Put AI into Practice
Explore AI funding opportunities and deep-dive into project highlights, outcomes, and lessons learned.
AI Deployment Strategies & Projects
AI Innovation Challenges
Innovation challenges are at the "proof of concept" technology phase where AI ideas are tested within a safe digital environment to confirm it is technically viable. The primary goal is to determine if the concept can solve a specific transportation challenge before an agency commits significant time or funding. The Intelligent Transportation Systems Joint Program Office (ITS JPO) undertakes a number of AI innovation challenges through the U.S. Department of Transportation (USDOT) Small Business Innovation Research Program and to meet other Departmental priorities.
AI Freight Solicitation
The USDOT Small Business Innovation Research (SBIR) FY26 Phase I solicitation invites small businesses to propose innovative AI solutions for freight transportation, improving safety, efficiency, and resilience. Opens Summer 2026.
AI Transportation Planning & Design (TPD) Project Briefs
The AI for Transportation Planning and Design (AI TPD) initiative is advancing a range of AI methods to collect, integrate and apply a range of data, particularly for active transportation. These tools will allow transportation practitioners to query complex datasets in natural language, asking questions such as "Where are the highest-risk pedestrian corridors?" and receive clear, visual responses. By combining AI analytics, geospatial visualization, and dynamic simulation, agencies can test scenarios, compare design options, and monitor performance across safety, mobility, and efficiency metrics. In Phase I, USDOT awarded $200,000 each to 12 small businesses to demonstrate novel multimodal data generation and AI processing methods in a pilot geography. Five companies are now undertaking further research and scaling their approaches to additional geographies under Phase II of this project.
AI Proof of Concepts & Prototypes
AI prototypes transition an AI solution from a purely digital setup into a functional system that is integrated into an agency's hardware and software. These tests are conducted under real-world conditions but within strictly controlled environments. This stage evaluates performance and reliability.
AI-Enhanced Predictive Analytics for Incident Management
The Incident Prediction Prototype combines several types of traffic and weather data to make reliable and proactive short-term forecasts of incident risk by road segments, offering Traffic Management Centers an analytical dashboard tool for improving incident prevention, travel alerts, and long-term roadway planning.
Coming soon
AI-Enhanced Traffic Counts
The Enhanced Mobility Traffic Counts Prototype leverages vehicle speeds, roadway characteristics, and temporal context data to forecast hourly traffic volumes for road segments lacking direct measurements. The modular, neural network-based tool offers transportation agencies a method for improving planning-level tasks such as infrastructure management, noise modeling, and congestion analysis.
Coming soon
AI-Video Analytics for Border Crossing Operations
The Border Crossing Wait Times Estimation Prototype integrates live video frames with a computer vision model to detect, classify, and track vehicles across defined start and end lines. Transportation agencies, especially those engaging with international ports of entry, may use this cost-effective, lane-specific analysis tool to generate real-time wait metrics that support operational decision-making.
Coming soon
AI Methods to Improve Categorization of Land Use Contexts
This Land Use Context Categorization Prototype automatically combines national building, census demographic, and road attribute data to assign land use context labels to road segments, offering transportation agencies an efficient, low-cost, and scalable analysis tool for context-sensitive applications such as safety analysis, multimodal planning, and emergency evacuation routing.
Coming soon
AI Pilot Deployments
AI pilot deployments move ready-to-test AI solutions into the real world on a limited, manageable scale. The focus is on observing how the technology interacts with actual conditions including users, vehicles, equipment, and freight flows while ensuring it integrates seamlessly with existing agency software.
Resources coming soon