Multirotor electric aerial vehicle model identification with flight data with corrections to physics-based models

Publication Date

7-1-2022

Document Type

Article

Publication Title

CEAS Aeronautical Journal

Volume

13

Issue

3

DOI

10.1007/s13272-022-00583-5

First Page

575

Last Page

596

Abstract

Developing standard, well-vetted methods for modeling and simulation, prediction of flying/handling qualities, and control system design is critical for improving safety and quality control of multirotor electric aerial vehicles. This paper explores two methods for modeling the dynamics of a small (56 cm, 1.56 kg) hexacopter at hover and forward flight. The first modeling method was system identification from flight data, the second method was a physics-based blade element model with 10 state Peter-He inflow. Evaluation of the fidelity for both the system-identification and physics-based models was completed by comparison to flight data at hover and forward flight. The results were used to classify the importance of key dynamic building blocks on the model fidelity, such as motor/rotor lag dynamics, inertia, and dynamic inflow.

Funding Number

W911W61120012

Funding Sponsor

Rensselaer Polytechnic Institute

Keywords

Electric VTOL (eVTOL), Flight mechanics, System identification

Department

Research Foundation

Link to Full Text

Share

COinS