Identifiers of reduced dimension in the problem of the unmanned aerial vehicle stabilization in perturbed atmosphere

Аuthors

Khrustaliov M. M.^*, Khalina A. S.^**

V.A. Trapeznikov Institute of Control Sciences of Russian Academy of Sciences, 65, Profsoyuznaya str., Moscow, 117997, Russia

*e-mail: mmkhrustalev@mail.ru
**e-mail: an.khalina@gmail.com

Abstract

Unmanned aerial vehicles (UAVs) have taken their rightful place in modern activities. The UAV control processes improvement is undoubtedly actual and allows employing more and more unmanned robot-vehicles.

The article considers the problem of a small UAV horizontal flight stabilizing in the vertical plane with account for angular motion. Due to the UAV smallness, the wind should be taken into account, and the control system should ensure high stabilization quality while being simple enough.

Stochastic quasilinear system was selected as a model for the control process description with account for the wind, as accounting for the ongoing processes most adequately, but, at the same time accessible for the effective analysis. Vertical and horizontal components of the wind disturbance are set by the modified Dryden-type generating filters

The authors suggested to employ minimum number of measured motion parameters to simplify the stabilization system.

Since that it is not always possible to stabilize the system by a limited set of state vector components for control, it is conventional to use the state identifiers. However, there several problems exist herewith. The generally accepted in practice formulation the problem of optimal control synthesis is based on the separation theorem. According to this theorem, the optimal control consists of an optimal filter, forming an estimate of the state vector, and optimal regulator determining the control under the assumption that the state vector is known exactly. The result of this theorem is strictly proven for linear systems only. As for the quasi-linear systems, to which the considered system is referred to, the separation theorem is inapplicable.

The presented paper proposes a natural approach to the joint synthesis of the control srategy and the identifier. In this case, the authors propose selecting both the control strategy parameters and the identifier parameters from the condition of the general stabilization quality criterion minimum. The proposed method feature consists in the absence of control and observation synthesis problems separation.

The general method for solving problem of optimal quasilinear stochastic system synthesis to find optimal control strategy in the presence of incomplete information on the system state vector was used. This method was earlier developed by the authors. It includes the necessary optimality conditions and the numerical method of the gradient type.

Keywords:

small unmanned aerial vehicle, optimal control, movement stabilization, wind perturbation suppression, low-dimensional identifier

References

1. Paraev Yu.I. Vvedenie v statisticheskuyu dinamiku protsessov upravleniya i fil’tratsii (Introduction to the statistical dynamics of control and filtration processes), Moscow, Sovetskoe Radio, 1976, 156 p.

2. Randal W. Beard, Timothy W. McLain Small unmanned aircraft: theory and practice. Princeton, NJ: Princeton University Press, 2012, 320 p.

3. Khalina A.S., Khrustalev M.M. Izvestiya RAN. Teoriya i sistemy upravleniya, 2017, no. 1, pp. 65 – 88.

4. Etkin B. Dynamics of atmospheric flight. John Wiley & Sons, 1972, 608 p.

5. Etkin B. Turbulent wind and its effect on flight // Journal of Aircraft, 1981, vol. 18, no. 5, pp. 327 – 345.

6. Parysheva G.V., Yaroshevskii V.A. Uchenye zapiski TsAGI, 2001, vol. XXXII, no. 1-2, pp. 102 – 118.

7. Lebedev G.N., Mikhailin D.A., Rumakina A.V. Trudy MAI, 2016, no. 91, available at: http://trudymai.ru/eng/published.php?ID=75637

8. Rybnikov S.I., Nguen Tkhan’ Shon. Trudy MAI, 2018, no. 98, available at: http://trudymai.ru/eng/published.php?ID=90450

9. Kvakernak Kh., Sivan R. Lineinye optimal’nye sistemy upravleniya (Linear Optimal Control Systems), Moscow, MIR, 1977, 656 p.

10. Di Masi G.B., Gombani A., Kurzhanski A.B. Modeling, Estimation and Control of Systems with Uncertainty, USA, Birkhäuser Boston, 1991, 472 p.

11. Mohinder S. Grewal, Angus P. Andrews. Kalman filtering: theory and practice using MATLAB, John Wiley & Sons, 2015, 640 p.

12. Simon D. Optimal State Estimation. Kalman, H_∞, and Nonlinear Approaches, John Wiley & Sons, 2006, 552 p.

13. Jazwinski A.H. Stochastic Processes and Filtering Theory, Academic Press, 1970, 376 p.

14. Sadomtsev Yu.V. Avtomatika i telemekhanika, 2006, no. 12, pp. 175 – 189.

15. Rom D.B., Sarachik P.E. The design of optimal compensators for linear constant systems with inacessible states, Transactions on Automatic Control, 1973, vol. AC-18, no. 5, pp. 509 – 512.

16. Blanvillain P.J., Johnson T.L. Specific-optimal control with a dual minimal-order observer-based compensator, International Journal of Control, 1978, vol. 28, no. 2, pp. 277 – 294.

17. Dai L. Singular control systems: Lecture notes in control and information scinces. Berlin, Heidelberg, Springer-Verlag, 1989, 334 p.

18. Zhang L., Huang B., Lam J. LMI synthesis of H₂ an mixed H₂/H_∞ controllers for singular systems, IEEE Transactions on Circuits and Systems-II: Analog and Digital Signal Processing, 2003, vol. 50, no. 9, pp. 615 – 626.

19. Kabanov S.A. Optimizatsiya dinamiki sistem pri deistvii vozmushchenii (Optimization of the dynamics of systems under the influence of perturbations), Moscow, FIZMATLIT, 2008, 200 p.

20. Korovin S.K., Fomichev V.V. Nablyudateli sostoyaniya dlya lineinykh sistem s neopredelennost’yu (State observers for linear systems with uncertainty), Moscow, FIZMATLIT, 2007, 224 p.

21. Il’in A.V., Korovin S.K., Fomichev V.V. Trudy Matematicheskogo instituta imeni V. A. Steklova RAN, 2008, vol. 262, pp. 87 – 102.

Download