Fig. 1.
Multiplex FunCub Airplane [23]
Tab. 1.
Aerodynamic non-dimensional coefficients
| ID | Par. | Value | ID | Par. | Vale | ID | Par. | Value |
|---|---|---|---|---|---|---|---|---|
| 1 | CD0 | 0.0177 | 5 | CLV | -0.0025 | 9 | Cmα | -1.6173 |
| 2 | CDv | 0.0136 | 6 | CLα | 4.2305 | 10 | Cmq | -8.0193 |
| 3 | CDα | 0.1223 | 7 | Cm0 | 0.0446 | 11 | Cmδe | -1.4830 |
| 4 | CL0 | 0.1518 | 8 | CmV | -0.0092 |
Fig. 2.
Flight Test in an R/C Model Runway
Fig. 3.
Flight Envelope for Tests Performed
Fig. 4.
Bode magnitude plot of the pitching moment for optimum frequency range determination
Fig. 5.
Longitudinal response dataset
Fig. 6.
Schematic of Output Error Method [8]
Fig. 7.
Schematic of Neural Network Training for Aerodynamic Coefficient Prediction
Fig. 8.
Schematic of a single Neuron in ANNs
Fig. 9.
Time histories of measured (blue line) and estimated parameters (red line), using OEM method
Fig. 10.
Time History of Estimated parameters: ANNs Prediction Cycle
Tab. 2.
Aerodynamic non-dimensional coefficients comparison
| ID | Parameter | Estimated value OEM | Relative error OEM | Estimated value ANNs | Relative error ANNs |
|---|---|---|---|---|---|
| 1 | CD0 | 0.0176 | 0.62% | 0.018 | 0.56% |
| 2 | CDV | 0.0131 | 4.04% | 0.013 | 2.94% |
| 3 | CDα | 0.1217 | 0.50% | 0.123 | 0.25% |
| 4 | CL0 | 0.1524 | 0.42% | 0.153 | 0.46% |
| 5 | CLV | -0.0035 | 40.80% | -0.003 | 24.00% |
| 6 | CLα | 4.2509 | 0.48% | 4.233 | 0.05% |
| 7 | Cm0 | 0.0443 | 0.61% | 0.045 | 1.12% |
| 8 | CmV | -0.0095 | 2.83% | -0.010 | 10.87% |
| 9 | Cmα | -1.5681 | 3.04% | -1.619 | 0.09% |
| 10 | Cmq | -6.9714 | 13.07% | -7.923 | 1.20% |
| 11 | Cmδe | -1.4163 | 4.50% | -1.492 | 0.61% |
Fig. 11.
Comparison of CL, CD and Cm between OEM and ANN