Join us to participate in the upcoming 2019 City:One Challenge. 

Safer, Smarter Corridors - AI to prevent crashes and improve mobility

Use Derq applications to create "smart corridors" where infrastructure predicts & prevents crashes, and improves flow of traffic

Photo of Will Foss
1 0

Written by

Derq uses Artificial Intelligence in intersections to predict crashes before they happen. We then use V2X technologies to warn cars of risks with enough time to avoid crashes. We propose building a "Safer, Smarter Corridor" in Grand Rapids where sensors are setup and roadside units are equipped with Derq's safety technology. This will reduce crashes, save lives, and improve traffic flow in Grand Rapids.

Describe who will use your solution (1,000 characters)

Cities, buses, and commercial vehicles use Derq's solutions to predict and prevent crashes. This helps them reduce the number of accidents, save lives, and reduce the costs and emissions from accidents and traffic. We can deploy off-the-shelf sensors and our algorithms at several intersections to show how these technologies and solutions can save lives today.

Describe your solution's stage of development

  • Pilot - you have implemented your solution in a real-world scenario
  • Ready to Scale - you have completed and expanded your pilot and are seeing adoption of your solution by your intended user

Tell us about your team or organization (500 characters) Derq exists to use AI to predict and prevent crashes. We are a Dubai and Detroit based startup with roots at MIT. We are currently using our applications with Michigan DOT and Dubai Road Authorities. Derq's underlying technology was developed during Dr. Georges Aoude's PhD at MIT, where he worked with automotive manufacturers and American Transportation authorities as part of the prestigious CICAS project to predict crashes 2+ seconds before they happened.

Size of your team or organization

  • 2-10

Funding Request

  • $100,000

Describe how you would pilot your idea (1000 characters)

Derq develops solutions to provide external road information to increase contextual awareness for vehicles / drivers. Derq's plan is to identify the riskiest intersections for pedestrians and vehicles. Derq will then deploy its V2X AI applications with infrastructure inside those corridors or paths and at those risky locations. and then onboard vehicles and fleets operating in that geography. Our process and implementation plan is as follows: -Select intersection sensors and RSUs (if none exist) -Procure and deploy sensors -Calibrate and train sensors / Derq algorithms -Configure and deploy event detection dashboard and train municipal users -Onboard vehicles and calibrate -Train fleet operators / drivers -Operate applications and expand deployments After piloting, we would license this technology to road operators and road users on an annual basis, and would scale up the number of intersections.

Describe how you would measure the success of your pilot (1000 characters)

We would measure success and modify our future efforts by tracking the number of predicted crashes / near-misses, the amount of seconds prior to event we predicted an event, the number of safety messages sent (true positives and false positives), and the feedback from users. We would then take feedback on the speed to alert, amount of advanced warning, and human-machine interface and update future programs based on that feedback

1 comment

Join the conversation:

Photo of Tom Bulten

Will: Thank you for posting the proposal for Derq to the City of Tomorrow Challenge in Grand Rapids. I'm one of the online facilitators here. Can you estimate how many intersections and how long a corridor you could monitor with a pilot project in Grand Rapids? A point of clarification for those of us that are amateurs: V2X refers to "vehicle-to-everything" communication systems. Those interested in where crashes have been occurring in Grand Rapids can use this tool: