Decentralized Uncertainty-Aware Active Search with a Team of Aerial Robots

2024-10-11

arXiv.org

AI Summary

Post-disaster communication outages, large-scale search areas, and unknown locations/numbers of survivors pose significant challenges for efficient search-and-rescue operations. Method: This paper proposes a decentralized multi-UAV active search system integrating uncertainty modeling with an approximate decentralized partially observable Markov decision process (POMDP) solver. It features dual-mode trajectory planning: (i) during communication outages, trajectories prioritize rapid area coverage and probabilistic re-visitation of high-uncertainty regions via stochastic exploration; (ii) upon communication restoration, distributed state estimation and collaborative planning enable information fusion and online task re-optimization. Contribution/Results: The system is the first to achieve hardware validation in real-world outdoor environments. Experiments demonstrate significantly higher search efficiency than greedy coverage under communication outages; comparable performance under nominal communication; and robust localization of multiple, heterogeneous, and数量-unknown targets.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Planning, Routing, and Scheduling: Planning under Uncertainty
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Decentralized Web and Fediverse systems

Abstract

Rapid search and rescue is critical to maximizing survival rates following natural disasters. However, these efforts are challenged by the need to search large disaster zones, lack of reliability in the communications infrastructure, and a priori unknown numbers of objects of interest (OOIs), such as injured survivors. Aerial robots are increasingly being deployed for search and rescue due to their high mobility, but there remains a gap in deploying multi-robot autonomous aerial systems for methodical search of large environments. Prior works have relied on preprogrammed paths from human operators or are evaluated only in simulation. We bridge these gaps in the state of the art by developing and demonstrating a decentralized active search system, which biases its trajectories to take additional views of uncertain OOIs. The methodology leverages stochasticity for rapid coverage in communication denied scenarios. When communications are available, robots share poses, goals, and OOI information to accelerate the rate of search. Extensive simulations and hardware experiments in Bloomingdale, OH, are conducted to validate the approach. The results demonstrate the active search approach outperforms greedy coverage-based planning in communication-denied scenarios while maintaining comparable performance in communication-enabled scenarios.

Problem

• Decentralized active search for unknown objects in large disaster zones
• Autonomous aerial robots operating in communication-denied environments
• Fusing multi-robot detections to localize objects with high accuracy

Innovation

• Decentralized active search with aerial robots
• Stochastic coverage in communication-denied scenarios
• Multi-robot fusion for OOI detection and localization