Robust Coordination Strategies for Heterogeneous UAS in Limited Communication Environments

Abstract

During the last three years, we further developed the time-critical cooperative control framework ,initiated under our prior AFOSR grants, to enable novel distributed trajectory-planning and online coordination capabilities for a fleet of cooperating vehicles that communicate over a lossy wireless network. In the field of trajectory planning, the main advances of this research include a distributed algorithm that balances the trajectory-planning workload among all agents in the fleet. The resulting algorithms combine tools from non-linear optimization and distributed programming methods, and leverage the polynomial structure of the trajectories and differential flatness properties of the system dynamics to compute the entire subdifferential analytically. Hence, reducing the computational cost as compared to existing bundle methods. Regarding time-critical coordination, the temporal specification and coordination capabilities of the framework have broadened significantly to accommodate the diversity of constraints required in realistic missions of interest to AFOSR. The distributed time-critical coordination law lets the vehicles adjust their speed in real time to meet the desired temporal and coordination constraints even in the presence of external disturbances. This research introduced a new classification of these constraints that has lead to unprecedented levels of flexibility and adaptability. Moreover, tools from Lyapunov stability, switched systems, and algebraic graph theory were leveraged to derive transient and steady-state performance bounds for some of the new time-critical coordination strategies. We also initiated new directions of research over the last two years on certified trajectory tracking for nonlinear and uncertain systems on a single-agent level. The certificates are amenable to incorporation within the results as mentioned earlier.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 06, 2021
Accession Number
AD1230403

Entities

People

  • Naira Hovakimyan

Organizations

  • University of Illinois Urbana–Champaign

Tags

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Robotics and Automation.