Scene Modeling and Augmented Virtuality Interface for Telerobotic Satellite Servicing
| Autor: | Will Pryor, Balazs Vagvolgyi, Simon Leonard, Wenlong Niu, Peter Kazanzides, Ryan Reedy, Anton Deguet, Louis L. Whitcomb |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0209 industrial biotechnology
Control and Optimization Computer science Mechanical Engineering Interface (computing) 05 social sciences Real-time computing Biomedical Engineering ComputerApplications_COMPUTERSINOTHERSYSTEMS 02 engineering and technology Solid modeling Computer Science Applications Visualization Human-Computer Interaction 020901 industrial engineering & automation Artificial Intelligence Control and Systems Engineering Teleoperation 0501 psychology and cognitive sciences Satellite Computer Vision and Pattern Recognition Augmented virtuality 050107 human factors |
| Zdroj: | IEEE Robotics and Automation Letters. 3:4241-4248 |
| ISSN: | 2377-3774 |
| DOI: | 10.1109/lra.2018.2864358 |
| Popis: | Teleoperation in extreme environments can be hindered by limitations in telemetry and in operator perception of the remote environment. Often, the primary mode of perception is visual feedback from remote cameras, which do not always provide suitable views and are subject to telemetry delays. To address these challenges, we propose to build a model of the remote environment and provide an augmented virtuality visualization system that augments the model with projections of real camera images. The approach is demonstrated in a satellite servicing scenario, with a multisecond round-trip telemetry delay between the operator on Earth and the satellite on orbit. The scene modeling enables both virtual fixtures to assist the human operator and augmented virtuality visualization that allows the operator to teleoperate a virtual robot from a convenient virtual viewpoint, with the delayed camera images projected onto the three-dimensional model. Experiments on a ground-based telerobotic platform, with software-created telemetry delays, indicate that the proposed method leads to better teleoperation performance with 30% better blade alignment and 50% reduction in task execution time compared to the baseline case where visualization is restricted to the available camera views. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|