Training Robot Swarms for Adaptive Foraging in Environments with Obstacles

In this work, we train adaptive and efficient foraging strategies for robot swarms in a large, unmapped search space with multiple randomly distributed box obstacles using the penalty-reward based NeuroEvolution of Augmented Topologies (NEAT), P-NeatFA. This model enables efficient multi-robot foraging behavior and obstacle avoidance by rewarding effective actions and penalizing inefficient ones, thereby minimizing redundant exploration and outperforming traditional stochastic foraging algorithms. We optimize foraging strategies and search patterns in robot swarms by training models that maximize cumulative rewards in three types of resource distribution environments. The evaluation focuses on the number of resources collected within a fixed time frame. We compare the performance of the CPFA with the P-NeatFA approach. Experimental results show that our enhanced P-NeatFA strategy significantly outperforms the CPFA. Additionally, we observed differences in the search patterns: robots with the trained foraging behavior tend to travel in straighter, more efficient paths. Robots can avoid random obstacles more efficiently. Robots outperform the CPFA when the robot swarm size increases. Building on this foundation, we can see the potential to develop a distributed, scalable, and secure foraging robot swarm framework by integrating distributed learning.