Parallel Self-Assembly for a Multi-USV System on Water Surface With Obstacles

Parallel self-assembly is an efficient approach to accelerate the assembly process for modular robots. However, these approaches cannot accommodate complicated environments with obstacles, which restricts their applications. We in previous work consider the surrounding stationary obstacles and propose a parallel self-assembly planning algorithm. With this algorithm, modular robots can avoid immovable obstacles when performing docking actions, which adapts the parallel self-assembly process to complex scenes. The algorithm was simulated in 25 distinct maps with different obstacle configurations and shows a significantly higher success rate, which is more than
$80\%$
, compared to the existing parallel self-assembly algorithms. For verification in real-world applications, we in this paper develop a multi-agent hardware testbed system. The algorithm is successfully deployed on four omnidirectional unmanned surface vehicles, CuBoats. The navigation strategy that translates the high-level discrete plan to the continuous controller on the CuBoats is presented. The algorithm’s feasibility and flexibility were demonstrated through successful self-assembly experiments on 5 maps with varying obstacle configurations.
Note to Practitioners
—This paper addresses deploying of self-assembly technologies for modular robots in practical environments with obstacles to facilitate overwater construction tasks or collective transportation systems. Stationary obstacles may severely influence the assembly planning and robot routing processes. Moreover, efficient task coordination, robot navigation, and structure formation are required for large-scale assembly tasks. The algorithm in this work allows all participating robots to navigate online and connect simultaneously to promote efficiency. The strategy presented here endows the robots’ assembly with obstacle-avoidance capability in dense environments. This work will interest those pursuing efficient assembly in scenes with surrounding obstacles. Our hardware experiments demonstrate a concept system and verify the real-time performance of the algorithm under limited computing power. The approach introduced here is not applicable to robots with heterogeneous shapes, three-dimensional target structures, or overcrowded environments with too many obstacles.