AI, along with advances in hardware technology, has greatly lowered the cost to captured infrastructure assets along a roadway using lidar and image capture. The more infrastructure owners are able to capture and extract using this approach, the larger their ROI on investing in the technologies. For this reason many state DOTs are for example deploying these mobile mapping systems on their road networks. Another key reason is that the technology improves safety by capturing information at highway speeds instead of putting humans next to active roadways.
In this challenge, we are asking students to take a section of roadway that was mobile mapped by WSB using the Trimble MX mobile mapping system and develop a model that would allow an infrastructure owner to extract and classify their some of their top assets -
The objective is to extract and classify four asset classes:
- Pavement: The travelled surface and its painted markings
- Utilities: Poles, overhead conductors, cabinets etc...
- Signs: Panels and the structures carrying them etc...
- Safety: Guardrails, barrier, rumble strips etc...
Event Schedule
- Kickoff — Thu, Sept 3, 4:00–6:00 PM, Student Commons 1600
- Submissions open — through Sept 11, 11:00 AM
- Demo Day / Judging — Fri, Sept 11, 11:00 AM–3:30 PM, North Classroom
- Awards Ceremony — Fri, Sept 11, 4:00–5:00 PM, North Classroom 1005
- Industry Symposium (winners showcase) — Wed, Sept 16, 9:00 AM–4:00 PM, Jake Jabs Center
The Technical Mandate: "Leverage Existing AI/ML Tools to Build an Efficient Model for Top Roadway Infrastructure Assets"
Using Trimble's Business Center Software and AI/ML tools for mobile mapping lidar/image data, students are challenged to develop their own model that extracts and classify's these four key assets. Students will have access to the mobile mapping lidar/image data and TBC tools for building their own extraction/classification model.
You will be provided about a 1 mile section of roadway data that has been mapped in Mannford OK using lidar/imagery with the Trimble MX9 mobile mapping system.
Trimble MX9 | Mobile Mapping Systems | Trimble Geospatial
Hosted by the AI Student Association at the University of Colorado Denver, in strategic partnership with Colorado Smart Cities Alliance, Trimble, and WSB, and in collaboration with Colorado State University, HDR, and SHPE.
Requirements
Each team should provide a brief presentation or written summary covering:
- Problem and scope — the asset type or workflow component you chose, and why it matters
- Proposed approach — the data, tools, methods, and intended outputs
- Evidence of exploration — diagrams, examples, experiments, screenshots, code, or prototype work
- Documented workflow — what was completed, how it was approached, and what remains
- Limitations and next steps — what you learned and how you'd continue or improve the work
Infrastructure owners needs these types of extraction/classification models to be consistent for each section of roadway. They also need the extraction and classification to provide consistent data accuracy and data completeness. Meaning does each asset extract with similar geospatial accuracy and does the classification of the asset provide the same attribution completeness from the roadway section.
Deliverables
- Extraction Model: trained model, configured workflow or code, in whatever tool you built it in. Asset Inventory: the assets you actually extracted, in the schema you defined, with the coordinate system stated explicitly
- Extraction and Classification: Procedure repeatable enough that another team could follow it on the next mile.
- Enhancement Procedure: How you’d improve it with more time, budget or data
- Presentation: Walk judges through everything above in a 10-minute presentation followed by a Q&
Prizes
1st Place
2nd Place
3rd Place
Devpost Achievements
Submitting to this hackathon could earn you:
Judges
Farnoush
Associate Professor
Paul Downing
AI Product Manager
Ed Shappell
VP - Strategy & Technology
Tyler Svitak
Executive director
Mohammed Mehany
Associate Professor
Steven Younkin
Chief Engineer
Judging Criteria
-
Rubric
Will be sent by email. Email AISA@ucdenver.edu if you haven't recieved it yet!
Questions? Email the hackathon manager
Invite others to compete
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.