Quadrotor Gray-Box Model Identification from High-Speed Flight Data

Sun, Sihao; Visser, Coen C. de; Chu, Q.P.

doi:10.2514/1.c035135

Cited by 45 publications

(45 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, the models are fragmented in the literature, with some focusing only on the axial flow condition, and others only considering a subset of the outputs in (1). In [13], a combination of first-principles and black-box modelling using stepwise regression was used given flight performance data. This is similar to our previous work [15], but the models were quite computationally expensive in the sense that they were not representable by analytical expressions, and thus must be solved numerically.…”

Section: Literature Reviewmentioning

confidence: 99%

“…• Sections 2 and 3: a first-principles-based analytical model is derived for (1). This model is parametrized by nine parameters, which can either be optimized over using labelled data generated from flight experiments as in [13][14][15], or against wind tunnel data as will be done herein. In contrast to the literature, this section provides a consolidated and analytical grey-box model for (1) without limiting restrictions on β.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Computationally Efficient Force and Moment Models for Propellers in UAV Forward Flight Applications

Gill

D’Andrea

2019

Drones

View full text Add to dashboard Cite

Two low-order, parametric models are developed for the forces and moments that a rotating propeller undergoes in forward flight. The models are derived using a first-principles-based approach, and are computationally efficient in the sense of being represented by explicit expressions. The parameters for the models can be identified either using supervised learning/grey-box fitting from labelled data, or can be predicted using only the static load coefficients (i.e., the hover thrust and torque coefficients). The second model is a multinomial model that is derived by means of a Taylor series expansion of the first model, and can be viewed as a lower-order lumped parameter model. The models and parameter generation methods are experimentally tested against 19 propellers tested in a wind tunnel under oblique flow conditions, for which the data is made available. The models are tested against 181 additional propellers from existing datasets.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Computationally Efficient Force and Moment Models for Propellers in UAV Forward Flight Applications

Gill

D’Andrea

2019

Drones

View full text Add to dashboard Cite

show abstract

“…Even though BEM outperforms simple quadratic models and often achieves accurate predictions, it does not account for any interaction between the flow tubes of different propellers or the frame [13]. Previous work has incorporated interaction effects using either static wind tunnel tests [33][34][35] where the vehicle is rigidly mounted on a force sensor, or by performing fast maneuvers in instrumented tracking volumes [36].…”

Section: Related Workmentioning

confidence: 99%

“…More elaborate models based on blade-element-momentum (BEM) theory manage to accurately model single rotors at high wind velocities, but they do not account for the aerodynamic interactions between rotors and the frame. Parametric gray-box models [13] aim to overcome these limitations by describing the forces and torques as a linear combination of library functions. While these models can perform well, their performance hinges on the appropriate choice of basis functions, which require human expert knowledge to design.…”

Section: Introductionmentioning

confidence: 99%

NeuroBEM: Hybrid Aerodynamic Quadrotor Model

Bauersfeld

Kaufmann

Foehn

et al. 2021

Robotics: Science and Systems XVII

Self Cite

View full text Add to dashboard Cite

Fig. 1: Long-exposure images depicting quadrotor trajectory tracking at speeds up to 65 km/h in a large-scale motion-capture system. The captured data is used to fit a hybrid quadrotor model combining blade-element-momentum (BEM) theory with a neural network compensating residual dynamics. This hybrid model reproduces the flown trajectories in simulation with a positional RMSE error reduction of over 50% compared to state-of-the-art.

show abstract

“…The aforementioned literature assumes that the drone is operated around the hovering condition and only limited aerodynamic effects are considered such as the rotational damping [13], [14]. However, in out-door applications, significant aerodynamic forces/moments on the quadrotor are present due to fast cruising speed and large wind disturbances [16], [17]. The system nonlinearity also becomes more significant due to the complex variation of rotor aerodynamic characteristics in high-speed conditions.…”

Section: Introductionmentioning

confidence: 99%

Incremental Nonlinear Fault-Tolerant Control of a Quadrotor With Complete Loss of Two Opposing Rotors

et al. 2021

Self Cite

View full text Add to dashboard Cite

In order to further expand the flight envelope of quadrotors under actuator failures, we design a nonlinear sensor-based fault-tolerant controller to stabilize a quadrotor with failure of two opposing rotors in the high-speed flight condition (> 8m/s). The incremental nonlinear dynamic inversion (INDI) approach which excels in handling model uncertainties is adopted to compensate for the significant unknown aerodynamic effects. The internal dynamics of such an underactuated system have been analyzed, and subsequently stabilized by re-defining the control output. The proposed method can be generalized to control a quadrotor under single-rotor-failure and nominal conditions. For validation, flight tests have been carried out in a large-scale open jet wind tunnel. The position of a damaged quadrotor can be controlled in the presence of significant wind disturbances. A linear quadratic regulator (LQR) approach from the literature has been compared to demonstrate the advantages of the proposed nonlinear method in the windy and high-speed flight condition.

show abstract

Quadrotor Gray-Box Model Identification from High-Speed Flight Data

Cited by 45 publications

References 19 publications

Computationally Efficient Force and Moment Models for Propellers in UAV Forward Flight Applications

Computationally Efficient Force and Moment Models for Propellers in UAV Forward Flight Applications

NeuroBEM: Hybrid Aerodynamic Quadrotor Model

Incremental Nonlinear Fault-Tolerant Control of a Quadrotor With Complete Loss of Two Opposing Rotors

Contact Info

Product

Resources

About