Joseph F. Horn

Optimal Path Planning of UAVs Using Direct Collocation with Nonlinear Programming

A trajectory generation algorithm using direct collocation with nonlinear programming is successfully demonstrated in simulation. Direct collocation, which approximates the states and controls with piecewise polynomials, has been widely used in space and manned aircraft applications, but has only seen limited use in UAV applications. The algorithm is successfully applied to the generation of a UAV trajectory that provides maximum viewing time for a camera mounted on the UAV. The target can be stationary or moving. Multiple UAVs are considered. In this case, the objective is to provide maximum sensor coverage time using a combination of the UAVs. No specific initial guesses are required to ensure the algorithm is successful.

A Small Semi-Autonomous Rotary-Wing Unmanned Air Vehicle (UAV)

Small radio controlled (R/C) rotary-wing UAVs have many potential military and civilian applications, but can be very difficult to fly. Small and lightweight sensors and computers can be used to implement a control system to make these vehicles easier to fly. To develop a control system for a small UAV, an 8-bit microcontroller has been interfaced with MEMS (Micro-Electro-Mechanical Systems) gyroscopes, an R/C transmitter and receiver, and motor drivers. A single angular degree of freedom test bed has been developed to test these electronics and successful pilot-in-the-loop PI control has been achieved for this test system. A quadrotor with a stability augmentation system that uses these electronics to control the vehicle has also been developed. The future goals of this research are to incorporate more sensors to increase the level of autonomy for UAV operation.

Simulation of Helicopter Shipboard Launch and Recovery with Time-Accurate Airwakes

A simulation of the helicopter/ship dynamic interface has been developed and applied to simulate a UH-60A operating from an LHA class ship. Time accurate CFD solutions of the LHA airwake are interfaced with a flight dynamics simulation based on the GENHEL model. The flight dynamics model was updated to include improved inflow modeling and gust penetration effects of the ship airwake. A maneuver controller was used to simulate pilot control inputs for specified approach and departure trajectories. The CFD solutions show significant time varying flow effects in the airwake. Time histories of the aircraft angular rate and pilot control activity indicate that the time varying nature of the airwake has significant effect on aircraft response and pilot workload.

Flight Testing a Real-Time Direct Collocation Path Planner

A path-planning algorithm using direct collocation with nonlinear programming is demonstrated in both simulation and flight tests. Direct collocation, which approximates the states and controls with piecewise polynomials, has been widely applied in space vehicles and manned aircraft, but has only seen limited use in unmanned aerial vehicle applications. The algorithm is successfully used to generate a path that produces maximal surveillance time of a moving or stationary ground target by a sensor mounted on an unmanned aerial vehicle while compensating for aircraft performance or mission constraints. Flight tests of the path-planning algorithm operating in real time onboard an unmanned aerial vehicle are also presented. These tests include surveilling a stationary and moving target with a video camera while compensating for any wind effects. Additionally, the effect of the use of road data in planning the path is simulated by tracking a second unmanned aerial vehicle flying a predefined pattern.

Intelligent Unmanned Air Vehicle Flight Systems

This paper describes an intelligent autonomous airborne flight capability that is being used as a test bed for future technology development. The unmanned air vehicles (UAVs) fly under autonomous control of both an onboard computer and an autopilot. The onboard computer provides the mission control and runs the autonomous Intelligent Controller (IC) software while the autopilot controls the vehicle navigation and flight control. The autonomous airborne flight system is described in detail. An IC architecture directly applicable to the design of unmanned vehicles is also presented. The UAVs may operate independently or in cooperation with one another to carry out a specified mission. The intelligent UAV flight system is used to evaluate and study autonomous UAV control as well as multi-vehicle collaborative control.

Simulation of Pilot Control Activity During Helicopter Shipboard Operations

A simulation of the helicopter/ship dynamic interface has been developed and applied to simulate a UH-60A operating from an LHA class ship. Time accurate CFD solutions of the LHA airwake are interfaced with a flight dynamics simulation based on the GENHEL model. The flight dynamics model was updated to include improved inflow modeling and gust penetration effects of the ship airwake. An optimal control model of a human pilot was used to simulate pilot control activity for a specified approach trajectory. The pilot model was designed so that the tracking performance could be tuned based on a desired crossover frequency in each control axis. The model was used to predict pilot workload for shipboard approaches in two different wind-over-deck conditions. Although further validation is needed, preliminary results show that the simulation results in similar workload trends as recent flight test studies. Introduction Helicopter shipboard launch and recovery operations continue to be a topic of interest for both civil and military operators. Ship-based helicopters regularly operate under hazardous flight conditions. Highly turbulent airwakes from the ship’s superstructure, low visibility, and moving flight decks make helicopter shipboard operations one of the most challenging, training intensive and dangerous of all helicopter flight operations. To ensure the compatibility of a particular helicopter and ship under various operating conditions, extensive dynamic interface flight-testing must be performed. This approach is both costly and limited by the availability of fleet assets and weather conditions. For example, the wind-over-the-deck (WOD) envelope defines the allowable wind conditions in terms of speed and azimuth for a given helicopter ship combination. The envelope is established by performing a series flight tests for all possible combinations of wind speed and azimuth (typically in 5 knot, 15 deg increments) and taking subjective pilot ratings. The WOD envelopes are often overly restrictive, because certain wind conditions never present themselves during tests. Better simulation tools for analyzing shipboard operations might be used for a variety of purposes. Simulation testing can be used for pilot training and to reduce flight test time and cost for establishing safe WOD envelopes. Ultimately piloted simulation testing might be used in acceptance testing of future rotorcraft and future ships. 1 In addition to the piloted simulation applications, the development of a non-real-time simulation tool for use in engineering and design would also be valuable. Such a simulation tool could be used to find optimal approach paths for safe landings, to design and test new flight control systems, and to identify possible trouble spots for future ship and helicopter combinations early in the design phase. Three key elements of this simulation tool include: CFD representations of the time varying ship airwake, a high fidelity flight dynamics model of the helicopter, and a feedback control model of the human pilot. Understanding and modeling the airwake from complex ship geometries presents a number of technical challenges. Some of the key flow features of the airwake are: unsteadiness, large regions of separated flow, vorticity, and low Mach number. Previous researchers have performed numerous experimental and computational studies of airwakes for different classes of ships. The U.S. Navy has done some experimental investigations using both full-scale tests and scalemodel tests in wind tunnels. 2 Unfortunately it is both difficult and costly to obtain high quality and complete sets of airwake data for all of the different WOD conditions on any given ship. Computational simulations of ship airwakes provide an attractive alternative and have been performed using a number of different numerical approaches by Tai, 3 Tattersall et al, Liu and Long, 5 Guillot and Walker, 6 and Reddy et al. Recently Sharma and Long 8 have simulated inviscid, steady and time-accurate flow over an LPD-17 ship using a modified parallel flow solver PUMA. They found good agreement between the steady-state solution ∗ Graduate Research Assistant, dylee@psu.edu † Assistant Professor, joehorn@psu.edu ‡ Graduate Research Assistant, nxs216@psu.edu § Professor, lnl@psu.edu Copyright

