Cooperative collision avoidance control and coordination for multiagent Lagrangian systems with disturbances

Deka, Shankar Anand

Permalink

https://hdl.handle.net/2142/90674 Copy

Description

Title

Cooperative collision avoidance control and coordination for multiagent Lagrangian systems with disturbances

Author(s)

Deka, Shankar Anand

Issue Date

2016-04-28

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

Stipanović, Dušan M.

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Mutliagent systems
• Collision avoidance
• Coordination

Abstract

Multi-agent systems like a network of autonomous robots, have tremendous potential in many military and civilian applications. But, even viewed as a pure academic problem, designing controllers for such complex systems is a matter of much interest. Controller design for multi-agent system might focus on achieving several objectives, such as formation control, coverage control, consensus, target capture, pursuit evasion etc., while all at the same time aiming to be optimal in some sense, or following certain constraints imposed by the environment or communication limitations. Whatever is the objective, we always want to have a safety guarantee for the agents; the agents should avoid collisions with themselves and any static obstacles, while performing an objective. This thesis studies one such controller, which guarantees collision avoidance among the agents, in presence of bounded disturbances, while the agents carry out a coordination objective. The agents are assumed to follow a Lagrangian dynamics. The collision avoidance controller takes up the second part of the thesis. In the first part of this thesis, a particular Lagrangian system, the Raven II surgical robot, is studied in with the aim of highlighting the process of modelling and identifying such system. This is done for two reasons. One because Lagrangian dynamics is commonly used to model the agents in a multi-agent system. And second reason that motivates the modelling Raven II in part I, is to aid in future research direction pertaining to the control of Raven II.

Graduation Semester

2016-05

Type of Resource

text

Permalink

http://hdl.handle.net/2142/90674

Copyright and License Information

Copyright 2016 Shankar Anand Deka

Owning Collections