In what concerns human-machine interaction, two main driving forces are performance and usability. In systems where the human operator integrates the control loop (HILhuman in the loop) one wishes to minimize its effort and fatigue, without compromising the performance.The paper proposes a design approach where the influence of the operator dynamics in the closed loop is taken into account, by means of dynamic compensation. A frequency analysis based procedure is used to obtain an operator model from experimental data.The human-machine model is used to synthesize a LQG based dynamic compensator which significantly improves the performance while reducing the operator effort.

Abstract-This

No abstract

Point-to-point reaching manual actions are present in numerous human-machine tasks. This is due to the fact that machines are commonly handled by a human operator through simple multiple interfaces. Therefore, task-performance evaluation methods based on extending the Fitts' law can be used as a skill estimator for the resulting human-machine system. This paper proposes a methodology for qualifying operator skill in point-to-point (p-t-p) man-machine operations, based on several task-performance index evaluation criteria. A 2-D real-time setup was built for the execution and evaluation of the p-t-p tracking experiences over a predefined process dynamics, using a pen tablet as the human-machine interface device. The analysis from the collected data reveals a correlation between operator performance and the fitting error to Fitts' law.

Traditionally Man-Machine Interfaces (MMI) are concerned with the ergonomic aspects of the operation, often disregarding other aspects on how humans learn and use machines. The explicit use of the operator dynamics characterization for the definition of the Human-in-the-Loop control system may allow an improved performance for manual control systems. The proposed human model depends on the activity to be performed and the mechanical Man-Machine Interface. As a first approach for model development, a number of 1-D manual tracking experiments were evaluated, using an analog Joystick. A simple linear human model was obtained and used to design an improved closedloop control structure. This paper describes practical aspects of an ongoing PhD work on cognitive control in Human-Machine systems.