Global Model Identification for a Coaxial Helicopter

Abstract

In this paper, the Blade Element Momentum Theory (BEMT) hast been applied to setup a nonlinear model structure for a coaxial helicopter with a maximum take-off weight (MTOW) of 600kg. The model includes unknown aerodynamic parameters, which are estimated using flight test data. An extensive flight test campaign, with a fully instrumented CoAx 600 helicopter has been carried out to generate the required flight test data for parameter estimation. In this study, we use the maneuvers performed in the vicinity of the hover flight condition for system identification. The nonlinear model can be trimmed and linearized for the analysis of the system dynamics and control design. This has been done for one trim point in the study and the corresponding pole plan has been provided.

On the Flight Test Campaign of a Coaxial Helicopter for the Development of an Unmanned Aerial System

“…For the model identification (see [1]) especially the velocities and accelerations of the helicopter as well as the pitch angles of the rotor blades are of importance. The velocities and accelerations are measured directly as mentioned above.…”

“…These maneuvers include doublets, steps and 3-2-1-1 maneuvers which are widely used for model identification (see [5] and [8]). The maneuver times are set using the estimated nonlinear model presented in [1], which is implemented with known physical helicopter data with a 6-DoF rigid body model of the CoAx 600 prototype with a dynamic rotor model and aerodynamic coefficients based on the blade profile and a lookup table. From this model, a set of linearized models for each maneuver point is calculated.…”

“…Analyzing the maneuver length ratio in comparison to the desired length (see Figure 10) it is shown that the maneuvers are varying around the 100 % line. This variation can be observed mostly during axis or maneuver length changes, as the pilot needs some time to get used to the new tempo (as each axis has its own tempo depending on the eigenfrequencies estimated by the a priori model [1]). But the variation is also beneficial as a higher frequency range is covered and therefore an erroneous estimation of the eigenfrequency can be covered to a certain degree.…”

“…Therefore, a manned flight test campaign is conducted during the development phase of the unmanned system. As detailed in [1] the system identification is performed in the time domain which determines a huge part of the conducted flight maneuvers. Many flight test campaigns in this scale have been performed for identification in the frequency domain as described in [2] [3] and [4].…”

Practical System Identification and Incremental Control Design for a Subscale Fixed-Wing Aircraft

“…System identification for aircraft is a wide topic, and many branches of the field are discussed in [4][5][6]. Recent advances in aircraft system identification include [7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]. Several methods above use frequency domain identification, which was also studied in [26][27][28][29].…”