TY - GEN
T1 - Modified elman recurrent neural network for attitude and altitude control of heavy-lift hexacopter
AU - Suprapto, Bhakti Yudho
AU - Mustaqim, Amsa
AU - Wahab, Wahidin
AU - Putro, Benyamin Kusumo
N1 - Funding Information:
ACKNOWLEDGMENT This work was supported by the Ministry of Research and Higher Education of Indonesia through Penelitian Disertasi Doktor (PDD) research fund and Hibah Publikasi International Terindeks untuk Tugas Akhir (PITTA) from Direktorat Riset dan Pengabdian Masyarakat (DRPM) Universitas Indonesia.
Publisher Copyright:
© 2017 IEEE.
PY - 2017/12/5
Y1 - 2017/12/5
N2 - Hexacopter is a member of rotor-wing Unmanned Aerial Vehicle (UAV) which has 6 six rotors with fixed pitch blades and nonlinear characteristics that cause controlling the attitude of hexacopter is difficult. In this paper, Modified Elman Recurrent Neural Network (MERNN) is used to control attitude and altitude of Heavy-lift Hexacopter to get better performance than Elman Recurrent Neural Network (ERNN). This Modified Elman Recurrent Neural Network has a self-feedback which provides a dynamic trace of the gradients in the parameter space. In the self-feedback, the gain coefficients are trained as connection weight. This connection weight could enhance the adaptability of Elman Recurrent Neural Network to the time-varying system. The flight data are taken from a real flight experiment. Results show that the Modified Elman Recurrent Neural Network can increase performance with small error and generate a better response than Elman Recurrent Neural Network.
AB - Hexacopter is a member of rotor-wing Unmanned Aerial Vehicle (UAV) which has 6 six rotors with fixed pitch blades and nonlinear characteristics that cause controlling the attitude of hexacopter is difficult. In this paper, Modified Elman Recurrent Neural Network (MERNN) is used to control attitude and altitude of Heavy-lift Hexacopter to get better performance than Elman Recurrent Neural Network (ERNN). This Modified Elman Recurrent Neural Network has a self-feedback which provides a dynamic trace of the gradients in the parameter space. In the self-feedback, the gain coefficients are trained as connection weight. This connection weight could enhance the adaptability of Elman Recurrent Neural Network to the time-varying system. The flight data are taken from a real flight experiment. Results show that the Modified Elman Recurrent Neural Network can increase performance with small error and generate a better response than Elman Recurrent Neural Network.
KW - Direct Inverse Control
KW - Elman Recurrent Neural Network
KW - Heavy-lift Hexacopter
KW - Modified Elman Recurrent Neural Network
UR - http://www.scopus.com/inward/record.url?scp=85045928571&partnerID=8YFLogxK
U2 - 10.1109/QIR.2017.8168502
DO - 10.1109/QIR.2017.8168502
M3 - Conference contribution
AN - SCOPUS:85045928571
T3 - QiR 2017 - 2017 15th International Conference on Quality in Research (QiR): International Symposium on Electrical and Computer Engineering
SP - 309
EP - 314
BT - QiR 2017 - 2017 15th International Conference on Quality in Research (QiR)
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th International Conference on Quality in Research: International Symposium on Electrical and Computer Engineering, QiR 2017
Y2 - 24 July 2017 through 27 July 2017
ER -