Multi Agent Active Search

Multi-agent active search focuses on developing algorithms that enable multiple autonomous agents to efficiently locate targets within an environment, often under conditions of uncertainty and limited resources. Current research emphasizes decentralized approaches, utilizing techniques like Thompson sampling and Monte Carlo methods to manage exploration-exploitation trade-offs and handle asynchronous communication between agents, while also incorporating factors like terrain awareness, cost optimization, and varying levels of detection and location uncertainty. This field is crucial for applications such as search and rescue, environmental monitoring, and surveillance, offering significant potential for improving the efficiency and robustness of autonomous systems in complex tasks.

Papers