Adaptive Foraging for Robotic Swarms

Several epuck robots during the experiments conducted at BRL (Bristol Robotics Lab).

Several epuck robots during the experiments conducted at BRL (Bristol Robotics Lab).

The aim of this work is to design, simulate, and implement, biological-inspired controllers in order to achieve efficient and dynamic task allocation mechanisms for multi-robot teams. These algorithms have the purpose of making a swarm of robots being able to adapt to unknown situations and abrupt changes within the environment.

In the future, swarm robotic systems will be an integral part of new transportation, logistic and even agricultural services. However, developing self-organized swarm robotics systems capable of adapting to environmental changes is a complex challenge. This project, developed an efficient labor division model, with the ability to regulate the distribution of work among swarm robots. This work extends the popular models in the literature and proposes a new Adaptive Response Threshold Model (ARTM). Experiments were carried out in simulation and in real-robot scenarios with the aim of studying the performance of this new adaptive model.  Several of the real-robot experiments can be seen in the following video:


 

Results presented in this project verified that the extended approach improves on the adaptability of previous systems. For example, by reducing collision duration among robots in foraging missions, our approach helps small swarms of robots to adapt more eefficiently to changing environments, thus increasing their self-sustainability. The following conference and journal papers were publish during the project: