| Summary: | 博士 === 國立臺灣科技大學 === 資訊工程系 === 105 === Telerobot navigation requires human operators to have sufficient knowledge on remote situation, known as remote perception. Insufficient remote perception increases the risk of telerobot failures. The enhancement of remote perception often relates to the delivery of visual feedback in operator interfaces. Most of the recent works about operator interfaces adopted viewpoint tethering to improve situational awareness due to its exocentricity. A tethered viewpoint provides both global awareness and local guidance while visualizing remote situations. The tethered view, adhering to the line-of-sight requirement, potentialy delivers lower visual momentum when visualizing dense remote environment such as during indoor teleoperation tasks. This problem occurs as the tethered view, unlike the bird's-eye view, visualizes incomplete spatial information due to line-of-sight exclusion. Operators are thus strangled with the limited information while developing their spatial mental model, especially due to visual discontinuity that happens as the result of view transitions between robot's movements.
This dissertation presents an approach to improve the visual momentum of a tethered viewpoint display by complementing the omitted spatial information. The approach works by augmenting simplified spatial information on the excluded areas of a tethered view. The cues are used to illuminate the basic spatial structure and thus help to preserve the visual continuity on the views. The spatial cues further assist the operators in building their spatial mental models effectively by lowering their dependency on their naturally limited working memory. The tethered viewpoint displays are thus hypotheticaly posses higher visual momentum.
Evaluation of the presented approach was performed using V-REP simulator to simulate a telerobot environment that streams RGBD images, which is then reconstructed into 3D surface models and visualized in the tethered viewpoint displays. The evaluation involved eighteen voluntary participants to perform remote navigation tasks under different view configurations. The results suggested that applying cue augmentation in tethered viewpoint displyas resulted in a higher state of visual momentum. The heightened visual momentum was exhibited by the improvement of three common measures on operator performance. The improved measures include lowered average mental workload, the enhancement of both spatial perception and situational awareness. In particular, the lowered effort, frustration, and temporal demand provided the most influence to the lowered mental workload in average.
|
|---|
