The scope of this report is the development of a mathematical model and a control system for a three degrees of freedom (3DOF) helicopter rig. This 3DOF-system offers a good simplification of a real world tandem helicopter for evaluating performanceof different automatic control principles. A mathematical model of the system is developed using free-body diagrams. This mathematical model is then linearized and a controller is developed by decoupling the system. Due to model imperfections external disturbances and similar effects integral action is added as well as feed-forward compensation to reduce nonlinear effects. After the controller has been decoupled the two different controllers are tuned. The Linear-Quadratic Regulator, described in section 3.6, is used for selectingstate-feedback gains. Due to the highly nonlinear nature of the system an Extended Kalman Filter is developed to estimate unmeasurable states. The model and controller is then implemented on the actual rig and evaluated. The results displayed that the elevation controller had good performance. The travel controller also showed good performance but not as good as the elevation controller. The main goal of this thesis was to develop a controller for the 3DOF helicopter system. The results clearly show that an LQR-controller is able to successfully control a system like this with decent performance characteristics despite the highly nonlinear system.

Thesis (M.S. in Computer Science) Naval Postgraduate School, September 1995. / "September 1995." Thesis advisor(s): Michael J. Zyda, John S. Falby. Includes bibliographical references (p. 107). Also available online.

No description available.

No description available.

Unmanned helicopters in recent years have gained much attention due to their potential in both civil as well as military applications. Helicopter is an inherently unstable system. As a result there is a growing need of developing a control structure that allows the helicopter to perform various applications while remaining stable throughout the flight. This thesis presents developments of a robust controller for the vertical channel of an unmanned helicopter while carrying and dropping a payload. In addition, a simulation platform is developed in Simulink that uses a nonlinear six degree of freedom helicopter model. Quantitative Feedback Theory, a frequency domain technique, is used to design a controller that meets specific performance criteria when uncertainties associated with different payload weights exist in the system. The controller performance is examined in simulation for an Xcell 60 helicopter for effective lifting and dropping of up to 10 lb payload. The performance is then compared with a traditional Proportional-Integral-Derivative controller. Further, the effect of actuator dynamics on the controller performance is also evaluated. Finally, a controller that is robust in minimizing the effect of actuator dynamics and the payload drop while keeping the helicopter stable in flight is designed.

Real-time estimates of weight and mass center location for helicopters are desirable for flight control and condition-based maintenance purposes. While methods to estimate mass parameters of helicopters have been developed, they often assume near-perfect knowledge of helicopter dynamics and have been validated only through simulated measurement data. The work described here aims to experimentally validate a method for weight and mass center estimation using an ALIGN T-REX 600e R/C helicopter. The estimation algorithm utilizes an extended Kalman filter (EKF) which estimates the helicopter states along with the weight and mass center location in real-time. Nonlinear system identification is performed using maximum likelihood estimation to create an accurate dynamic model for use in the EKF. Results show that given a reasonably accurate dynamic model, weight, stationline mass center location, and buttline mass center location can be reliably estimated in non-descending flight conditions. Weight estimation is shown to be robust to sudden weight changes during flight, whereas stationline and buttline mass center estimates are marginally robust to sudden shifts in the mass center location. Waterline mass center proved to be unobservable for the axial flight maneuvers conducted. Detailed flight test studies characterize estimation error in weight and three-dimensional mass center position using the EKF formulation.

