Flight Science & Technology Research Group

The Adaptive Pilot Model in Simulation Fidelity

Department Home Page
 
Courses
 
Simulation Facilities

FST Staff

Research

Publications

Aircraft Library

News & Events

Vacancies

HEADSTART

University home page

Feedback: mdw@liv.ac.uk

The level of fidelity of Flight Simulators, or, more generally Synthetic Training Devices (STD), determines their fitness for purpose and is quantified in documents like JAR-STD-1H in terms of performance criteria for the individual components, e.g. the motion/visual/sound systems, the mathematical model. 

Component fidelity is important but the standards also require piloted assessment of the integrated system with typical mission sorties flown covering the training aspects for which the system will be used.  Subjective opinion here is important too because it reflects the value that an experienced pilot places on the level of realism.  Quantifying overall simulation fidelity is more difficult however, but is equally important because, arguably, component or sub-system fidelity can only be properly related to fitness for purpose if connected by measure to the whole.

Attempts to quantify overall simulation fidelity within the framework of handling qualities engineering have been presented in a number of forms in recent years.  Hess and colleagues have developed an approach based on pilot-aircraft modelling and introduced the handling qualities sensitivity function as the basis of a quality metric.  McCallum et al propose the use of the ADS-33 performance standards for deriving metrics.  Within the JSHIP project, Wilkinson and Advani, and Roscoe and Thompson present an approach using comparative measures of performance and control activity, correlated with handling qualities ratings given for the same tasks flown in simulation and flight.  In all these approaches, the philosophy has been to develop a rational and systematic approach to identifying differences between tasks performed in simulation and flight, hence directing attention to simulation deficiencies.  While JAR-STD 1H is directed at the training community, fidelity criteria are equally applicable to simulation in design, research and development.  In these areas, flight simulation can be a primary source of data from which knowledge is derived, decisions are made and significant resources committed.

  This research continues the development of an approach for quantifying overall simulation fidelity based on an analysis of pilot visual guidance strategy, identifying the control loops utilised, levels of abruptness and the cues available to support anticipation. The premise is that if the control strategy adopted to perform the same flying task is ‘equivalent’ in flight and simulation, then the fidelity is good and the training device fit for purpose.  The meaning of equivalent is developed in terms of what we describe as the Adaptive Pilot Model (APM) concept, whereby the combined pilot and aircraft is modelled and comparisons made of model parameters identified from the same curve fitting process applied to data from flight and simulation tests.  As with previous studies, the research is thus concerned with approximations for describing the behaviour of the combined pilot-aircraft system.  However, in the present work, it is assumed that the pilot adapts control strategy during the manoeuvre, with the adaptation reflected in the changing model parameters.  Thus the changing pilot gains relating to velocity and distance control, for example, are tracked through the manoeuvre.  The concept can be extended under the premise that motion control by the pilot follows temporal rather than spatial guidance principles.  The results gained through analysing the way in which helicopter pilots manoeuvre indicate that pilots strictly have no need for velocity or distance information, per se, when manoeuvring close to a surface.  Instead, they use information about time to close on surfaces, t(t), to make judgements about relative motion and control requirements.

 Areas for innovation in the research include;  

• Prediction and understanding of changes of pilot control strategy during manoeuvres through APM

• Extend guidance APM to include stabilisation

• Develop simulation fidelity criteria

• Explain different piloting techniques through the APM concept

Figure 1 - Response to Longitudinal Cyclic in Hover, Comparison of Flight Test and FLIGHTLAB Simulation – Bo105

Members:

Professor Gareth Padfield 

Dr Mark D White

Mr Rob Armstrong

Go back to FST homepage