Decentralized floating object transportation using a swarm of autonomous surface vehicles

In this work, a multi-robot system maneuvers floating objects through the so-called caging strategy, where the target objects are non-self-propelled and have different shapes and inertia. To achieve this, a swarm of non-holonomic autonomous surface vehicles acts as pusher-boat caging and pushing objects of various shapes without relying on specific information about their shape, inertia, or initial orientation. Then, the floating object is maneuvered to a desired position and orientation. Numerical simulations use a nonlinear maneuvering model for all the bodies, while cooperative object transportation uses a homogeneous multi-robot system with direct communication and decentralized control. Tasks use target, repulsion, propulsion, and object transportation algorithms. The navigation of the swarmed object uses a proportional integral derivative (PID) controller and waypoint navigation. Finally, the researchers carried out numerical simulations to evaluate the performance and verify the proposed strategy. The results show that during the object transportation, the different-shaped objects were successfully caged, handled, and transported, demonstrating the robustness of this approach was successful in all cases following the proposed rules. In addition, simulations showed the effectiveness of the proposed method, demonstrating transportation without the use of any extra mechanism, only through contact forces.