Search

DOI: 10.2514/1.C000204 Human pilot control during the landing approach is modeled using a stochastic switched model, which is a combination of the conventional linear regression model and a hidden Markov model. Using this model, the time histories of pilot control can be categorized in several states, such as pitch stabilizing mode. First, the selection of modelinputsisdiscussed.Basedondataobtainedin flightsimulatorexperiments,thepilotapproachcontrolmodelis constructed for several cases, with different wind conditions applying an expectation-maximization algorithm. The obtained models are analyzed with respect to the timing of state transitions and the stochastic-switched-model gain parameters. The current findings suggest that the proposed analysis method has a great potential to reveal a pilot’s decision-making process.

Human pilot landing control has been analyzed by the authors using neural network modeling. Although the previous research considered only longitudinal control, analysis of the lateral control is the focus of this paper. The lateral control can be quite difficult under crosswind conditions, because the combination of two lateral control methodologies (crab method and wing-low method) is needed for a successful landing. To analyze lateral control, this paper makes a first step: a lateral pilot control model is established. The necessary visual cues for lateral control are defined, and neural network models for aileron and rudder controls are constructed. The adequacy of the proposed neural network model is checked by Monte Carlo simulations.

T HIS Note presents a real-time flight trajectory optimization method. The flight trajectory that minimizes the fuel consumption or flight time of an aircraft can be solved with optimization. Because the optimization is time-consuming, the optimal flight trajectory needs to be obtained before flight. This, however, cannot cope with unexpected situations in flight. Thus, the real-time optimization, which optimizes the flight trajectory with the transition of the flight state, is important. The final goal of our study is to establish the real-time trajectory generation applicable to emergency landing approaches. Though the present flight management system provides the optimal flight path for a flight plan in normal operations, it cannot operate in emergency situations. This Note presents a fundamental real-time trajectory optimization algorithm and its flight validation. Many studies about real-time flight-path generation have produced a flight path by connecting the trim conditions in prestored databases [1,2]. In these studies, computer-assisted simulations were performed, but there were few actual flight experiments. Additionally, it is difficult to generate nonstational trajectories and the wind effects were not considered. In Japan, the Society of Japanese Aerospace Companies has promoted research on the development of a fault-tolerant flight control system. In a proceeding study, Suzuki et al. proposed a flight trajectory search method composed of a realtime A* algorithm and a random tabu search method [3]. They also conducted flight experiments. This random-based method, however, lacks convergent stability, so it often does not produce the appropriate trajectory. We have taken over this study, and in this Note we apply a direct collocation method including some schemes to solve the preceding problems. Actual flight experiments are also conducted to prove the effectiveness of the method. The experimental aircraft is MuPAL, the Multi-Purpose Aviation Laboratory, developed by JAXA (Japan Aerospace Exploration Agency) [4]. It has a high-precision Global Positioning System/inertial navigation system and a three-axis airspeed sensor. The system allows the estimation of the wind velocity and direction during flight. MuPALalso has a tunnel-in-the-sky display, which was developed by JAXA [5]. The experiments were conducted under manual tracking control using the display. In the future, the trajectory generated by the real-time optimization method will be tracked by an automaticflight control systembeing developed in the research on the fault-tolerant flight control system.

A new design for a tail-sitter vertical takeoff and landing unmanned aerial vehicle was proposed. A nonlinear mathematical model of the vehicle dynamics was constructed by combining simple estimation methods. The flight characteristics were revealed through a trim analysis and an optimized transitional flight path analysis by using the mathematical model. The trim analysis revealed the existence of a flight path constraint to avoid stall; the vehicle could not descend in low-speed flight without high-lift devices such as flaps and slats. These devices improved the descentperformance. Inparticular, slatsprovidedasubstantialimprovement; theyenabled adescent rateof2 m=s. In the optimized transitional flight path analysis, a level outbound transition without high-lift devices was achieved althoughatrimmedlevel flightatlowspeed,aswasshowninthetrimanalysis,wasnotpossible;thiswasbecausethe outboundtransition wasanaccelerative flight.Onthe contrary,without high-lift devices, the vehicle could notavoid climbing to avoid stall during inbound transitions. The slats provided a satisfactory improvement during the transition and enabled a level inbound transition. These results showed the necessity of leading-edge slats for the proposed tail-sitter vertical takeoff and landing unmanned aerial vehicle.

