TY - GEN
T1 - A swarm-intelligence approach to oil spill mapping using unmanned aerial vehicles
AU - Ball, Zachary
AU - Odonkor, Philip
AU - Chowdhury, Souma
N1 - Publisher Copyright:
© 2017, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.
PY - 2017
Y1 - 2017
N2 - Swarm of Unmanned Aerial Vehicles (UAV) is an emerging paradigm, where a collaborative team of simple UAVs are used to perform complex mapping missions at a lower expense and/or greater effectiveness compared to a sophisticated manned/unmanned aircraft. This paper capitalizes on this potential by proposing a swarm-intelligence inspired approach to map the boundary of offshore oil spills. A new probability map concept is developed, where an occupancy grid is probabilistically updated based on whether an UAV is recording the image of water, oil, or both. Sharing actionable information among UAVs in the form of occupancy grid updates can significantly reduce computing and communication overheads. The state of the probability map is used to control the exploitive and explorative components of the velocity vector that drives the waypoint planning decisions of each UAV. The flight of UAVs is divided into scouting, aggressive oil mapping, and boundary tracing phases. Various dispatch strategies and parameter settings are explored to identify suitable resource allocation for mapping efficiency. Simulated case studies are developed assuming quadrotor UAVs, with pertinent range and speed constraints. Resulting simulations showed that this method is able to map simple/complex and small/large oil spills with a greater than 95% success rate.
AB - Swarm of Unmanned Aerial Vehicles (UAV) is an emerging paradigm, where a collaborative team of simple UAVs are used to perform complex mapping missions at a lower expense and/or greater effectiveness compared to a sophisticated manned/unmanned aircraft. This paper capitalizes on this potential by proposing a swarm-intelligence inspired approach to map the boundary of offshore oil spills. A new probability map concept is developed, where an occupancy grid is probabilistically updated based on whether an UAV is recording the image of water, oil, or both. Sharing actionable information among UAVs in the form of occupancy grid updates can significantly reduce computing and communication overheads. The state of the probability map is used to control the exploitive and explorative components of the velocity vector that drives the waypoint planning decisions of each UAV. The flight of UAVs is divided into scouting, aggressive oil mapping, and boundary tracing phases. Various dispatch strategies and parameter settings are explored to identify suitable resource allocation for mapping efficiency. Simulated case studies are developed assuming quadrotor UAVs, with pertinent range and speed constraints. Resulting simulations showed that this method is able to map simple/complex and small/large oil spills with a greater than 95% success rate.
KW - Oil spill mapping
KW - Particle swarm optimization
KW - Swarm intelligence
KW - Unmanned aerial vehicles (UAV)
UR - http://www.scopus.com/inward/record.url?scp=85088061285&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85088061285&partnerID=8YFLogxK
U2 - 10.2514/6.2017-1157
DO - 10.2514/6.2017-1157
M3 - Conference contribution
AN - SCOPUS:85088061285
SN - 9781624104497
T3 - AIAA Information Systems-AIAA Infotech at Aerospace, 2017
BT - AIAA Information Systems-AIAA Infotech at Aerospace, 2017
T2 - AIAA Information Systems-Infotech At Aerospace Conference, 2017
Y2 - 9 January 2017 through 13 January 2017
ER -