Fixed-time Synchronization of Networked Uncertain Euler-Lagrange Systems

Yi Dong, Zhiyong Chen

Abstract

This paper considers the fixed-time control problem of a multi-agent system composed of a class of Euler-Lagrange dynamics with parametric uncertainty and a dynamic leader under a directed communication network. A distributed fixed-time observer is first proposed to estimate the desired trajectory and then a fixed-time controller is constructed by transforming uncertain Euler-Lagrange systems into second-order systems and utilizing the backstepping design procedure. The overall design guarantees that the synchronization errors converge to zero in a prescribed time independent of initial conditions. The control design conditions can also be relaxed for a weaker finite-time control requirement.

Introduction

In the coordination of multi-agent systems—such as swarms of unmanned aerial vehicles or collaborative industrial manipulators—synchronization is a fundamental objective. While asymptotic and finite-time control strategies are well-established, they often suffer from convergence rates that are sensitive to initial conditions. This paper presents a robust control framework achieving fixed-time synchronization, ensuring that all agents track a dynamic leader within a bounded time interval $T$ that is independent of the agents' starting positions.

The authors address three primary challenges:

  1. Fixed-time Stability: Convergence within a deterministic time window regardless of initial state.
  2. Parametric Uncertainty: Robustness against unknown or time-varying physical parameters (e.g., mass, friction) within the Euler-Lagrange dynamics.
  3. Directed Communication: Stability guarantees over one-way communication links, which are more restrictive than undirected networks.

Methodology

The proposed solution utilizes a two-stage "Observer-Controller" architecture.

Distributed Fixed-time Observer

Since the leader's state is not globally available, each follower agent implements a distributed fixed-time observer. This observer estimates the leader’s trajectory by communicating with neighboring agents. The authors prove that if the network topology contains a spanning tree with the leader as the root, the estimation error vanishes in a fixed time $T_1^*$.

The directed communication network topology used in the simulation, featuring a leader (node 0) and six follower agents (nodes 1-6).

Robust Backstepping Controller

With a reliable estimate of the leader's state, a backstepping controller is designed to drive the tracking error to zero. The Euler-Lagrange systems are transformed into second-order models, and high-gain domination is employed to counteract parametric uncertainties. The controller incorporates non-linear power functions to accelerate convergence across all error scales, ensuring the tracking error reaches zero within a fixed time $T_2^*$.

Results and Discussion

The effectiveness of the proposed method was evaluated using a numerical simulation of six 2-link robotic manipulators. Each robot was assigned distinct, unknown physical parameters within defined bounds. The communication topology followed the directed graph shown in the methodology section.

Synchronization performance showing position errors ($q_i - q_0$) and velocity errors ($\dot{q}_i - \dot{q}_0$) for all six agents.

As shown in the simulation results, the distributed observer converges in approximately 2.1 seconds, followed by full synchronization of positions and velocities by 2.3 seconds. This confirms the theoretical prediction that the total settling time is bounded and independent of initial offsets.

Conclusion

This work provides a rigorous framework for fixed-time synchronization of uncertain robotic systems over directed networks. While the high-gain approach ensures robustness, future work could address practical limitations such as actuator saturation and sensor noise. Additionally, exploring event-triggered communication protocols could enhance the efficiency of the proposed distributed observer.

Summary updated

Abstract

This paper considers the fixed-time control problem of a multi-agent system composed of a class of Euler-Lagrange dynamics with parametric uncertainty and a dynamic leader under a directed communication network. A distributed fixed-time observer is first proposed to estimate the desired trajectory and then a fixed-time controller is constructed by transforming uncertain Euler-Lagrange systems into second-order systems and utilizing the backstepping design procedure. The overall design guarantees that the synchronization errors converge to zero in a prescribed time independent of initial conditions. The control design conditions can also be relaxed for a weaker finite-time control requirement.
BibTeX
@misc{dong2022fixed,
  author = {Yi Dong and Zhiyong Chen},
  title = {{Fixed-time Synchronization of Networked Uncertain Euler-Lagrange Systems}},
  year = {2022},
  eprint = {2202.07866},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2202.07866},
}
Figure