Distributed time-critical coordination strategies for unmanned aerial systems in cluttered environments

Puig Navarro, Javier

Permalink

https://hdl.handle.net/2142/112988 Copy

Description

Title

Distributed time-critical coordination strategies for unmanned aerial systems in cluttered environments

Author(s)

Puig Navarro, Javier

Issue Date

2021-07-07

Director of Research (if dissertation) or Advisor (if thesis)

Hovakimyan, Naira

Doctoral Committee Chair(s)

Hovakimyan, Naira

Committee Member(s)

• Liberzon, Daniel
• Salapaka, Srinivasa
• Langbort, Cedric
• Allen, B. Danette

Department of Study

Aerospace Engineering

Discipline

Aerospace Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Unmanned aerial systems
• UAV
• UAS
• UxS
• distributed coordination
• cluttered environments
• time critical
• safety critical systems
• consensus
• trajectory generation
• uncertainty
• uncertain

Abstract

This thesis addresses the problem of cooperative motion planning and control for a group of cooperating unmanned aerial systems through cluttered and uncertain environments, subject to a broad range of coordination and temporal constraints. The proposed solution expands the type of time-critical missions that can be automated using cooperative motion control frameworks. This work introduces the use of novel geometric queries to aid a sample-based motion-planning algorithm guide the growth of a rapidly-exploring random tree through the narrow passages in cluttered and uncertain scenarios. To this effect, specific silhouette and tolerance verification queries are designed for the geometric objects that represent vehicle motion and environmental obstacles. The combination of the silhouette-informed path planner with a CNC-inspired path-smoothing method, and a centralized cooperative speed-assignment algorithm yields a set of C2 continuous trajectories that maintain safe separation with all uncertain obstacles and cooperating peers, meet desired mission constraints, and satisfy a set of simplified dynamic constraints. The vehicles are then tasked to follow their assigned paths and coordinate online to meet mission objectives, desired inter-agent spacing constraints, and temporal constraints—such as a time of arrival or a window of arrival. The thesis introduces two types of inter-agent spacing constraints—tight and loose coordination—and three types of temporal constraints—unenforced, relaxed, and strict—that result in six general time-critical coordination strategies. This thesis presents six distributed coordination protocols to enforce this range of constraints. These coordination protocols rely on a lossy communication network that can be disconnected pointwise in time at all times, but is connected in an integral sense over a sliding temporal window. This work derives transient and steady-state performance bounds for the tight coordination protocols. Simulation results through a cluttered urban-like environment, where vehicles are subject to wind disturbances, corroborate the theoretical results.

Graduation Semester

2021-08

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/112988

Copyright and License Information

2021 by Javier Puig Navarro. All rights reserved.

Owning Collections