Classifying Human Pilot Skill Level Using Deep Artificial Neural Networks

To fully optimize the synergy between human operators and machines in modern day’s highly automated vehicle control tasks, a real-time quantitative feedback of skill level is required. Direct feedback of skill level could be used to enable scalable levels of autonomy of the controlled system, or to provide a warning when sudden skill level...

Identification of manual control behaviour to assess rotorcraft handling qualities

Flight safety has been a fundamental aspect of aircraft, and the future demand for wider usage of aerial operations leads to more focus on the flight safety. Particularly rotorcraft require high standards of flight safety due to their inherent features, such as complicated rotary mechanisms, close-to-ground operations, and complex aerodynamic...

A Cybernetic Approach to Assess the Training of Manual Control Skills

This paper presents a cybernetic approach to assess the training of manual control skills in simulators. The approach uses multi-channel pilot models that separate pilots’ responses to visual and motion stimuli. This allows for a quantitative analysis of pilots’ use of visual and motion cues for manual aircraft control, as well as the evolution...

Modelling transport pilot control behaviour and the effect of motion during a simulated stall recovery manoeuvre

Loss Of Control In-flight (LOC-I) is currently the largest cause of accidents for the world wide commercial jet transport fleet. To decrease the occurrence of this type of accident, procedures have been modified and pilot training has been increased. However, current training devices lack the appropriate flight models and motion capabilities to...

Objective Evaluation of Flight Simulator Motion Cueing Fidelity Through a Cybernetic Approach

Compared to aircraft, flight simulators are severely limited in their motion envelopes. Presenting the true aircraft motion one-to-one on flight simulators is generally impossible and it is therefore common practice that these motion stimuli are only presented in reduced and attenuated form. Because of a limited understanding of human motion...

Modeling Human Multimodal Perception and Control Using Genetic Maximum Likelihood Estimation

This paper presents a new method for estimating the parameters of multi-channel pilot models that is based on maximum likelihood estimation. To cope with the inherent nonlinearity of this optimization problem, the gradient-based Gauss-Newton algorithm commonly used to optimize the likelihood function in terms of output error is complemented with...