Event-based Reconfiguration Control for Time-varying Formation of Robot Swarms in Narrow Spaces

TL;DR

The paper tackles time-varying formation control for robot swarms navigating narrow environments by introducing a decentralized event-based reconfiguration controller (ERC) that switches between formation and tailgating modes. It integrates five artificial potential field–based behaviors (formation, tailgating, migration, inter-agent avoidance, obstacle avoidance) and uses environment-width sensing to adapt the formation via a scaling factor $\kappa$, with stability proven through Lyapunov analysis. Empirical results from simulations and software-in-the-loop tests show ERC outperforms fixed-behavior approaches in success rate, speed, travel time, and energy efficiency, while preserving high directional agreement among agents. The approach is scalable, distributed, and supported by open-source code, making it practical for real-world swarm robotics in confined spaces.

Abstract

This study proposes an event-based reconfiguration control to navigate a robot swarm through challenging environments with narrow passages such as valleys, tunnels, and corridors. The robot swarm is modeled as an undirected graph, where each node represents a robot capable of collecting real-time data on the environment and the states of other robots in the formation. This data serves as the input for the controller to provide dynamic adjustments between the desired and straight-line configurations. The controller incorporates a set of behaviors, designed using artificial potential fields, to meet the requirements of goal-oriented motion, formation maintenance, tailgating, and collision avoidance. The stability of the formation control is guaranteed via the Lyapunov theorem. Simulation and comparison results show that the proposed controller not only successfully navigates the robot swarm through narrow spaces but also outperforms other established methods in key metrics including the success rate, heading order, speed, travel time, and energy efficiency. Software-in-the-loop tests have also been conducted to validate the controller's applicability in practical scenarios. The source code of the controller is available at https://github.com/duynamrcv/erc.

Figures (12)

• Figure 1: A robot swarm traveling through a narrow space: (a) Motion paths of the robots using the rigid formation control 736776Vsrhelyi2018, which collide with surrounding obstacles; (b) Motion paths of the robots using our proposed approach, which safely navigate through the narrow space.
• Figure 2: Illustration of a robot in the swarm having a local range sensor with sensing area $S_s$ of radius $r_s$ (solid white circle), alert area $S_a$ of radius $r_a$ (dashed gray circle), and set $\mathcal{M}_i=\{o\}$ of the nearest data point (green dot) to an obstacle.
• Figure 3: The proposed time-varying formation switching between the task and safe configurations to navigate through a narrow space
• Figure 4: The proposed event-based reconfiguration control with two modes and five individual behaviors
• Figure 5: Environment's width estimation

Theorems & Definitions (7)

• Definition 1
• Definition 2
• Remark 1
• Theorem 1
• proof
• Remark 2
• Remark 3