Research Papers

Swarm Coordination Under Conflict and Use of Enhanced Lyapunov Control

[+] Author and Article Information
Daniel A. Sierra

Biomedical Engineering Program, University of Connecticut, Storrs, CT 06268; Universidad Industrial de Santander, Bucaramanga 680002, Colombiadasierra@uis.edu.co

Paul McCullough

Department of Mechanical Engineering, University of Connecticut, Storrs, CT 06268paul.t.mccullough@gmail.com

Nejat Olgac1

Department of Mechanical Engineering, University of Connecticut, Storrs, CT 06268olgac@engr.uconn.edu

Eldridge Adams

Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT 06268eldridge.adams@uconn.edu


Corresponding author.

J. Dyn. Sys., Meas., Control 133(2), 021004 (Feb 11, 2011) (8 pages) doi:10.1115/1.4003213 History: Received February 25, 2009; Revised September 11, 2010; Published February 11, 2011; Online February 11, 2011

We consider hostile conflicts between two multi-agent swarms. First, we investigate the complex nature of a single pursuer attempting to intercept a single evader (1P-1E), and establish some rudimentary rules of engagement. The stability repercussions of these rules are investigated using a Lyapunov-based stability analysis. Second, we extend the modeling and stability analysis to interactions between multi-agent swarms of pursuers and evaders. The present document considers only swarms with equal membership strengths for simplicity. This effort is based on a set of suggested momenta deployed on individual agents. The control of group pursuit is divided into two phases: the approach phase during which the two swarms act like individuals in the 1P-1E interaction, and the assigned pursuit phase, where each pursuer follows an assigned evader. A simple, single-step dissipative control strategy, which results in undesirable control chatter, is considered first. A distributed control logic is then introduced, in order to ameliorate the chatter problems. In this new logic, the dissipative control action is spread out over a time window. A wide range of case studies is tested in order to quantify the parametric effects of the new strategy.

Copyright © 2011 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Figure 1

The momentum profiles as in Eq. 1 for bep=6, cep=0.5, bpe=10, cpe=1

Grahic Jump Location
Figure 2

Capture time as a function of the initial distance between agents ds=‖y(t=0)‖ for δep=1.67 and δpe=2.37

Grahic Jump Location
Figure 3

Sample momentum profiles for interaction between like members, for app=0.1, bpp=30, cpp=5, aee=0.1, bee=30, cee=10. Note that d=‖u‖ and d=‖v‖, for gpp(.) and gee(.), respectively.

Grahic Jump Location
Figure 4

Function trans(d), from Eq. 8, d0=3

Grahic Jump Location
Figure 5

Phase 1 analysis for the parameter set in Table 1. δEP=12.25 and δPE=13.42. Left: net moment on the swarm and right: approach time tapp corresponding to the duration of phase 1. Note that d=‖y¯‖ and ds=‖y¯(t=0)‖.

Grahic Jump Location
Figure 6

Windowed damping strategy

Grahic Jump Location
Figure 7

Depiction of the windowed technique with update in damping

Grahic Jump Location
Figure 8

Damping control momenta on pursuer x2 for L=1 and L=6

Grahic Jump Location
Figure 9

(A) The Lyapunov function variations for L=1, L=6, and no damping and (B) the corresponding violation flags




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In