Decision control system

• Main Author: Sekar, Karthik.
• Other Authors: Suresh Sundaram
• Format: Theses and Dissertations
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Aeronautical engineering
• Online Access: http://hdl.handle.net/10356/53093
• Institution: Nanyang Technological University

Since the remote controlled UAVs have certain limitations dictated by weather conditions, controller skill, equipment modernization etc., autonomous UAVs have been used to replace the remote controlled UAVs in recent time but these unmanned drones are not as skilful as their human controlled counterparts. To use autonomous drones an effective command decision system (task allocation between the UAVs) has to be setup so as to make these unmanned drones as efficient as the manned drones. The command decision system is based on the strategies developed using game theory whose primary function is to create a reasoning model that mimics a human modus operandi. This paper discusses about the efficient task allocation in-between the members of a UAV squadron also referred to as a swarm under different scenarios. In the first scenario the UAV are required to attack any target that they come across while carrying out search operations in an unknown territory. Here the targets destroyed, using different strategies that are developed from negotiation scheme and greedy algorithm is studied. Based on the performance (outcome- time taken to destroy the targets) the game theories are compared. In the second scenario the UAVs have to escort the bomber/transport through a hostile territory assumed to be completely under enemy radar surveillance. The interceptors that come within range of the bomber/transport are destroyed by the UAVs using different strategies developed from the Borda count and the greedy Algorithm method and the mission success is studied. The different strategies are compared and it is found that negotiation scheme performs better than the greedy algorithm in the search and destroy scenario while the strategy developed using Borda count performs better than the strategy developed using the greedy algorithm in the Escort scenario.