ACADSTAFF UGM

CREATION
Title : Deep Q-Network Configuration and Performance for a Power Line Inspection Autonomous Quadrotor in a Simulated World
Author :

Muhammad Idham Ananta Timur, M.Kom. (1) Prof. Dr. Ir. Jazi Eko Istiyanto, M.Sc. (2) Dr. Andi Dharmawan, S.Si., M.Cs. (3) ROIS NUR HAKIM (4) AHMAD SHIDDIQ N (5)

Date : 0 2022
Keyword : Deep Q-Network, UAV, Reinforcement Learning, power line monitoring Deep Q-Network, UAV, Reinforcement Learning, power line monitoring
Abstract : Monitoring power line across mountainous landscape is always a daunting task. Usually, pilot-driven Unmanned Aerial Vehicle (UAV) such as quadrotor is one of the most viable option to do so. However, this scenario put a high demand on the pilot both in availability and costs. Therefore, this study attempts to adopt autonomous approach using Deep Q-Network algorithms to maintain UAV position against the power line. Sum of rewards per episodes and loss function analysis are used for the evaluation. The result shows that adjusting number of action and kernel size could significantly accelerate learning process while maintaining an acceptable sum of reward. On the other side, bigger kernel size and more actions will slow the learning process down but gaining even better sum of rewards.
Group of Knowledge : Sistem Informasi Geografi (SIG)
Original Language : English
Level : Internasional
Status :
Published
Document
No Title Document Type Action
1 Cek Similarity.pdf
Document Type : [PAK] Cek Similarity
[PAK] Cek Similarity View
2 Full Dokumen.pdf
Document Type : [PAK] Full Dokumen
[PAK] Full Dokumen View
3 Bukti Korespondensi QuadRotor.pdf
Document Type : [PAK] Bukti Korespondensi Penulis
[PAK] Bukti Korespondensi Penulis View
4 DEEP Q-NETWORK CONFIGURATION AND PERFORMANCE FOR A POWER LINE INSPECTION AUTONOMOUS QUADROTOR IN A SIMULATED WORLD.pdf
Document Type : [PAK] Cek Similarity
[PAK] Cek Similarity View
5 full-dokumenpdf.pdf
Document Type : [PAK] Full Dokumen
[PAK] Full Dokumen View