TY - GEN
T1 - Real-Time Environmental Forecasting for Autonomous Aircraft
AU - Carmeli, Guy
AU - Moshe, Boaz Ben
AU - Ferrier, Bernard
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022/1/1
Y1 - 2022/1/1
N2 - The research intends to examine the feasibility of predicting a ship's environmental conditions in real time in order to maximize the efficiency and safety of landing autonomous aircraft on its deck. The ship state is represented by 2 main axes: Roll and Pitch. The study will deal with predicting these 2 axes a few seconds ahead, which will allow landing on the ship more safely. According to conversations with pilots, and after looking at accidents that occurred while landing helicopters on ships, there seems to be a real need to increase safety conditions when making manned or autonomous landings. The research will include the development of an artificial intelligence platform that will enable forecasting the pitch and roll conditions on deck. The forecast data of the ship's position will be one of the main factors to be transmitted in real time to the aircraft; knowledge of the ship's immediate and future position will facilitate and ensure a soft landing of the aircraft on its deck. The ability to predict the ship's future conditions will equip the ship and the drone with a technological advantage, as the platform will enable the aircraft to plan its landing and perform it more safely.
AB - The research intends to examine the feasibility of predicting a ship's environmental conditions in real time in order to maximize the efficiency and safety of landing autonomous aircraft on its deck. The ship state is represented by 2 main axes: Roll and Pitch. The study will deal with predicting these 2 axes a few seconds ahead, which will allow landing on the ship more safely. According to conversations with pilots, and after looking at accidents that occurred while landing helicopters on ships, there seems to be a real need to increase safety conditions when making manned or autonomous landings. The research will include the development of an artificial intelligence platform that will enable forecasting the pitch and roll conditions on deck. The forecast data of the ship's position will be one of the main factors to be transmitted in real time to the aircraft; knowledge of the ship's immediate and future position will facilitate and ensure a soft landing of the aircraft on its deck. The ability to predict the ship's future conditions will equip the ship and the drone with a technological advantage, as the platform will enable the aircraft to plan its landing and perform it more safely.
KW - CNN Convolution Neural Network
KW - Dense Feed forward Fully connected Neural Network
KW - IMU Inertial Measurement Unit
KW - LSTM Long short-Term memory
KW - WMAPE Weighted mean absolute percentage error
UR - http://www.scopus.com/inward/record.url?scp=85134165963&partnerID=8YFLogxK
U2 - 10.1109/ICAPAI55158.2022.9801570
DO - 10.1109/ICAPAI55158.2022.9801570
M3 - Conference contribution
AN - SCOPUS:85134165963
T3 - 2022 International Conference on Applied Artificial Intelligence, ICAPAI 2022
BT - 2022 International Conference on Applied Artificial Intelligence, ICAPAI 2022
PB - Institute of Electrical and Electronics Engineers
T2 - 2022 International Conference on Applied Artificial Intelligence, ICAPAI 2022
Y2 - 5 May 2022
ER -