Mission simulators are in widespread use for the evaluation of military systems under circumstances of appropriate realism. This thesis reports on a series of investigations into the architectures which could result from a synthesis of existing simulation methodologies with novel computing techniques being developed largely for the needs of the internet. The techniques selected were component architectures and mobile agent systems. A framework for simulation based on a component architecture is presented. Entitled MulTiSIM, it permits models to be distributed over a network, and for their interactions to be unaffected by changes in physical distribution and model type. Instances of models can be assembled into arbitrarily complex distributed structures to permit modelling of complex entities, while a degree of structural transparency over interactions with such entities is maintained. Examples of real-time simulators developed using this framework are presented, including a driving simulator for the Thrust super-sonic car and a helicopter mission simulator at DERA. The thesis goes on to suggest various generic roles for mobile agents in medium and large-scale simulations. These roles include communications management functions, dynamic control over model distribution and mediation of specific interactions. Prototypes of the communications management and mediation roles are described, the latter being implemented in the form of what has here been termed an 'ambassador system', employing only a sub-set of the facilities normally required for a system of frilly autonomous mobile agents, while representing the specific interests of a simulation model within a remote operator's station.

We study two different operational scenarios for a regional air ambulance service-company which has bases in Northern California. Two of these bases serve the land areas encompassed roughly in a circular area of radius 100 miles centered in Gilroy and Salinas, respectively; with a large part of their coverage areas reachable from either base. The base in Salinas currently operates one helicopter only from Thursday to Monday, whereas the base in Gilroy operates one helicopter 24/7. The company is considering extending the operation of one helicopter to 24/7 for its Salinas base. In this study we analyze the operational impacts of that extension, and develop a framework that can be applied towards the study of the ambulance assignment problem faced by small operators. / pa/cb Original. 10/06/05. updated 09/09/2011.

No description available.

Helicopter aeroelastic analysis is highly complex and multidisciplinary in nature; the flexibility of main rotor blades is coupled with aerodynamics, dynamics and control systems. A key component of an aeroelastic analysis is the vehicle trim procedure. Trim requires calculation of the main rotor and tail rotor controls and the vehicle attitude which cause the six steady forces and moments about the helicopter center of gravity to be zero. Trim simulates steady level flight of the helicopter. The trim equations are six nonlinear equations which depend on blade response and aerodynamic forcing through finite element analysis. Simulating the behavior of the helicopter in flight requires the solution of this system of nonlinear algebraic equations with unknowns being pilot controls and vehicle attitude angles. The nonlinear solution procedure is prone to slow convergence and occasional divergence causing problems in optimization and stochastic simulation studies. In this thesis, an attempt is made to efficiently solve the nonlinear equations involved in helicopter trim. Typically, nonlinear equations in mathematical physics and engineering are solved by linearizing the equations and forming various iterative procedures, then executing the numerical simulation. Helicopter aeroelasticity involves the solution of systems of nonlinear equations in a computationally expensive environment. The Newton method is typically used for the solution of these equations. Due to the expensive nature of each aeroelastic analysis iteration, Jacobian calculation at each iteration for the Newton method is not feasible for the trim problems. Thus, the Jacobian is calculated only once about the initial trim estimate and held constant thereafter. However, Jacobian modifications and updates can improve the performance of the Newton method. A comparative study is done in this thesis by incorporating different Jacobian update methods and selecting appropriate damping schemes for solving the nonlinear equations in helicopter trim. A modified Newton method with varying damping factor, Broyden rank-1 update and BFGS rank-2 update are explored using the Jacobian calculated at the initial guess. An efficient and robust approach for solving the strongly coupled nonlinear equations in helicopter trim based on the modified Newton method is developed. An appropriate initial estimate of the trim state is needed for successful helicopter trim. Typically, a guess from a simpler physical model such as a rigid blade analysis is used. However, it is interesting to study the impact of other starting points on the helicopter trim problem. In this work, an attempt is made to determine the control inputs that can have considerable effect on the convergence of trim solution in the aeroelastic analysis of helicopter rotors by investigating the basin of attraction of the nonlinear equations (set of initial guess points from which the nonlinear equations converge). It is illustrated that the three main rotor pitch controls of collective pitch, longitudinal cyclic pitch and lateral cyclic pitch have significant contribution to the convergence of the trim solution. Trajectories of the Newton iterates are shown and some ideas for accelerating the convergence of trim solution in the aeroelastic analysis of helicopter are proposed.