AR Sensor Overlay for Machine Identification in Processing Plants
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In processing plants, technicians face difficulties in identifying specific machinery that requires servicing due to the lack of integration between sensor and diagnostic data with real-world scene data, especially when accessing physical machinery.
Innovation Solution
An augmented reality interface system that uses sensors with machine-readable indicia and a camera to capture imagery, process sensor data, and generate an augmented display overlaying data visualizations onto the real-world scene, allowing technicians to quickly identify machines associated with aberrant readings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a central control unit with graphical user interface is used to manage sensor data, then data acquisition and control capabilities are improved, but technician access to dynamic display and management of data is lost when physically accessing machinery
Solution Approach 1:
The patent introduces an augmented reality interface as an intermediary between the central control system and the technician. The AR interface captures the physical scene through a camera, overlays sensor data visualizations onto the viewed machinery, and provides touch-activated information retrieval. This mediator enables technicians to access real-time sensor data and diagnostic information directly at the machinery location without needing to return to the central control room, thus resolving the contradiction between centralized data management and field accessibility.
2Productivity
If schematic display of data is used for high-level process control, then overall process management is improved, but identification of particular physical machines becomes difficult
Solution Approach 1:
The patent merges the schematic data visualization with the real-world visual scene by overlaying sensor data graphs, readings, and diagnostic information directly onto the image of the physical machinery captured by the camera. This combination allows technicians to see both the actual machine and its corresponding data simultaneously in the same field of view, eliminating the need to switch between schematic displays and physical inspection, thus resolving the contradiction between high-level process control and specific machine identification.
Solution Approach 2:
The patent transitions from two-dimensional schematic displays to a three-dimensional augmented reality experience where data visualizations are superimposed onto the spatial context of the actual machinery. By adding the dimension of spatial overlay, the system maintains the data-rich presentation of schematic displays while anchoring the information to the specific physical location and appearance of each machine, thereby solving the identification difficulty while preserving process control efficiency.
3Reliability
If multiple sensors are deployed across processing plants, then monitoring coverage is improved, but complexity of identifying which sensor corresponds to which machine increases
Solution Approach 1:
The patent employs color-coded data visualizations and annotations that change or highlight based on sensor status and machine identity. Different sensors and their corresponding machines can be distinguished through color-coding schemes, making it easier for technicians to quickly identify which sensor readings belong to which physical machine. This visual differentiation reduces the cognitive load and complexity of matching multiple sensors to multiple machines while maintaining comprehensive monitoring coverage.
Data Source
AI summary
System, method, and media for an augmented reality interface for sensor applications. Machines making up a particular production or processing facility are instrumented with one or more sensors for monitoring their operation and status and labeled with machine-readable tags. When viewed by a technician through an augmented reality display, the machine-readable tags can be recognized using a computer-vision system and the associated machines can then be annotated with the relevant sensor and diagnostic data. The sensors may further form a mesh network in communication with a head-mounted display for an augmented reality system, eliminating the needs for centralized networking connections.