Evolution of Solitary and Group Transport Behaviors for Autonomous Robots Capable of Self-Assembling

by Roderich Groß and Marco Dorigo

Abstract

Group transport is being performed in many natural systems and has become a canonical task for studying cooperation in robotics. We simulate a system of simple, insect-like robots that can move autonomously and grasp objects as well as each other. We use artificial evolution to produce solitary transport and group transport behaviors. We show that robots, even though not aware of each other, can display effective group transport behaviors. Thereby, they can benefit from behaving differently from robots engaged in solitary transport. The best group transport behaviors yielded by half of the evolutions let the robots organize into self-assembled structures. This provides evidence that self-assembly can provide adaptive value to individuals that compete in an artificial evolution based on task performance. We conclude the paper by discussing potential implications for evolutionary biology and robotics.

evolution of group transport

Video Recordings (copyrights are with Roderich Groß)

Group transport of a heavy prey (1250g) by 5 robots:
  1. self-assembling behavior - evolved neural network controller that lets robots self-assemble:
    Video 1 (2.3 MB) (robots capable of group transport)
    Video 2 (6.1 MB) (interesting behavior)
  2. non-self-assembling behavior - evolved neural network controller that does not let robots self-assemble:
    Video 1 (2.2 MB) (robots incapable of group transport)

References

  1. Groß R. and Dorigo M., Evolution of Solitary and Group Transport Behaviors for Autonomous Robots Capable of Self-Assembling, Adaptive Behavior (accepted for publication)
  2. Groß R. and Dorigo M. Evolving a Cooperative Transport Behavior for Two Simple Robots, In Liardet P., Collet P., Fonlupt C., Lutton E., and Schoenauer M., editors, Artificial Evolution - 6th International Conference, EA 2003, Revised Selected Papers, volume 2936 of Lecture Notes in Computer Science, Springer Verlag, Berlin, Germany (2004) 305-317

© 2003-2007 Roderich Groß