- 408-924-7560
- mineta-institute@sjsu.edu
- Donate
Artificial Intelligence for Pedestrian and Bicyclist Safety: Using AI to Detect and Report Near-Miss Collisions
According to the latest reports by the US Department of Transportation, the Insurance Institute for Highway Safety (IIHS), and the National Highway Traffic Safety Administration (NHTSA), the number of traffic fatalities has significantly increased in 2021 and reached a 16-year high. According to the reports, 43% of the victims were pedestrians and cyclists. Understanding the movement of people, bicycles, and their interaction with vehicles is critical to avoid traffic accidents and improve safety. Currently, there is no efficient automated system in the state of California for detecting and predicting pedestrians/bicyclists collisions risks in major urban areas.
According to the Occupational Safety and Health Administration (OSHA), a “Near-Miss” collision is an incident where, given a slight shift in time or position, serious damage or injury easily could have occurred. Near-miss collisions are dangerous traffic situations that are rarely reported to the authorities. Nevertheless, they are important indicators to identify potential risks and prevent actual collisions with personal injury or property damage in future. Fortunately, with an effective detection and reporting system we will be able to take advantage of near-miss collisions to help prevent actual collisions in future.
The main goal of this project is to design and develop an automated system based on advanced artificial intelligence (AI) and computer vision to improve the safety of pedestrians and bicyclists. In this project, we develop automated AI-powered systems to monitor, measure, and control the traffic flow particularly at intersections, and detect and identify Near-Miss Collisions for pedestrians and bicyclists as an important indicator to recognize and measure actual risks. This project proposes an end-to-end system including a series of image/video processing, computer vision algorithms, machine learning and deep learning, and optimal state estimator and tracking algorithms. The information generated by the proposed system allows us to improve safety measures for pedestrians and bicyclists as well as optimizing the flow of traffic and travel time.
Principal Investigator:
Mohammad Pourhomayoun, Ph.D.
PI Contact Information:
California State University Los Angeles
Dates:
June 2023 to May 2024
Implementation of Research Outcomes:
In this project, we will design and develop novel algorithms and models and develop AI-powered systems to monitor, measure, and control the traffic flow particularly at intersections, and detect and identify Near-Miss Collisions for pedestrians and bicyclists as an important indicator to recognize and measure actual risks. This project outlines a comprehensive system that encompasses various components, including image and video processing algorithms, computer vision algorithms, machine learning models, deep learning models, and state-of-the-art state estimation and tracking algorithms. The data generated by this integrated system not only enhances safety measures for pedestrians and cyclists but also facilitates the optimization of traffic flow and travel time.
Impacts/Benefits of Implementation:
Improving the safety of pedestrians and bicyclists has always been one of the priorities of transportation officials in California. Two of the main goals of Pedestrian and Bicycle Safety Branch under Caltrans’ Traffic Operations are developing programs to improve the safety of transportation infrastructures for pedestrians and bicycles; and encouraging research and technology transfer in the field of pedestrian and bicycle fatality.
The proposed research project will facilitate these goals by providing a platform to enhance and improve safety measures for pedestrians and bicyclists in California. Vision Zero, a national and state initiative to reduce and eliminate traffic fatalities, has been adopted by a number of cities and counties in California. This project will help the transportation authorities to enhance the advancement of Vision Zero implementation across the state, which directly benefits Californians, and support the state’s effort to reduce traffic fatalities to zero. Through innovative research and practical solutions, it seeks to create safer roadways, encourage active transportation choices, and reduce the risks associated with walking and cycling. By leveraging the research outcomes, we can not only enhance safety but also improve the overall quality of life for communities across California. This project is a testament to our dedication to creating a safer and more sustainable future for the state's residents and aligning our efforts with the ambitious vision of Vision Zero.
Project Number:
2350
-
Contact Us
SJSU Research Foundation 210 N. 4th Street, 4th Floor, San Jose, CA 95112 Phone: 408-924-7560 Email: mineta-institute@sjsu.edu
