Edge Computing and Sensor Integration for Real-time Rural Road Management

Real-time access to key data from interactive environmental sensors is critical to increasing safety on rural roads and improving efficiency for municipal governments. Existing Environmental Sensor Stations and Road Weather Information Systems are expensive and complicated to build and operate. These challenges prevent governments from building or accessing scalable solutions especially in rural areas. The result is expensive labor-intensive methods for checking roads and assets before maintenance and repair operations. There is a significant need and opportunity to integrate affordable in situ traffic and environmental sensors with edge computing devices to enable real-time data processing allowing for immediate notification and decision-making. This project will develop an integrated platform that combines cellular-connected cameras with edge compute capability for image and environmental sensor processing. Custom machine learning models will allow for efficient transmission of only relevant information saving end-users time and money. The objectives of the project are to (1) identify and prioritize which environmental sensors yield highest impact in increasing safety and efficiency (atmospheric surface hydrologic); (2) integrate those sensors into the Barn Owl EdgeCam platform; (3) develop machine learning models and heuristics to analyze the data and determine action steps of notifications or automations (ex.precipitation sensor triggers image capture); (4) test the platform with a customer in a large rural environment to evaluate weatherization power/solar capabilities transmissions and user interface; (5) develop a commercialization plan for broad adoption by municipal customers and private industry. The need for this solution is great and we anticipate rapid commercialization.