Enhancing Border Security using Heterogeneous UV Swarm with AI and Graph Theory

With increasing border emergencies, the application of novel border patrol and surveillance techniques is crucial. One of the prosperous novel techniques for border surveillance is the application of Artificial Intelligence (AI) and swam robotics. The authors proposed ?DUST: an ultra-large scale miniature UV-based AI platform for border protection? in 2021-SLA (grant number: 21STSLA00009-01-00). This proposal introduces a similar idea by extending the surveillance domain from the ground border to the ground, air, water surface, and under-the-water border. Therefore, one introduces an AI platform for border security that uses a Heterogeneous Unmanned Vehicle Swarm (HUVS) to protect and monitor the border on the ground, air, water surface, and below the water. The proposed HUVS is equipped with different kinds of unmanned vehicles, including Unmanned Aerial Vehicles (UAVs) with different levels of sophistication, Unmanned Ground Vehicles (UGVs), Unmanned Water Surface Vehicles (UWSVs), and Unmanned Sub-surface Vehicles (USVs). While the backbone of our proposal is using different types of robots to secure borders, the research heavily relies on AI (supervised learning and reinforcement learning), machine vision, control engineering, sensory data fusion, and cybersecurity. Also, for swarming, we will concentrate on graph theory to generate a centralized, decentralized, and hybrid communication protocol within the HUVS.