Achieving Human-Inspired Drift Diffusion Consensus in Swarm Robotics

We present a human inspired approach to collective decision-making in swarm robotics, leveraging a social drift diffusion model that models the decision making process of group of humans. We adapt its principles to robotic swarms to address collective perception tasks. Our method introduces the social factor parameter that allows direct control over the trade-off between decision speed and accuracy. It enables robotic swarms to reach consensus on environmental characteristics more efficiently, with the possibility to prioritize either speed or accuracy, depending on the task requirements. Experimental simulations across various environmental complexities demonstrate our method’s superior performance compared to traditional algorithms like the voter model and majority rule. The results highlight the effectiveness of human-inspired decision-making mechanisms in enhancing the capabilities of swarm robotics.