Rezende Adriano M C, Gonçalves Vinicius M, Pimenta Luciano C A
Graduate Program in Electrical Engineering, Universidade Federal de Minas Gerais, Pampulha, Belo Horizonte, MG, Brazil.
Graduate Program in Electrical Engineering, Universidade Federal de Minas Gerais, Pampulha, Belo Horizonte, MG, Brazil.
ISA Trans. 2021 Mar;109:126-140. doi: 10.1016/j.isatra.2020.09.019. Epub 2020 Oct 7.
This paper presents a MILP (mixed integer linear programming) based formulation for the coordination of multiple robots. We consider robots that must follow closed intersecting paths persistently. We propose an off-line planning of velocity profiles preventing the need of online collision avoidance maneuvers or path replanning. Our robot model considers minimum and maximum speed constraints, which allows our strategy to be applied to fixed-wing aerial robots. We also deal with three important other issues which are usually disregarded in the related literature: minimum spatial separation; acceleration limits; and uncertainties on the speeds and positions. Simulations with up to 48 robots show the efficiency of our strategy. A real experiment with 3 actual e-puck robots is presented in order to demonstrate the robustness of our formulation in a real world scenario.
本文提出了一种基于混合整数线性规划（MILP）的多机器人协调方法。我们考虑的机器人必须持续遵循封闭的相交路径。我们提出了一种速度剖面的离线规划方法，避免了在线碰撞避免操作或路径重新规划的需要。我们的机器人模型考虑了最小和最大速度约束，这使得我们的策略能够应用于固定翼空中机器人。我们还处理了其他三个在相关文献中通常被忽视的重要问题：最小空间间隔；加速度限制；以及速度和位置的不确定性。多达48个机器人的仿真结果表明了我们策略的有效性。为了证明我们的方法在实际场景中的鲁棒性，我们进行了3个实际电子鼠机器人的真实实验。
