Effective UAV patrolling for swarm of intruders with heterogeneous behavior

Abstract

The phenomenal growth in the utilization of commercial unmanned aerial vehicles (UAVs) or drones leads to an urgent need for new approaches to ensure safety in the sky. Effective aerial surveillance requires patrolling swarms to react according to the various behaviors demonstrated by intruding swarms, but existing approaches are not practical when dealing with a large number of drones. Specifically, predicting the behaviors or planned paths of the intruding swarms is highly challenging as intruders may perform evasive strategies to avoid detection. Therefore, this work utilizes heuristic search strategies and investigates how various intruder behaviors affect the search performance. To investigate the search performance, a swarm versus swarm simulator is developed. Using the simulator, first, a comparative study is performed to evaluate how intruders’ behaviors can affect the performance of the patrolling swarm. Subsequently, three approaches, including single-objective optimization, multi-objective optimization, and Lévy flight, are compared in terms of their detection performance in a bounded space. The results suggest that multi-objective optimization outperforms both single-objective optimization and Lévy flight-based approaches. Furthermore, our results show that intruders have a lower chance of being tracked when moving in a dense crowd, and this finding reaffirms the schooling behaviors of fish. In a specific simulation scenario, the total percentage of detection is above 90%. However, the detection percentage is highly related to other factors such as search space, number of patrolling UAVs, and the intruders’ behaviors.

Optimal Configuration of Heterogeneous Swarm for Cooperative Detection with Minimum DOP Based on Nested Cones

“…As shown in Figures 33 and 34 Comparing Figure 32 with Figures 33 and 34, it is not difficult to find that the change trend of the objection function is quasi-synchronous with the sum of PDOPs of nodes in 0 0 . 3 Comparing Figure 32 with Figures 33 and 34, it is not difficult to find that the change trend of the objection function is quasi-synchronous with the sum of PDOPs of nodes in DG2 due to the greater amplitude of the change in the sum value of DG2 than that of DG1. This phenomenon indicates that if β is not the optimal value as designed, the sum PDOP of the group would be more influenced.…”

“…Correspondingly, a swarm composed of multiple unmanned platforms with different motion characteristics or detection sensors is called a heterogeneous swarm. Diverse sensors and carrier platforms in heterogeneous swarms give them better capabilities and more approaches when swarms are detected in complex environments [3][4][5].…”

Flying Watchdog-Based Guard Patrol with Check Point Data Verification

“…The patrolling method can identify suspicious behavior by utilizing single or multiple object optimization techniques to anticipate human movement [24,25]. This approach leverages heuristic search strategies to explore the impact of various intruder behaviors on search performance [26,27]. Critical parameters in the pattern of search within this method encompass the size of the search area, the number of patrolling drones, camera positioning, and the behavior of potential intruders.…”