| Virtual environment technologies, such as helmet-mounted displays (HMDs), are challenged by problems involving time delay--the time between an input to a system, and its corresponding output. An experiment was conducted to evaluate two methods of time delay compensation--algorithmic prediction and perceptual adaptation--during a time-delayed, head-slaved tracking task using an HMD. Predictive algorithms attempt to compensate for time delays by predicting future head position in order to update images effectively in the HMD. Perceptual adaptation refers to the ability of humans to adapt to the time delay by modifying perceptual-motor coordination. Ten subjects were assigned at random to either a perceptual adaptation or algorithmic prediction condition. All subjects participated in four experimental sessions during which they attempted to center a reticle over a moving circular target using a Kaiser Electronics SimEye 2500 HMD. The practice and pre-exposure sessions featured a nominal delay of 45 msec. The third session served as an exposure session and introduced a longer time delay of 100 msec. The final session served as a post-exposure condition, again using the nominal time delay of 45 msec. Tracking performance was evaluated in terms of root mean squared error, and percentage of time on target, and the adequacy of the adaptation and prediction solutions was evaluated by several comparisons of tracking efficiency within and between sessions. Tracking performance was equal for the adaptation and prediction groups during the practice and pre-exposure sessions. The addition of time delay during the exposure session resulted in degraded tracking efficiency for both groups; however, performance was superior for the prediction group as compared to the adaptation group. In addition, both groups demonstrated improvement in tracking performance during the exposure session. A comparison between the pre- and post-exposure phases revealed increases in tracking efficiency for the prediction group, but not for the adaptation group. The results indicated that the algorithmic prediction solution is superior to the perceptual adaptation solution for compensating for the effects of time delay in a helmet-mounted display. |
