Bucket brigade foraging improves upon homogeneous foraging by reducing spatial interference between robots, which occurs when robots are forced to work in the same space. Robots must spend time avoiding one another instead of carrying out useful work. Bucket brigade foraging algorithms restrict the motion of each robot to at most some fixed distance from its starting location. We reproduce the performance of known bucket brigade foragers, and then present a new controller in which robots adapt the size of their foraging area in response to interference with other robots, improving overall performance. This approach also has the potential to cope with nonuniform resource distributions.

Interference reduction through task

Swarm techniques, where many simple robots are used instead of complex ones for performing a task, promise to reduce the cost of developing robot teams for many application domains. The challenge lies in selecting an appropriate control strategy for the individual units. This work explores the effect of different control strategies of varying complexity and of various environmental factors on performance of a team of robots at a foraging task when using physical robots (the Minnesota Distributed Autonomous Robotic Team). Specifically we study the effect of localization and of simple communication techniques on task completion time using two sets of foraging experiments. We also present results for task performance with varying team sizes and target distribution. As indicated by the results, control strategies with increasing complexity reduce the variance in the performance, but do not always reduce the time to complete the task. 1

Recherche et développement pour la défense Canada

1.1 Approach and outcomes............................... 5

All rights reserved. However, in accordance with the Copyright Act of Canada, this work may be reproduced, without authorization, under the conditions for Fair Dealing. Therefore, limited reproduction of this work for the purposes of private study, research, criticism, review, and news reporting is likely to be in accordance with the law, particularly if cited appropriately. ii

We present a supervised learning from demonstration system capable of training stateful and recurrent behaviors, both in the single agent and multiagent case. Furthermore, behavior complexity due to statefulness and multiple agents can result in a high dimensional learning space, which can require many samples to learn properly. Our approach, which relies heavily on both per-agent behavior decomposition and structuring agents into a tree hierarchy, can significantly reduce the number of samples and make such training feasible. We demonstrate our system in a simulated collective foraging task where all the agents execute the same behavior set. We also discuss how to extend our approach to a heterogeneous case, where different subgroups of agents perform different behaviors. 1