{"title":"An augmented-reality-assisted immersive system for robotic arms\n teleoperation","authors":"C. Loconsole, D. Leonardis, A. Frisoli","doi":"10.54941/ahfe1002831","DOIUrl":null,"url":null,"abstract":"In recent years Virtual Reality (VR) and Augmented Reality (AR)\n Head-Mounted Displays (HMDs) have been used to provide an immersive,\n first-person view in real-time in robotic teleoperation\n applications.However, there are two main critical issues in robotic\n teleoperation for remote manipulation and real-world interaction: firstly,\n it is challenging to correctly perceive the distance and the dimensions\n during fine manipulation tasks; secondly, latency in video streaming can\n affect the immersive experience of the operator. To address these issues, in\n this paper we propose a novel system that allows the operator to immersively\n teleoperate robotic arms using his/her hand motion and gesture.More in\n detail, the operator wears a HMD equipped with hand tracking and head\n orientation estimation technologies. The hand tracking and gesture\n recognition are used by the operator to generate control commands for remote\n robotic arms, whereas the head orientation angles are used to control the\n orientation of a remote robotic base on which is installed a RGB-Depth\n (RGB-D) camera. The latter allows the operator an improved immersive\n experience since he/she can freely rotate his/her head resulting in a\n corresponding movement of the camera on the remote side. Furthermore, the\n video/depth streaming of the RGB-D camera is used for three main objectives:\n i) appropriately filtering the RGB image through spatial depth information\n for reducing the number of significant pixels to be sent to the HMD thus\n reducing the latency; ii) providing spatial cues overlaid on the 2D image to\n facilitate remote interaction (basic AR); iii) generating additional\n augmented information with the real-world environment for specific tasks\n (advanced AR). On the remote side of the system, the robotic base of the\n RGB-D camera is integral with the base system of two robotic arms, each of\n them equipped with a robotic hand for object manipulation and\n interaction.The proposed system can be used for several tasks in several\n scenarios including social applications and high-risk situations. In the\n former scenario, an operator can remotely interact with a person by\n providing assistance or by playing board games. In the latter scenario, our\n system, suitably integrated with a robotic base allowing its mobility, makes\n possible some teleoperated activities such as door opening, valve turning\n and switch operating without local human intervention.","PeriodicalId":269162,"journal":{"name":"Proceedings of the 6th International Conference on Intelligent Human Systems Integration (IHSI 2023) Integrating People and Intelligent Systems, February 22–24, 2023, Venice, Italy","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 6th International Conference on Intelligent Human Systems Integration (IHSI 2023) Integrating People and Intelligent Systems, February 22–24, 2023, Venice, Italy","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.54941/ahfe1002831","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
In recent years Virtual Reality (VR) and Augmented Reality (AR)
Head-Mounted Displays (HMDs) have been used to provide an immersive,
first-person view in real-time in robotic teleoperation
applications.However, there are two main critical issues in robotic
teleoperation for remote manipulation and real-world interaction: firstly,
it is challenging to correctly perceive the distance and the dimensions
during fine manipulation tasks; secondly, latency in video streaming can
affect the immersive experience of the operator. To address these issues, in
this paper we propose a novel system that allows the operator to immersively
teleoperate robotic arms using his/her hand motion and gesture.More in
detail, the operator wears a HMD equipped with hand tracking and head
orientation estimation technologies. The hand tracking and gesture
recognition are used by the operator to generate control commands for remote
robotic arms, whereas the head orientation angles are used to control the
orientation of a remote robotic base on which is installed a RGB-Depth
(RGB-D) camera. The latter allows the operator an improved immersive
experience since he/she can freely rotate his/her head resulting in a
corresponding movement of the camera on the remote side. Furthermore, the
video/depth streaming of the RGB-D camera is used for three main objectives:
i) appropriately filtering the RGB image through spatial depth information
for reducing the number of significant pixels to be sent to the HMD thus
reducing the latency; ii) providing spatial cues overlaid on the 2D image to
facilitate remote interaction (basic AR); iii) generating additional
augmented information with the real-world environment for specific tasks
(advanced AR). On the remote side of the system, the robotic base of the
RGB-D camera is integral with the base system of two robotic arms, each of
them equipped with a robotic hand for object manipulation and
interaction.The proposed system can be used for several tasks in several
scenarios including social applications and high-risk situations. In the
former scenario, an operator can remotely interact with a person by
providing assistance or by playing board games. In the latter scenario, our
system, suitably integrated with a robotic base allowing its mobility, makes
possible some teleoperated activities such as door opening, valve turning
and switch operating without local human intervention.