A neural network modeling approach has been developed to analyze the pilot's use of visual cues for landing a transport airplane. Time sequences of the visual cues and pilot control inputs obtained by using a flight simulator can be analyzed to quantitatively estimate the relationship between the visual cues and the pilot control inputs. In this paper, visual cues such as the horizon, runway shape, and runway marker are compared based on their importance. By using a flight simulator, neural network models are obtained in all cases wherein a pilot intentionally alters his or her attentiveness to the visual cues. The contribution ratios analysis reflects the attentiveness to each visual cue. The Monte Carlo landing simulation shows the difference in robustness of each obtained neural network model. It is confirmed that the timely choice of the appropriate visual cue is necessary for a smooth and safe landing.

A neural-network-modeling approach is applied to analyze human-pilot control inputs during the landing phase in the visual approach. Flight data that contain the aircraft state variables and pilot control inputs are recorded using a flight simulator. The time history of visual cues and control inputs is utilized as teaching data for neural networks that can emulate the movements of a human pilot. A genetic algorithms approach is proposed to improve the generalization ability of the network by determining the network structure and initial values of its parameters. Generalization capabilities are evaluated by analyzing the flight data of a personal-computer-based simulator. The contribution ratios of each visual cue and their sensitivities to the control inputs of a pilot are estimated by analyzing the obtained neural-network models using a training simulator. The obtained results reveal that the proposed method can be used for analyzing the skill of a pilot.

An aircraft design and its flight trajectory problem should be formulated simultaneously when an exact flight mission is determined or an extremely high-performance level is required. This article presents a block diagonal sequential quadratic programming approach by introducing interface variables between the two problems to efficiently solve simultaneous optimization problems. As numerical examples, the design of a sailplane for a flight distance competition is studied. The need of simultaneous optimization will be demonstrated by comparing a conventional design process; i.e., a lift-drag ratio maximization approach.

Simultaneous design optimization is considered for structure and control parameters of a wing with a gust load alleviation (GLA) control system. The application of a goal programming (GP) formulation to the design synthesis of an aeroservoelastic system is carried out by the use of a simple mathematical model in conjunction with a wind-tunnel model having a GLA control system. Numerical applications are based on a cantilever wing having an aileron surface controlled by wing-tip accelerometer feedback signals. System equations are obtained in the form of state equations, thus enabling the statistical characteristics of both the gust-induced wing stress and the control surface deflection angle to be evaluated using their standard deviations. The wing spar height and the controller feedback gain are simultaneously optimized to obtain the minimum spar weight while satisfying the following structure and control design constraints: 1) the spar stress is limited with the control system either on or off; 2) the control surface deflection angle is restricted; and 3) system stability should be guaranteed by incorporating a controller stability margin. Numerical examples demonstrate the successful application of a GP formulation for the simultaneous structure/control design synthesis by specifying priorities to the conflicting design constraints.

\0 0 3 A > /, 0 IV In the present paper, the finite element technique is applied to the calculation of three-dimensi onal steady incompressible potential flow around a wing-body combination, the main emphasis being placed upon the study of the aerodynamic interference effect between the wing and the body. The formulation of the problem is made in the form of an integral equation following Morino's method. By the use of the finite element technique combined with the collocation method, this integral equation is reduced to a set of linear algebraic equations. By solving this set, the pressure distribution over the surface of the wing-body combination is obtained. Wind tunnel experiments have been conducted at National Aerospace Laboratory of Japan in pace with the numerical analysis. It has been found that agreement between the numerical results obtained by the present method and those obtained by the wind tunnel experiments is very encouraging.