2021/05/13 by Ki Myung Brian Lee, Felix H. Kong, Lee, Ki Myung Brian +11 · 1 citation
Computer Science · #Distributed Control Multi-Agent Systems #FOS: Computer and information sciences #Optimization and Search Problems #Robotic Path Planning Algorithms #Robotics (cs.RO)
paper · pdf · doi:10.48550/arxiv.2105.06118
openalex publication_date 2021/05/13 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Heterogeneous multi-robot systems are advantageous for operations in unknown\nenvironments because functionally specialised robots can gather environmental\ninformation, while others perform tasks. We define this decomposition as the\nscout-task robot architecture and show how it avoids the need to explicitly\nbalance exploration and exploitation~by permitting the system to do both\nsimultaneously. The challenge is to guide exploration in a way that improves\noverall performance for time-limited tasks. We derive a novel upper confidence\nbound for simultaneous exploration and exploitation based on mutual information\nand present a general solution for scout-task coordination using decentralised\nMonte Carlo tree search. We evaluate the performance of our algorithms in a\nmulti-drone surveillance scenario in which scout robots are equipped with\nlow-resolution, long-range sensors and task robots capture detailed information\nusing short-range sensors. The results address a new class of coordination\nproblem for heterogeneous teams that has many practical applications.\n