Our platform is able to model the position, speed and classification of vehicles in real-time, to intelligently detect and validate Events on the road and to predict times and locations with an elevated accident risk. We utilise Machine Vision to identify both static and moving objects, weather and light conditions and to automatically verify alerts reported from open and floating-car sources. We use Natural Language Processing (NLP) to make sense of social media feeds. In combination, these methods enable rapid, data-driven decisions, reduce human error and save time, emissions and lives.
Information about the condition of the road comes in all shapes and forms — cameras, weather reports, social media, radar loops, IoT sensors, navigation apps and connected vehicles. Inherently, any individual data source has its limitations, uncertainties and blind spots. However, when combined, they tell a comprehensive story about what’s happening on the road. Data Fusion allows road network operators to leverage this certainty. Using custom knowledge graphs and optimal estimation, we assimilate data, measurements, information and their associated uncertainties from a multitude of sources and derive real-time and actionable insights about the road and its users.
Our scalable, event-driven architecture enables performant, cost-effective and reliable data ingestion, processing and storage. Our system can stream unlimited cameras at 30 frames per second straight to the Cloud, as well as consuming Open-Data and Sensor signals to power algorithms for accident detection, verification and automated response. What’s more, we will back-up all 2.5 billion frames every day and can replay conditions from any historical date. We continuously run pattern detection algorithms on this data to alert Operators to present traffic anomalies. Our objective is to remove the overhead of high-volume, high-velocity data in order for its value to be realised.