IEEE IECON2025 SYPA Winner - Robot Swarm Exploration via Dual Calibration скачать в хорошем качестве

IEEE IECON2025 SYPA Winner - Robot Swarm Exploration via Dual Calibration

This video presentation describes the work in the paper titled: A Dual Calibration Framework for Exploring Environments using Heterogeneous Robot Swarms Authors Yun Gao, Hao Gao, Yiding Ji, Jinni Zhou, and Yang Shi. Yun Gao is with the Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Hong Kong SAR, China, and also with the Robotics and Autonomous Systems Thrust, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China. Hao Gao and Yiding Ji are with the Robotics and Autonomous Systems Thrust, Jinni Zhou is with the Base of Red Bird Master of Philosophy, the Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China. Yang Shi is with the Department of Mechanical Engineering, University of Victoria, Victoria, Canada. Paper abstract Exploring complex environments using heterogeneous robot swarms (RSs) is a considerable challenge in terms of coordination, sensing, and information fusion. Existing approaches suffer from a lack of systematic analysis that fully exploits the complementary capabilities of heterogeneous agents. To bridge this gap, we propose a novel spatial calibration framework that integrates both virtual and physical calibration mechanisms to enable coordinated operation between two distinct robot swarms, RS-A and RS-B. RS-A, characterized by high mobility and a broad field of view, performs continuous, large-scale monitoring and identifies candidate regions of interest. RS-B, equipped with high-precision sensors, is dispatched to these regions to conduct fine-grained data collection and return accurate environmental information, facilitating comprehensive environmental mapping. To this end, we develop a distributed control method for spatial partitioning, position optimization, and information exchange within the swarm, based on improved coverage control and a flooding-based broadcast algorithm for intra-swarm communication. We further design a control architecture that enables inter-swarm collaboration. The proposed framework effectively addresses the limitations of homogeneous RSs in environmental exploration by integrating fast, coarse-grained surveillance with slow, fine-grained investigation through heterogeneous coordination. Finally, the effectiveness of our proposed framework is validated through simulation results.