Identification of multimodal control behavior in pursuit tracking tasks

In manual control, a pursuit display may support the use of a multimodal “pursuit” control strategy by the human operator. This paper evaluates two methods that may be used to directly estimate describing functions for such multimodal human operator control dynamics in pursuit tracking. The first is a previously developed frequency-domain method based on Fourier coefficients. The second method makes use of linear-time invariant ARMAX models. An experiment is described in which participants performed tracking tasks with quasi-random multisine target and disturbance forcing functions, for single and double integrator controlled elements and with compensatory and pursuit displays. The experiment data confirms the findings from previous experiments, where it was found that multimodal pursuit control dynamics are adopted in control of systems with double integrator dynamics, but not for single integrator control tasks. Furthermore, the direct multimodal identification methods were found to give improved insight into the internal organization and dynamics of the human operator in pursuit tracking.