Inspection of infrastructure using static sensor nodes has become a well established approach in recent decades. In this work, we present an experimental setup to study a binary inspection task using vibration-sensing mobile sensor nodes. The objective is to identify the predominant tile type in a \(1\text { m} \times 1\text { m}\) flat tiled surface composed of vibrating and non-vibrating tiles. A swarm of miniaturized robots performs the inspection. Each robot is equipped with an onboard IMU for vibration sensing and infrared sensors for collision avoidance. We adopt an existing decision-making approach from the literature that leverages a Bayesian algorithm for updating robots’ belief based on their observations and supports two information sharing strategies. Building on this, we introduce a new information sharing strategy designed to accelerate the decision-making. To optimize the algorithm parameters, we develop a simulation framework calibrated to our real-world setup in the high-fidelity Webots robotic simulator. We evaluate all three information sharing strategies through simulation (100 trials) and real-world (15 trials) experiments. Moreover, we test the effectiveness of our optimization by evaluating swarms with optimized and non-optimized parameters in increasingly complex environments that differ from scenarios seen during optimization. Results show that our proposed information sharing strategy consistently reduces decision time compared to the two baseline strategies. Additionally, optimized parameters yield robust performance across different environments, while non-optimized parameters perform adequately in simple scenarios but degrade in complex settings.
