In swarm robotic systems, task allocation is a challenging problem aiming to decompose complex tasks into a series of subtasks. We propose a self-organizing method to allocate a swarm of robots to perform a foraging task consisting of sequentially dependent subtasks. The method regulates the proportion of robots to meet the task demands for given tasks. Our proposed method is based on the response threshold model, mapping the intensity of task demands to the probability of responding to candidate tasks depending on the response threshold. Each robot is suitable for all tasks but some robots have higher probability of taking certain tasks and lower probability of taking others. In our task allocation method, each robot updates its response threshold depending on the associated task demand as well as the number of neighbouring robots performing the task. It relies neither on a centralized mechanism nor on information exchange amongst robots. Repetitive and continuous task allocations lead to the desired task distribution at a swarm level. We also analyzed the mathematical convergence of the task distribution among a swarm of robots. We demonstrate that the method is effective and robust for a foraging task under various conditions on the number of robots, the number of tasks and the size of the arena. Our simulation results may support the hypothesis that social insects use a task allocation method to handle the foraging task required for a colony's survival.
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Materials Science(all)