A Distributed Dynamic Framework to Allocate Collaborative Tasks Based on Capability Matching in Heterogeneous Multirobot Systems

• Published in: IEEE transactions on cognitive and developmental systems 2024-02, Vol.16 (1), p.251-265
• Main Authors: Lee, Hoi-Yin, Zhou, Peng, Zhang, Bin, Qiu, Liuming, Fan, Bowen, Duan, Anqing, Tang, Jingtao, Lam, Tin Lun, Navarro-Alarcon, David
• Format: Article
• Language: English
• Subjects: Algorithms; Capability modeling; cognitive systems; Collaboration; Hardware; Knowledge representation; Matching; Motion planning; multi-robot systems (MRSs); Multiple robots; resource allocation; Resource management; Resource utilization; Robot kinematics; Robot sensing systems; Robots; task allocation; Task analysis; Task complexity
• ISSN: 2379-8920; 2379-8939
• DOI: 10.1109/TCDS.2023.3264034

Collaboration among a group of robots with heterogeneous capabilities is an important research problem that enables to combine different robot functionalities, and thus, conducts complex tasks that may be difficult to achieve by a single robot with limited resources. In this article, we propose a new distributed task allocation framework based on the capability matching of heterogeneous robots. The framework is composed of an ontological dynamic knowledge graph model and a hardware control scheme to model the capability and optimize resource utilization for collaborative tasks. We introduce an intuitive hardware control scheme based on a dynamic knowledge graph that resolves possible conflicts between the hardware control of different types of robots. Action sequences are produced by a task and motion planning algorithm to collaboratively perform the assigned task. The performance of the proposed methodology is evaluated by both simulations and hardware experiments.