Predicting Performance in Robotic Search and Tag


In swarm robotics, sometimes more targets result in less tagging due to interference. We investigated several static metrics and their relationship to tagging performance in a swarm robotic cellular automata simulation. The performance metric measures the percentage of targets tagged within a fixed number of steps. The independent metrics measure initial tagging potential, target interference potential, and the robot density. The first metric is the normalized sum of the Manhattan distances between each robot and each target and is meant to capture the potential for tagging early in the simulation that can heavily impact performance. The second metric measures the potential negative impact of target interference on robots, which has the effect of slowing or even trapping robots, distracting them from untagged targets. Finally, the third metric is the global robot density in the world and actually has a positive influence on tagging performance up to the point of saturation. The values of these metrics are taken form the initial world states. We discuss the metrics and simulation and our results show how the three independent metrics in combination predict the dependent metric.

  • Abstract
  • 1. Introduction
  • 2. Related Work
  • 3. Experiment Design
  • 4. Results
  • 5. Future Work
  • 6. Conclusion
  • References
Topics: Robotics

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