Deterministic Networking Empowered Robotic Teleoperation

Abstract

Robotic teleoperation has seen widespread adoption across industries. Advances in technologies like the Internet of Things (IoT), Artificial Intelligence (AI), and Extended Reality (XR) are making teleoperation more flexible. The integration of visual, audio, tactile, and immersive interfaces enhances situational awareness, enabling teleoperators to interact effectively with complex remote environments and make informed decisions. However, challenges persist, particularly in environments characterized by constrained network resources. The limited bandwidth, delays, and intermittent communication can disrupt the teleoperator’s interaction with the robot. This study aims to comprehensively understand the scenarios of industrial ground robotic teleoperation and the intricacies of its network to effectively enhance teleoperation performance. Initially, we introduce the multimodal perception-enhanced robotic teleoperation, accompanied by an analysis of the Quality of Service (QoS) of all involved streams. Subsequently, we introduce a Deterministic Robotic Teleoperation (Det-RT) system, along with a deterministic traffic flow scheduling framework designed for real-time remote environment perception. Finally, we evaluate the proposed scheduling solution in a simulated environment to assess its performance within the Det-RT system. The results obtained demonstrate the capability of our solution to deliver high-quality teleoperation performance.