Citation Context

...us hover. Both LaCivita et al. [13] and Ng et al. [17] achieved sustained autonomous hover and accurate flight in regimes where the helicopter’s orientation is fairly close to upright. Roberts et al. =-=[18]-=- and Saripalli et al. [19] achieved vision based autonomous hover and landing. In contrast, autonomous flight achievements in other flight regimes have been very limited. Gavrilets et al. [9] achieved...

Citation Context

...ility of classical Kalman and extended Kalman filter algorithms. In prior work it has been shown that angular velocity and orientation (or attitude) can be estimated from the output of a low cost IMU =-=[13]-=-, [8], [6] for a vehicle that is in quasistationary, or hover, flight. Conversely, translational position and velocity can not be estimated from a low cost IMU for more than a few seconds due to unbou...

Citation Context

...are general, and can be applied to modeling large classes of vehicular dynamics. 1 Introduction In the last few years, considerable progress has been made in finding good controllers for helicopters. =-=[7, 9, 2, 4, 3, 8]-=- In designing helicopter controllers, one typically begins by constructing a model for the helicopter’s dynamics, and then uses that model to design a controller. In our experience, after constructing...

Citation Context

... systems. For example the YAMAHA RMAX radio controlled helicopters use high quality optical gyros for their stability augmentation systems. However, military specification inertial measurement units are expensive, often subject to export restrictions, and not suitable for commercial applications. More recently, the focus on new low cost aerial robotic systems, has lead to a strong interest in attitude estimation algorithms. Traditional linear Kalman filter techniques [9] (including EKF techniques [5], [7]) have proved difficult to apply robustly to applications with low quality sensor systems [8]. The inherent non-linearity of the system and non-Gaussian noise encountered in practice can lead to very poor behaviour of such filters. More sophisticated stochastic filtering techniques such as particle filters and even unscented filters place too much computational load on the low cost processing systems associated with the commercial applications considered. The majority of filtering techniques used and implemented to estimate the attitude of UAVs are based on linear complementary filters with frequency domain analysis [1]; a tradeoff between a good short term precision given by gyroscop...

Citation Context

...roperties for linear systems [12]. Extended Kalman Filtering has been studied for a range of aerospace applications [7], [9], [6], [16]. Such filters, however, have proved difficult to apply robustly =-=[14]-=-, [4], [16]. In practice, many applications use simple linear single-input single-output complementary filters [16], [5]. In recent work, a number of authors have developed nonlinear analogous of sing...

Citation Context

...oximately 17̓ x 13̓, an aperture set tosf/8 and focus set to infinity.sA sensor platform was constructed, consisting of thesDragonfly camera and lens, a GPS receiver, and an inertialsmeasurement unit =-=[20]-=- which was used to measure cameraspose. This was mounted atop a ridge near Mary CairncrosssReserve, a location with an elevation of approximately 1000ft,saround 2nmi SE from the township of Maleny.sFr...