Motion scaling on one- and two-axis compensatory control tasks

Abstract

One-and two-axis compensatory control tasks were performed both with and without motion inputs. Additional tests were made in which the motion input, compared with the visual input, was reduced in scale. L i t t l e o r no difference in error measurements was observed in the single-axis-motion/no-motion tests. However, a decrease in normalized tracking error and an increase in the closed-loop system frequency were observed when motion was added to the two-axis tests. In the tests with reduced motion scale, the normalized tracking error remained constant for full, 1/2, and 1/4 motion scaling but increased with a further reduction in motion scaling. In general, the results show that motion may o r may not be an aid in controlling a compensatory tracking task, depending on the difficulty of the task and on the requirements of the mission. Also it is shown that for cases in which motion is beneficial, it is not always necessary to supply full-scale motion cues.

Dual-Axis Manual Control: Performance Degradation, Axis Asymmetry, Crossfeed, and Intermittency

“…Performance degradation is also reflected in the decreased crossover frequency. Whereas some studies suggest a degradation ratio in relation to the corresponding single axis ω c such as 1/ √ 2 [10], such a consistent ratio has not been found here.…”

“…1) Performance degrades in dual-axis tasks compared to the independent single axis tasks, indicated through an increase in error variance, a smaller crossover frequency, and a smaller phase margin [1], [7], [17]. 2) Performance degradation in dual axis is smaller with physical motion [10], which can be indicated by a decrease in error variance, an increase in crossover frequency and an increase in phase margin. 3) Asymmetrical human control behavior in each axis can be seen from unequal HC parameters in each axis, as well as from differences in error variance, control variance, crossover frequency, and phase margin between the roll and pitch axis [9], [1].…”

“…Intermittency is present in dual axis [21], which can be indicated through a consistently lower VAF, and consistently more occurrences of peaks. 6) Intermittency in dual axis is mitigated with motion [10], which can be indicated by a decrease in peak time and an increase in VAF.…”

“…Furthermore, a focus on one axis or a consistent prioritization has been observed [9], an effect referred to as axis asymmetry. While some studies have postulated that such asymmetry may be explained by a systematic reduction in operator aggressiveness (reduced crossover frequency) compared to the single-axis case [10], [11], others have proposed that the characterization of multi-axis control should include task interference phenomena, such as those resulting from divided attention (e.g., switching between axes), prioritization between axes-axis asymmetry [9] and time varying axis prioritization-intermittency [1]. A number of earlier investigations [1], [5], [7], have proposed to analyze and model crossfeed between axes, which occurs when operators are unable to fully decouple their separate tasks.…”

“…When one axis has a larger error than the other, this axis can be (temporarily) prioritized over the other [1], [9]. Intermittent behavior has been found to be smaller with the use of motion [10], thereby producing more consistent, linear timeinvariant pilot behavior. Intermittency is traditionally lumped up with pilot remnant [2], and has been found to proportionally increase with each additional axis used [6].…”