Advancing Vision Zero Road Safety with Operational AI and Smart City Data

点击翻译按钮获取中文摘要

Vision Zero is a road safety strategy aimed at achieving no road user fatalities or serious injuries with a specific focus on vulnerable road users (VRUs) such as pedestrians and cyclists. Each year in Canada, about 2,000 people are killed and 165,000 are injured (10,000 seriously), while using our road transportation system. The annual social costs of the motor vehicle collisions in terms of loss of life, medical treatment, rehabilitation, lost productivity, and property damage are measured in tens of billions of dollars - estimated at 40.7 billion (or 2.1% of GDP) for Canada in 2017.The goal of the proposed research is to advance the planning, designing and evaluation of various road safety strategies and safety countermeasures by leveraging emerging data sources and recent advances in artificial intelligence. For example, the emergence of video- and laser-based smart city sensors provides an opportunity to monitor traffic "conflicts" - situations where there is an increased risk of collision. Then, artificial intelligence can use these data to predict collisions in the future. Similarly, these data can be used to develop simulation models to test road safety projects and policies that do not yet exist in reality. Hence, the proposed research will develop state-of-the-art tools and techniques for quantitatively assessing the road safety effects of decisions made in the planning, design, operations, and maintenance of transportation systems.The proposed research is anticipated to directly enhance the Vision Zero plans developed concurrently by the partnering cities of Kitchener and Waterloo. Similarly, Miovision, a leading and partnering smart city sensor provider, is anticipated to commercialize promising developments for applications in other Canadian cities.