Many kinds of bio-inspired tasks have been tested with swarm robotics and task partitioning is one of the challenging subjects. In nature, it is well known that some colonies of social insects such as honeybees, termites, and ants use task partitioning strategies for their survival. In this paper, we demonstrate an effect of the task partitioning strategy called bucket brigade, which uses the direct transfer of materials or food between a pair of workers. We propose a task partitioning strategy based on the moving speeds of agents for the foraging task. We test various environmental conditions and compare the performance between task partitioning groups and non-partitioning groups. The experimental results show that task partitioning may not always be the best solution for foraging performance. However, when there exists a transfer bottleneck at a central location such as the entrance of the nest, task partitioning can be an effective strategy for reducing the traffic jam and improving the overall foraging performance of a group. The bucket brigade sequenced from the slowest agents (near the food source) to the fastest agents (near the nest) can particularly improve performance significantly in the region with traffic congestion near the nest. Generally, many social insect colonies consist of a number of members, and the entrances of colony nests always suffer from heavy traffic congestion. Our experimental results support the hypothesis that several social insects use one of the task partitioning strategies based on bucket brigades in their foraging tasks.
Bibliographical noteFunding Information:
This work was supported by a National Research Foundation of Korea grant, funded by the Korean government (MEST) (No. 2014R1A2A1A11053839).
All Science Journal Classification (ASJC) codes
- Molecular Medicine
- Engineering (miscellaneous)