Raven Connected partners with uOttawa-Kanata North Innovation Centre paving the way for industry-shaping AI-enhanced vehicle safety
Bridging the gap between research and industry, Raven Connected joins forces with the University of Ottawa on a new project to accelerate development of an AI/ML framework to bring the latest safety and connected vehicle features to market.
May 19, 2022
Raven Connected announced today that it has partnered with the University of Ottawa’s Smart Connected Vehicles Innovation Centre (SCVIC). The joint initiative will include the creation of an Artificial Intelligence (AI) and Machine Learning (ML) framework to allow for the rapid deployment of Raven Edge AI and Raven Cloud AI based image and sensor analysis for safety critical and productivity applications. Integrated within the Raven video telematics platform, this framework and associated models will facilitate detection, alerts and reports to achieve optimal outcomes for fleet operations and other stakeholders.
Employing an array of novel and best in class technologies, including TensorFlow and AWS SageMaker, a primary goal of the project is to improve driving behaviors and reduce insurance claims, while avoiding personal injury and loss of use costs for small and medium businesses.
Distracted driving contributes to more than 20% of crashes in provinces across Canada. Therefore, the first model being developed focuses on mobile device use by a driver while the vehicle is in motion. Real-time alerts will be triggered on the edge, helping drivers to maintain their attention on the road and avoid costly fines.
“One of the biggest challenges in this field of research is access to good data.” explains uOttawa Faculty of Engineering Professor Burak Kantarci who leads the SCVIC. “Through our partnership with Raven, we can leverage their sample datasets and real world experience to test and refine the models.”
Mitigation of false positive and false negative detections is a key driver for the project. The University of Ottawa is using seed data to construct the models which will be automatically enhanced by way of continuous feedback from users of the Raven platform. Already the team is seeing accuracy of event detections in the high 90th percentile.
“Our goal is to make this technology accessible to smaller businesses which represent 80% of the fleet market.” noted Dan Carruthers, CEO of Raven Connected. “The collaboration with the SCVIC research team and their diverse experience in the areas of wireless networks, IoT, mobile computing, sensor fusion, AI and machine/deep learning for mobile systems will accelerate our time to market and ultimate performance of the system.”
The framework the teams are creating is a foundation to then rapidly train and deploy new models to the Raven Edge and Raven Cloud. This framework will also create the opportunity for road facing object detection and categorization for the benefit of Smart City applications. Tagged with location, accelerometer and other sensor metadata, this information becomes highly valuable for road condition monitoring and other municipal asset management use cases.
For inquiries: Natalie Jackson, email@example.com