0.753 $\pm$ 0.091	0.752 $\pm$ 0.091	0.223 $\pm$ 0.066	683.6 $\pm$ 269.4
FULL OBS	0.721 $\pm$ 0.071	0.721 $\pm$ 0.072	0.253 $\pm$ 0.089	655.4 $\pm$ 106.7
EUCLIDEAN	0.586 $\pm$ 0.120	0.583 $\pm$ 0.121	0.353 $\pm$ 0.092	887.9 $\pm$ 237.7
LOCAL-NO-GATE	0.569 $\pm$ 0.090	0.568 $\pm$ 0.089	0.389 $\pm$ 0.106	669.1 $\pm$ 223.2
CTDE MAPPO	0.006 $\pm$ 0.004	0.006 $\pm$ 0.004	0.159 $\pm$ 0.111	2806.8 $\pm$ 139.1
APF+PN	0.125 $\pm$ 0.007	0.125 $\pm$ 0.007	0.875 $\pm$ 0.007	124.9 $\pm$ 2.0

TABLE III: Main Stage-5 benchmark (mean

\pm

std over 3 seeds, 500 episodes/seed). OURS-LITE , FULL OBS , and LOCAL-NO-GATE are inherited parameter-shared decentralized PPO-family variants.

Fig. 4: Velocity sweep. OURS-LITE maintains the strongest success–collision trade-off under increasing evader speed, while EUCLIDEAN suffers persistent high-step inefficient pursuits.

Fig. 5: Yaw-rate sweep. The proposed configuration preserves a robust mid-regime margin and shows graceful degradation as maneuver authority shrinks.

Fig. 6: Observation-noise sweep (

\sigma\in[0,0.20]

). OURS-LITE remains strongest across all tested noise levels.

Fig. 7: Action-delay sweep (0–3 steps). OURS-LITE preserves the best absolute robustness profile under latency perturbation.

Fig. 8: 3D Z-axis tactical slicing. The team performs cross-layer vertical sweep: upper-layer pressure constrains vertical escape while lower-layer agents close horizontal corridors, yielding a communication-free topological enclosure.

Panel of Figure 12 — Fig. 9: Zero-shot transfer on five unseen Urban Canyon maps (density 0.08–0.24). Left: map examples from easy to hard density. Right: seed-wise and mean success, collision, average steps, and reward without fine-tuning. OURS-LITE degrades gracefully with density, remaining near 61% success and below 40% collision at density 0.24, consistent with structural transfer rather than fixed-map memorization.

Multi-Agent Systems

Multi-Agent Systems studies how autonomous agents coordinate, compete, communicate, and learn in shared or interdependent tasks.

See all 668 papers

Dec-MARVEL: Decentralized Multi-Agent Exploration without Communication under Budget Constraints
Jul 10, 2026Janghyun Cho, Jimmy Chiun, Guillaume Sartoretti +1Aerial Robotics Multi-Robot Exploration
Dept. of Artificial Intelligence, Sogang University. · Dept. of Mechanical Engineering, College of Design and Engineering, National University of Singapore. · Dept. of Electronic Engineering, Sogang University.
Robust Multi-Agent Pursuit with Preemptive Communication via Reachability-Based Control Barrier Functions
Oct 15, 2026Madison Book, Justin BradleyMulti-Agent Systems Multi-Agent Communication
Department of Computer Science, NC State University, Raleigh, NC, USA
Decentralized End-to-End Multi-AAV Pursuit Using Predictive Spatio-Temporal Observation via Deep Reinforcement Learning
Mar 25, 2026Yude Li, Zhexuan Zhou, Huizhe Li +6Multi-Agent Systems Multi-Robot Coordination
The authors are with the School of Intelligence Science and Engineering, the Guangdong Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics, and the Shenzhen Key Lab for Advanced Motion Control and Modern Automation Equipments, Harbin Institute of Technology, Shenzhen, Guangdong 518055, China

Jul 10, 2026J/K move · Enter open · S save

Dec-MARVEL: Decentralized Multi-Agent Exploration without Communication under Budget Constraints

Janghyun Cho, Jimmy Chiun, Guillaume Sartoretti, Changjoo Nam

Dept. of Artificial Intelligence, Sogang University. · Dept. of Mechanical Engineering, College of Design and Engineering, National University of Singapore. · Dept. of Electronic Engineering, Sogang University.

Aerial Robotics Multi-Robot Exploration Multi-Robot Coordination Decentralized MARL Autonomous Robot Exploration Decentralized Multi-Agent Motion Planning

Open paper AskPDF arXiv

Multi-UAV exploration is often constrained by unreliable communication, limited field-of-view sensing (e.g., lightweight onboard camera), and finite travel budgets that require each robot to reserve enough budget to return to its base. We present Dec-MARVEL, a decentralized, communication-free exploration policy that jointly selects waypoints and sensing headings from each robot's local map, remaining budget, and opportunistic observations of teammates. Across 900 held-out trials spanning three team sizes (2, 4, 8 robots) and three travel budgets (720, 800, and 1024 m) against four baselines, Dec-MARVEL achieves the highest or tied-highest exploration and success rates and the shortest reported travel to 99% merged coverage among successful trials in all nine budget--team configurations. Under our tightest 720 m budget, it reaches 46%, 84%, and 94% success for 2, 4, and 8 robots, versus 17%, 57%, and 85% for the strongest baseline in each case. Physical-robot experiments show that the resulting waypoint-heading plans are executable and remain budget-feasible.

Abstract

Multi-Agent Systems

Explore similar work

Dec-MARVEL: Decentralized Multi-Agent Exploration without Communication under Budget Constraints

Robust Multi-Agent Pursuit with Preemptive Communication via Reachability-Based Control Barrier Functions

Decentralized End-to-End Multi-AAV Pursuit Using Predictive Spatio-Temporal Observation via Deep Reinforcement Learning