This paper presents an approach to the dynamic leader selection problem in autonomous non-holonomic mobile robot formations when the current leader enters a failure state. Our method is based on a tree structure coupled with a modified version of the Nonlinear Model Predictive Control (NMPC) that allows for behavior change at the controller level. An explanation of the control algorithm, behavior selection, and leader selection structure is given, after which the results of both simulations and experiments using a three robot formation are shown and discussed.

中文翻译：

---

基于非线性模型预测控制的多机器人系统动态领导者分配

本文提出了一种解决当当前领导者进入故障状态时自主非完整移动机器人编队中的动态领导者选择问题的方法。我们的方法基于树形结构，并结合了非线性模型预测控制（NMPC）的改进版本，该模型允许在控制器级别更改行为。给出了控制算法，行为选择和领导者选择结构的解释，然后显示并讨论了使用三个机器人阵型的仿真和实验结果。