2D–3D Spatial Registration for Remote Inspection of Power Substations

Abstract

Remote inspection is critical for smart factories, power systems and undersea and space exploration, among other domains. However, these applications have conflicting requirements: operators should experience high situation-awareness, implying a considerable amount of data to be presented, while having a minimal sensory load, not to compromise the time to make decisions. Recent research suggests computer vision inspection and the adoption of virtual reality (VR) as an alternative to traditional SCADA interfaces. Nevertheless, although VR may provide a good representation of a substation’s state, it lacks some real-time information, available from online field cameras and microphones. This work discusses a method to augment virtual environments of power substations with field images, enabling operators to promptly see a virtual representation of the inspected area’s surroundings. In addition, the system interacts with a SCADA database, continuously comparing the equipment states against the ones inferred by processing the field images. Whenever a discrepancy is found, a virtual camera can be teleported to the affected region, speeding up system reestablishment. Our results concern the registration accuracy and performance impact for a simple scenario. The collected metrics suggest good registration levels and low impact on real-time rendering performance.