ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

A Heuristic Learning Algorithm for Preferential Area Surveillance by Unmanned Aerial Vehicles

Ramasamy, Manickam and Ghose, Debasish (2017) A Heuristic Learning Algorithm for Preferential Area Surveillance by Unmanned Aerial Vehicles. In: JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS, 88 (2-4, 1). pp. 655-681.

[img] PDF
Jou_Int_Rob_Sys_88-2_655_2017.pdf - Published Version
Restricted to Registered users only

Download (4MB) | Request a copy
Official URL: http://doi.org/10.1007/s10846-017-0498-5


A heuristic learning algorithm is presented in this paper to solve the problem of persistent preferential surveillance by an Unmanned Aerial Vehicle (UAV). The algorithm helps a UAV perform surveillance as per quantitative priority specifications over a known area. It allows the specification of regional priorities either as percentages of visitation to be made by a UAV to each region or as percentages of surveillance time to be spent within each. Additionally, the algorithm increases the likelihood of target detection in an unknown area. The neighborhood of a detected target is suspected to be a region of a high likelihood of target detection, and the UAV plans its path accordingly to verify this suspicion. Similar to using the target information, the algorithm uses the risk information to reduce the frequency of visits to risky regions. The technique of using risk map to avoid risky regions is adapted from the existing geometric reinforcement learning technique. The effectiveness of this algorithm is demonstrated using simulation results.

Item Type: Journal Article
Additional Information: Copy right for this article belongs to the SPRINGER, VAN GODEWIJCKSTRAAT 30, 3311 GZ DORDRECHT, NETHERLANDS
Department/Centre: Division of Mechanical Sciences > Aerospace Engineering(Formerly Aeronautical Engineering)
Date Deposited: 03 Nov 2017 10:54
Last Modified: 03 Nov 2017 10:54
URI: http://eprints.iisc.ac.in/id/eprint/58141

Actions (login required)

View Item View Item