We are leveraging advanced LiDAR sensing technology to enhance safety at intersections by enabling real-time detection and tracking of vehicles and pedestrians. Our research focuses on deploying infrastructure-mounted LiDAR sensors to capture high-resolution, 3D spatial data at signalized intersections. This data is processed locally to identify dynamic objects, classify road users, and monitor movement patterns within predefined zones. The processed information is then transmitted to a centralized Traffic Operations Center (TOC), where it is recorded, analyzed, and used to identify potential safety risks, particularly involving pedestrian interactions with turning or conflicting vehicles. By combining LiDAR’s precise object detection capabilities with centralized analytics, our system aims to provide transportation agencies with actionable insights for proactive safety interventions, improved signal timing strategies, and long-term planning for pedestrian-friendly infrastructure.
This research introduces a low-cost, efficient, and scalable approach for the frequent collection of roadway maintenance asset data using mobile devices. The core technology is a smartphone-based toolset that captures video and GPS data while mounted on standard DOT fleet vehicles, enabling seamless integration into routine operations. Leveraging artificial intelligence (AI), the system automatically extracts asset-related information from the recorded videos. Specifically, the AI models are designed to assess the condition of pavement, striping, and signage, as well as detect roadway debris, trash, and litter. Additionally, the research includes the development of a prototype capable of identifying and recording information on roadside barriers and guardrails, supporting more proactive and data-driven maintenance planning.