Entropy fuzzy system identification for the heave flight dynamics of a model-scale helicopter
Date
2020
Authors
Santoso, F.
Garratt, M.A.
Anavatti, S.G.
Hasanein, O.
Stenhouse, T.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Journal article
Citation
IEEE/ASME transactions on mechatronics, 2020; 25(5):2330-2341
Statement of Responsibility
Conference Name
Abstract
This paper studies non-linear system identification of a small scale and flybar-free unmanned helicopter, the Trex450 chopper, built using commercial off-the-shelf components. We employ the real-time input-output data, obtained from human-controlled flight tests, operating the aircraft under severe ground effects during the vertical flight maneuvers. We highlight the performance of the entropy fuzzy system identification in regards to the performance of several non-linear system identification techniques (i.e. the Takagi-Sugeno (TS) Fuzzy systems, adaptive neuro-fuzzy identification system (ANFIS), and non-linear ARX (NARX)) as benchmarks. Our research confirms the efficacy of the entropy fuzzy identification technique. Despite being non-linear, the proposed fuzzy model is relatively simple, transparent, and highly accurate in representing the complex behaviors of our unmanned helicopter. Yet another major advantage of the proposed technique is the ability to avoid the overfitting, an essential requirement in modeling, as the system is capable of achieving a delicate balance between accuracy and complexity in the acquired model.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright 2019 IEEE