Neural Network Based Trajectory Optimization for Unmanned Aerial Vehicles

A neural network approximation to direct trajectory optimization methods is presented. The method uses neural networks to approximate the dynamics and objective equations integrated over a given time interval. The trajectory is then built recursively and treated as a nonlinear programming problem. The method is compared to a direct collocation method as well as more recent pseudospectral methods and shows competitive results while being computationally faster. In addition, a neural network provides a continuously dierentiable function approximation which may be advantageous when a discontinuous objective function is used in a nonlinear solver. A surveillance trajectory planning problem for an unmanned aerial vehicle is given as an example application and results are presented for all three methods.

Adaptive Model Inversion Control of a Helicopter with Structural Load Limiting

An adaptive control system capable of providing consistent handling qualities throughout the operational flight envelope is a desirable feature for rotorcraft. The adaptive model inversion controller with structural load limit protection evaluated here offers the capability to adapt to changing flight conditions along with aggressive maneuvering without envelope limit violations. The controller was evaluated using a nonlinear simulation model of the UH-60 helicopter. The controller is based on a well-documented model inversion architecture with an adaptive neural network (ANN) to compensate for inversion error. The ANN was shown to improve the tracking ability of the controller at off-design point flight conditions; although at some flight conditions the controller performed well even without adaptation. The controller was modified to include a structural load-limiting algorithm to avoid exceeding prescribed limits on the longitudinal hub moment. The limiting was achieved by relating the hub moment to pitch acceleration. The acceleration limits were converted to pitch angle command limits imposed in the pitch axis command filter. Results show that the longitudinal hub moment response in aggressive maneuvers stayed within the prescribed limits for a range of operating conditions. The system was effective in avoiding the longitudinal hub moment limit without unnecessary restrictions on the aircraft performance.

Detection and Avoidance of Main Rotor Hub Moment Limits on Rotorcraft

Future rotorcraft will require systems that allow carefree maneuvering. This has driven the need for advanced algorithms that predict the onset of structural limits and cue the pilot using tactile feedback. This paper presents a new method for the detection and avoidance of the main rotor hub moment limit. The dynamic nature of the hub moment limit makes it a challenging problem. An algorithm was developed which uses linear models to estimate constraints on longitudinal and lateral cyclic stick positions that ensure the transient response of hub moments remains bounded within prescribed limits. The system was tested using a high-fidelity non-linear simulation of a UH-60A helicopter. The simulation was run using prescribed force inputs and a pilot model to simulate maneuvers. The system was shown to be quite successful in constraining stick travel in multiple axes to prevent hub moment limit violations. The algorithm appeared to be relatively robust and could operate in real-time with a 10 ms time frame. The most critical conditions occurred during control reversals, at which point the system effectively imposed rate limits on the stick motion.

Near Real-Time Simulation of Rotorcraft Acoustics and Flight Dynamics

In this paper, a near-real-time rotorcraft flight dynamics-acoustics prediction system is presented. The highfidelity PSU-WOPWOP rotor noise prediction code is coupled with the GENHEL flight simulation code, which provides low-fidelity blade loading and motion. This system is an initial step intended to investigate the feasibility of real-time rotorcraft noise prediction and to demonstrate the utility of such a system. Limited acoustic validation is shown for a contemporary-design four-bladed main rotor in level flight. A complex 80-s maneuver was used to demonstrate the potential of the coupled system. This realistic maneuver includes a climb, coordinated turn, and level flight conditions. The noise predictions show changes in main rotor noise radiation strength and directivity caused by maneuver transients, aircraft attitude changes, and the aircraft flight—but do not include the effect of blade-vortex-interaction noise. A comparison of the total noise with the thickness and loading noise components helps explain the noise directivity. The computations for a single observer were very fast, although the algorithm is not currently organized as a real-time computation.