Robust Self-Reconfiguration for Fault-Tolerant Control of Modular Aerial Robot Systems

TL;DR

This paper tackles fault tolerance in Modular Aerial Robotic Systems by addressing controllability during intermediate self-reconfiguration steps. It introduces a controllability-margin ($CM$) framework for control-constrained dynamics, and develops algorithms to (i) select CM-maximizing target configurations, (ii) construct minimal controllable subassemblies containing faults, and (iii) plan disassembly/assembly sequences that preserve controllability. The approach yields higher practical controllability, fewer reconfiguration steps, and improved trajectory tracking across complete unit failures and rotor degradations, outperforming a baseline method. The findings have practical impact on safer, more reliable fault-tolerant flight of modular aerial swarms, with open-source code available for replication and extension.

Abstract

Modular Aerial Robotic Systems (MARS) consist of multiple drone units assembled into a single, integrated rigid flying platform. With inherent redundancy, MARS can self-reconfigure into different configurations to mitigate rotor or unit failures and maintain stable flight. However, existing works on MARS self-reconfiguration often overlook the practical controllability of intermediate structures formed during the reassembly process, which limits their applicability. In this paper, we address this gap by considering the control-constrained dynamic model of MARS and proposing a robust and efficient self-reconstruction algorithm that maximizes the controllability margin at each intermediate stage. Specifically, we develop algorithms to compute optimal, controllable disassembly and assembly sequences, enabling robust self-reconfiguration. Finally, we validate our method in several challenging fault-tolerant self-reconfiguration scenarios, demonstrating significant improvements in both controllability and trajectory tracking while reducing the number of assembly steps. The videos and source code of this work are available at https://github.com/RuiHuangNUS/MARS-Reconfig/

Figures (8)

• Figure 1: MARS is tasked to track a spiral trajectory with two faulty units. The faulty propellers are marked red. (a) MARS crashed after the complete failure of two units (all rotors are broken). (b) MARS can track trajectories after self-reconfiguration.
• Figure 2: Self-reconfiguration flow after the failure of drone unit 1 in a 3$\times$2 assembly. In Case 1, simply connecting a normal unit adjacent to the complete faulty unit is not controllable, as the torque generated by the gravity and thrust cannot be compensated, resulting in negative CM. In Case 2, the minimum controllable sub-assembly containing the faulty unit is identified to assist in transferring the faulty unit.
• Figure 3: Self-reconfiguration flow after failure of unit No.3 in a 3$\times$2 assembly. Black and purple lines show different disassembly and reassembly choices, and the green line indicating the highest CM case from our algorithm. Full disassembly moves and reconfigures units one by one, while partial disassembly uses sub-assemblies.
• Figure 4: Dynamical simulation of self-reconfiguration processes in CoppeliaSim. First row: (a)--(e), full disassembly approach. Second row: (f)--(h), partial disassembly approach. Additional demonstrations, including rotor failures and alternative configurations, are available at https://github.com/RuiHuangNUS/MARS-Reconfig.
• Figure 5: Yaw angle and IMU-measured accelerations of the faulty unit during the self-reconfiguration (full disassembly approach, corresponding to (a)-(e) in Fig. \ref{['fig:Simulation snapshots of self-reconfiguration']}).

Theorems & Definitions (2)

• Lemma 1: Practical Controllability du2015controllability
• Lemma 2