Brain-computer interface for human-multirobot strategic consensus with a differential world model
- PDF / 5,609,876 Bytes
- 19 Pages / 595.224 x 790.955 pts Page_size
- 46 Downloads / 150 Views
Brain-computer interface for human-multirobot strategic consensus with a differential world model Yaru Liu1 · Wei Dai1 · Huimin Lu1 · Yadong Liu1 · Zongtan Zhou1 Accepted: 18 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In a distributed multi-robot system, the world model maintained by each robot is inconsistent due to measurement errors from onboard sensors, which will produce different and even incorrect strategies. In this paper, we propose an advanced interaction approach for human-multirobot strategic consensus. First, an opinion dynamics model is used to find the consistent multirobot strategy, which is not necessarily the best choice due to the inaccurate world model. When the human receives the strategy from the robots, he/she can accept or reject it and reselect the strategy via a brain-computer interface (BCI). Of course, human judgment may be incorrect, and the BCI has false detections. Thus, the robots do not directly accept the human strategy but add it to the opinion dynamics model as a new node and recalculate the final consistent strategy. In addition, we developed a custom-designed simulation system based on the Robot Operating System and Gazebo to realize and evaluate the human-multirobot interaction. The extensive simulation results show that the proposed approach can significantly improve the correct rate of strategy selection compared with robot-only or human-only control, as well as the traditional human-robot interaction methods and other strategic consensus models. Keywords Human-multirobot interaction · Brain-computer interface · Opinion dynamics model · Distributed multi-robot system
1 Introduction With the expansion of the execution environment and increasing task complexity, the multi-robot system (MRS) has been widely used because of its enormous advantages compared to the single-robot system (SRS) [18, 43]. First, the MRS is a parallel system, which has higher execution efficiency when performing tasks. Second, through cooperation and communication, the MRS can become a quite capable mobile robot system. Finally, the fault tolerance of a distributed MRS is significantly stronger. When one robot in this system becomes trapped, Yaru Liu and Wei Dai contribute equally to this work. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10489-020-01963-2) contains supplementary material, which is available to authorized users. Zongtan Zhou
[email protected] 1
College of Intelligence Science and Technology, National University of Defense Technology, Changsha, Hunan, China
the remaining robots will replace its role and continue to complete the task [7]. However, there are measurement errors in sensors, and communication delays are inevitable among robots. The strategy obtained by MRS is not absolutely correct, so human participation is required. Researchers have recently shown an increasing interest in another aspect of robot control: human-robot interaction. While there are well-established interfaces
Data Loading...
