Behavior-driven Indoor Multi-Target UAV-based Detection System: Intelligent RL-based Approach.

Advancements in UAV technology have revolutionized various industries, enabling their widespread deployment in embedded systems, autonomy, control, security, and communication. Autonomous UAVs distinguish themselves by their ability to make informed decisions by anticipating future scenarios and drawing on past experiences. Our primary focus is on analyzing a monitoring system consisting of a mission area, an autonomous UAV, a charging station, and multiple dynamic targets. The mission area is divided into distinct zones, and a time-slotted scheme facilitates UAV movement and recharging. Specifically, this paper presents a multitarget detection system based on reinforcement learning (RL)-based behavior-driven UAVs designed for indoor environments. Simulation results demonstrate that our proposed RL-based detection system significantly outperforms reference systems in terms of detection rate and convergence.