Crowd-AI Sensing Based Traffic Analysis for Ho Chi Minh City Planning Simulation

This activity is in response to NSF Dear Colleague Letter Supporting Transition of Research into Cities through the US ASEAN (Association of Southeast Asian Nations Cities) Smart Cities Partnership in collaboration with NSF and the US State Department. Ho Chi Minh City (HCMC), an ASEAN city in Vietnam, is well-known for its traffic congestion and high density of vehicles, cars, buses, trucks, and a swarm of motorbikes (7.3 million motorbikes for more than 8.4 million residents) that overwhelm city streets. Large-scale development projects have exacerbated urban conditions, making traffic congestion more severe. Additionally, traffic congestion is one of the leading contributors to noise and dust pollution in the city. Altogether, traffic congestion poses major barriers to urban quality of life, but the solutions are complex. There are two main problems with traffic in HCMC. First, HCMC, like other dense urban areas, needs significant financial and technical resources to solve its traffic and infrastructure problems. Second, given that traffic monitoring is carried out by a limited number of staff who watch traffic activities from thousands of camera feeds on multiple screens, there are limits to the number and effectiveness of responses that personnel are able to offer in response to real-time traffic problems.

The goal of this project is to use visual crowd AI sensing for the HCMC planning simulator. The project will make use of the city camera system (crowd-AI sensing) for traffic analysis in real-time. It seeks to detect “anomaly events” such as traffic violations, traffic jams, and accidents, with reduced intervention from monitoring staff, allowing staff, in turn, to better respond to traffic problems as they arise. A city planning simulator will be developed upon the analyzed traffic data. The simulator will be used to support metropolitan transportation planning. Project findings will not only address specific urban challenges and the innovative technical solutions needed to solve them in HCMC, but also will provide models use in other contexts, including U.S. cities where traffic, congestion, and urban infrastructure challenges can benefit from AI.

The project will be validated by professionals in HCMC who can evaluate its effectiveness for detecting anomaly events with reduced human observation, who are better able to respond to traffic problems as a result of implementing aspects of the project, and who can make use of the project data for traffic analysis.