This paper presents a new method to measure wing deflections in flight for small UAVs. It employs a pair of high resolution stereo cameras and LED wing markers, as well as a small form factor computer for control and storage. Post-processing of all the data is done off-line. Accuracy benchmark tests are conducted. Finally, theoretical discussion of wing shape reconstruction is presented. The method employs numerical optimization by minimizing the difference between numerical geometrically nonlinear slender beam equations and observed markers points with associated uncertainties.Nomenclature c x ,c y = principal focal point C = circularity d = disparity e = error metric f x ,f y = focal length (in pixels) F = focal length F i ,M i = design variables (point forces and moments in UM/NAST) J = cost functional k 1 ,k 2 ,k 3 = radial distortion parameters k = nodal points where point forces and moments are applied N = number of marker points p 1 ,p 2 = tangential distortion parameter p i = beam deflection computed by UM/NAST q i = measured beam deflection Q = reprojection matrix r = radial distance R = rotation matrix t = translational vector T x = x component of rotation matrix x,y = image coordinates (distorted) x c ,y c = image coordinates (pin hole model) X,Y,Z = physical coordinates X cb,i = known checkerboard coordinates σ i = error weight

Recently, there is a growing interest in Systems of Systems (SoS), their architecture, security and application domains. However, their specific characteristics such as the operational independence of SoS constituent systems (CS), the absence of central authority and their emergent behavior make the modeling of their structure, behavior and security a complex task. One of the current main security challenges in the context of SoS is the cascading attack problem. The challenge is to predict the concatenation/sequence of CS's vulnerabilities that could be triggered resulting in destructive cascading failures and take corrective actions to reduce the cost, development time and effect of later changes. In this paper, we propose a domain specific modeling language (DSML) to represent SoS security architecture. Having SoS security models will enable the discovery, analysis and resolution of cascading attacks, in the architecture phase, preventing development time and cost wastage. Following a Model Driven Engineering (MDE) approach, we generate a graphical editor for our DSML and use it to model a Smart Campus case study.

Modern manned and unmanned aircraft designs have a lightweight and flexible structure to increase the flight performance. This trend is continuing with more flexible aircraft structures that possess a nonlinear behaviour. Typically those very flexible are equiped with a flight control system. This paper addresses trajectory control in longitudinal and lateral motion of a highly flexible aircraft with nonlinear dynamics. A new flight control design approach is introduced that is based on the classical multi-loop control law structure with specific considerations for the nonlinear structural behaviour. The inner loops of the flight control laws shall ensure stability and an optimal shape of the aircraft. The nonlinearities are summed up as uncertainties. Control design strategies in the frequency domain are used to achieve the design objectives. The outer loops are based on the classical multi-loop concept for autopilots. The design method and results in a linear and nonlinear simulation of the very flexible unmanned aerial vehicle X-HALE are presented.

This paper presents two different methods for estimating wing shape and rigid body attitude using displacement (stereovision reconstruction) and orientation information (inertia measurement units) at discrete points along a very flexible wing. First, theoretical relationship between the system states (strain, strain rate, body velocity, attitude etc.) and the sensor measurements are derived. Using the derived relationship, a nonlinear least squares fit is developed to obtain wing shape snapshots as well as rigid body attitude. The second method employs a Kalman filter to obtain both shape and rate information. The first method is more accurate since it employs nonlinear strain-displacement relationship at the cost of higher computational cost. However, in the presence of noise in the sensor measurement, the Kalman filter is faster and performs better in terms of accuracy.

No abstract