Effect of Jerk and Acceleration on the Perception of Motion Strength

Abstract

In a flight simulator, the calculated aircraft motions are scaled down and filtered to fit within the envelope of the simulator motion system. A number of recent flight and ground simulation studies have reported that the simulator motion was too strong, when in fact, the motion was scaled down and filtered. This paper puts forth the hypothesis that this could be due in part to the motion drive algorithm and vehicle model exaggerating the jerk. To test the plausibility of this hypothesis a paired-comparison experiment was run to determine if the subjective impression of motion strength is a function of both the acceleration and jerk of the motion. The experiment found that the level of jerk and acceleration contributed to the perceived strength of motion, with larger jerks and accelerations leading to increased motion strength. In addition, the duration of the acceleration had a significant effect on the perceived motion strength, with longer durations leading to increased motion strength. Although the relationship between jerk and motion strength suggests that exaggerated jerk in the simulator could lead to the preference for scale factors less than one, the strength of the relationship strongly suggests that it does not entirely account for the preference. Nomenclature A i = nominal acceleration of motion condition i, m=s 2 A nom = acceleration used to normalize the linear model, m=s 2 a i = total score of motion condition i (number of times selected) D i = length of constant-acceleration phase of motion, s D n = score variable for model testing J i = nominal jerk level of motion condition i, m=s 3 J nom = jerk level used to normalize the linear model, m=s 3 l = actuator length command, m n = total number of comparisons between two conditions p = probability of null hypothesis S x = simulator fore-aft displacement command, m s = Laplace variable T J = duration of the constant jerk section of the motion condition, s T s = duration of half of the motion condition, s t = number of motion conditions V x = motion-based fore-aft velocity, m=s x c = commanded X acceleration x i = intermediate X ij = total number of times that condition i was preferred over condition j A = perceived motion strength change due to a change in acceleration A = normalized perceived motion strength change due to a change in acceleration D = perceived motion strength change due to a change in the duration of constant acceleration i = perceived motion strength of motion condition i, no order effect i = perceived motion strength of motion condition i, order effect J = perceived motion strength change due to a change in jerk J= normalized perceived motion strength change due to a change in jerk = model order effect = simulator pitch command, rad i = preference probability for condition i ij = probability that condition i is selected over condition j, no order effect ij = probability that condition i is selected over condition j, when i is the first condition = standard deviation 2 = 2 distribution ! = angular jerk, time derivative of angular ac...

Perception Model Analysis of Flight Simulator Motion for a Decrab Maneuver

“…This phenomenon is not found for angular motion [11,27]. In previous studies the dynamics of the washout filter may have had an unknown contribution to this effect [28]. However, in the present study we can be sure that it is the amplitude, rather than the temporal behavior of the linear motion cue that was overestimated.…”

Human discrimination of translational accelerations

“…This finding implies that the sensory system dealing with the perception of whole-body translational movement is sensitive to a combination of the acceleration and rate of change of acceleration (jerk) of the motion stimulus (Gundry 1978;Benson et al 1986). This sensitivity to jerk was also observed in supra-threshold motion amplitudes by Grant and Haycock (2008). In order to further investigate this phenomenon, this study also includes estimation of difference thresholds at different frequencies.…”

Optimising the roll-sway motion cues available from a short stroke hexapod motion platform

“…This phenomenon has not been reported in any of the previous studies and may be a particular issue with short-stroke electric motion systems, which have the potential to react much faster to motion commands than larger hydraulic or hybrid systems. The higher acceleration onset capability, or 'jerk' rate, of modern electric motion systems can sometimes be perceived as too 'strong' or 'sharp' (24) . One method of reducing this effect is, of course, to increase the corresponding sway motion travel to compensate for false sway cues induced by roll motion, particularly those induced by transient roll accelerations.…”