Fully distributed event-triggered control for multi-robot systems based on modal space framework

In tasks that require improved mechanical strength and load-bearing capacity, such as the handling of heavy or large-volume objects, multi-robot collaborative control is of utmost importance. In this paper, a novel control framework is introduced for multi-robot cooperation, aiming to address the challenges presented by dynamic coupling, anisotropy, the lack of velocity information, and the significant network transmission load within large-scale robot cooperation systems. This framework draws upon insights from both vibration theory and control theory. Firstly, a novel decoupling modal space is presented for multi-robots to complete a collaborative task, which means each control channel is not affected by the coupling of other control channels. Then, a distributed filter is designed for systems to avoid the use of velocity measurement information, which ensures that the output feedback control of multiple robots is realized and makes the estimation error converge to zero uniformly, exponentially and globally. Moreover, distributed adaptive event-triggered protocols are developed that are independent of network scale and do not rely on global information. In this study, the controller and communication do not need to be continuously updated. Finally, experimental demonstrations are provided to show the effectiveness of the proposed control algorithms.