Optimizing Multi-Agent Search with Non-Uniform Sensor Effectiveness in Distributed Quadcopter Systems.

This article explores the vital role of distributed multi-robot systems (DMRS) in applications such as search and rescue, surveillance, and military operations. In particular, we focus on developing a method for multi-agent search using a quadcopter unmanned aerial vehicle (UAV) equipped with a downward-facing camera. Unlike existing studies, our model includes a unique model for searching for the best camera and achieving maximum performance in the scene. We introduce an uncertainty distribution that reflects the lack of information to capture the uncertainty in the search space. Using the concept of the hub Voronoi configuration, our approach optimizes the deployment of the quadcopter to reduce confusion. The distribution and detection process continues until the average uncertainty reaches a threshold, which means the detection target is successful and reliable. We present an in-depth study of the different parameters in the search for a good camera and propose a test setup for the model's performance. The multiple quadcopter search strategy was implemented and simulated using ROS/Gazebo and Matlab allowed its performance on various parameters to be verified in real experiments. Simulation results demonstrate the effectiveness of this strategy and provide insight into the impact of the study on aspects such as camera performance and number of detection quadcopters. The simulation platform we have created is an important tool for further testing and benchmarking optimization in real life. This research helps to improve the understanding of multi-sensory search strategies, especially when the sensor search efficiency is unequal.