Detroit Automated Driving Systems Shuttle Program—Comprehensive Safety Tests and AI Driven Data Analyses to Identify, Characterize, and Mitigate High-Risk Scenarios for Autonomous Shuttles

Document Type

Conference Proceeding

Publication Date

1-1-2026

Abstract

Under the US Department of Transportation's Autonomous Driving System (ADS) Demonstration program, the City of Detroit Office of Mobility Innovation launched an autonomous shuttle pilot project in 2024 to explore the promise of self-driving technology in helping residents facing mobility barriers get to their destinations. With active support and collaboration from the City of Detroit, the American Center for Mobility and Michigan Technological University have conducted comprehensive road safety and readiness analysis to shape the evolution of the ADS safety framework. With over 83 h of on-road data from the Detroit shuttle deployment route, which has been collected with ACM/MTU fully instrumented vehicles, the data was then processed with Driving Analytics, an AI/ML-driven data analytics platform developed by MTU/ACM, to identify edge cases and safety-critical situations. In this study, the outcomes and findings will be presented in the form of methodology for data collection, processing, identification of critical driving scenarios, and creation of safety maps to point out localized high-risk zones. The safety map approach aids in informed route planning and thereby enhances overall safety through risk reduction. The outcome of this study will also be beneficial to frame policy and legislation through measurable metrics and highlight adoption challenges to be addressed for large-scale ADS deployment.

Publication Title

International Conference on Transportation and Development 2026 Transportation Planning and Operations Selected Papers from the International Conference on Transportation and Development 2026

ISBN

[9780784487020]

Link to Full Text

Share

COinS