AI’s Mild Ride: RoadBotics Puts AI on Pothole Patrol

by Lauren Finkle

National Pothole Day is Jan. 15. Its timing is no accident.

All over the Northern hemisphere, potholes are at their suspension-wrecking, spine-shaking worst this month.

Thanks to AI, one startup is working all year long to alleviate this menace. Benjamin Schmidt, president and co-founder of RoadBotics, is using the tech to pave the way to better roads.

His startup — a member of the NVIDIA Inception startup accelerator program — is identifying areas at risk of potholes, so city governments can improve roads before damage worsens.

Schmidt spoke with AI Podcast host Noah Kravitz about how RoadBotics is working with over 160 governments across the world to collect and analyze video data to improve preventative maintenance.

 Key Points From This Episode:

  • Using smartphones placed against car windshields, RoadBotics collects and analyzes video data to assign each road a score, which local governments can use to inform infrastructure decisions.
  • RoadBotics protects privacy by blurring people, cars and other sensitive data so only roads are analyzed.
  • Early this year, RoadBotics will be release an app so anyone can use smartphones to collect data and submit to their neural network to help improve analysis.


“The sooner you can detect [surface distresses], the sooner you can put a cheaper intervention in now that really just saves the life of the road.” — Benjamin Schmidt [5:00]

“RoadBotics was founded at exactly the right moment with the right tech, the right hardware. So we’re now in this sweet spot where we can actually deploy a solution” — Benjamin Schmidt [6:46]

You Might Also Like

How Deep Learning Will Reshape Our Cities

Lynn Richards, president and CEO of the Congress for New Urbanism, and Charles Marohn, president and co-founder of Strong Towns, weigh in on the benefits of using AI to design cities, and simulating designs in VR prior to construction.

How AI Will Revolutionize Driving

Danny Shapiro, senior director of automotive at NVIDIA, explains the capabilities necessary for autonomous driving, from object detection to AI to high performance computing.

Where Is Deep Learning Going Next?

Bryan Catanzaro, head of applied deep learning research at NVIDIA, explains his journey in AI from UC Berkeley, to Baidu, to NVIDIA. He’s striving for AI that works so seamlessly that users don’t even notice it, and he explains how GPUs are helping to make that happen.

Tune in to the AI Podcast

We’re available through iTunesGoogle Play MusicGoogle PodcastsCastboxCastro, DoggCatcher, OvercastPodbayPocket Casts, PodCruncher, PodKicker, SpotifyStitcher and Soundcloud. If your favorite isn’t listed here, drop us a note.

 The AI Podcast SpotifyThe AI Podcast Google Podcasts The AI Podcast Apple Podcasts

Make the AI Podcast Better

Have a few minutes to spare? Fill out this listener survey. Your answers will help us make a better podcast.

Featured image credit: Santeri Viinamäki, some rights reserved.