DESIGN AND IMPLEMENTATION OF A FLIGHT CONTROL SYSTEM FOR AN UNMANNED ROTORCRAFT USING RPT CONTROL APPROACH Guowei Cai, Biao Wang, Ben M. Chen, and Tong H. Lee ABSTRACT In this paper, we apply a so-called robust and perfect tracking (RPT) control technique to the design and implementation of the flight control system of a miniature unmanned rotorcraft, named HeLion.To make the presented work self-contained, we will first outline some background knowledge, including mainly the nonlinear flight dynamics model and the inner-loop flight control system design. Next, the highlight of this paper, that is, the outer-loop flight control system design procedure using RPT control technique, will be detailed. Generally speaking, RPT control technique aims to design a controller such that (i) the resulting closed-loop system is asymptotically stable, and (ii) the controlled output almost perfectly tracks a given reference signal in the presence of any initial conditions and external disturbances. Since it makes use of all possible information including the system measurement output and the command reference signal together with all its derivatives (if available) for control, RPT control technique is particularly useful for the outer-loop layer of an unmanned aircraft. Both simulation and flight-test results will be presented and analyzed at the end of this paper, and the efficiency of the RPT control approach will be evaluated comprehensively. Key Words: Unmanned aerial vehicles, flight control systems, robust control, tracking control. I. INTRODUCTION During the last two decades, miniature unmanned- aerial-vehicle (UAV) helicopters have gained great attention in academic circles worldwide. Some unique features such as low cost, good maneuverability, and easy maintenance, make them an ideal experimental platform for various research purposes. Their growing popularity in the last several years has been further revealed by some successful and impressive implementations (see, for example, [1,12,23,29]). The automatic flight control system is essen- tial for a UAV to carry out flight missions with minimal or even without interference from human pilots. The classical single-input/single-output (SISO) feedback control method (i.e., PD or PID control) is one of the most common choices because of its simplicity in structure with less requirement on the accuracy of the dynamical model of the UAV. Exam- ples include the CMU-R50 UAV helicopter [24], in which a SISO PD control law is adopted and further optimized using CONDUIT for both hovering and forward flight, and the Ursa Major 3 UAV helicopter [27], in which a SISO PID control is implemented for automatic hovering. To improve flight control performance, many researchers are devoted to the study of implementing more advanced control techniques on the miniature rotorcraft UAVs. For example, a flight control system using a MIMO (multi-input/multi-output) H• control approach has been designed and implemented for their mini rotorcraft UAVs in [34]. It is reported that the resulting system has clearly outperformed the classical method. Other cases reported in the literature include systems designed by using: (i) a decentralized decoupled model predictive approach [28]; (ii) a neural network method [15,32]; (iii) adaptive control techniques [11,22]; (iv) a fuzzy logic approach [20]; (v) m-synthesis [33]; (vi) an approximate lin- earization method [21]; (vii) nonlinear control methods [3,26]; (viii) a differential geometry technique [19]; (ix) H• control [16,17]; (x) a learning control technique [14]; (xi) intelligent control methods [31]; and (xii) a sliding mode control technique [13], to name a few. After decades of development, although there is a vast number of works that have been performed along these lines, many are still in the simulation stage. They are still not ready for reliable and mature implementations onto real platforms. Recently, Cai et al. [4,5] (see also Peng et al. [25]) have proposed a flight control scheme consisting of three parts, Manuscript received March 7, 2011; revised September 5, 2011; accepted Novem- ber 6, 2011. Guowei Cai is with Temasek Laboratories, National University of Singapore, Sin- gapore 117411. Biao Wang, Ben M. Chen (corresponding author) and Tong H. Lee are with the Department of Electrical & Computer Engineering, National University of Singapore, Singapore 117576 (e-mails: tslcaig@nus.edu.sg; elewb@nus.edu.sg; bmchen@ nus.edu.sg; eleleeth@nus.edu.sg). Asian Journal of Control, Vol. 15, No. 1, pp. 1–25, January 2013 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/asjc.504 © 2012 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society