The annual Automated Vehicles Symposium was held again this week in San Francisco. Attendance continues to grow in this TRB sponsored event, topping more than 1500 manufacturers, academics, and other practitioners.
Breakout Session #18 was titled, “Reading the Road Ahead: Infrastructure Readiness.” As the name implies, this session focused on our roads and what must be done to prepare them for autonomous vehicles.
The first section looked at the State of Machine Vision Systems. Both speakers: Jaap Vreeswijk of MAP Traffic Management and Tom Alkim of the Dutch Road Authority talked about results in Europe. They both pointed out road conditions vary from facility to facility and even from one section to another. And for that reason they suggested that a system should be developed to tell drivers what level of automation is supported as conditions change.
This would also apply to temporary changes such as work zones. As autonomous vehicles approach a work zone the driver would be told to take control as a work zone is just ahead. Once through the work zone, the driver could return control to the vehicle so long as that road segment supported it.
The actual mechanics for this process was not discussed. Digital maps could notify us of upcoming changes in the roadway requiring more or less human control. But it could only do so for work zones under 2 possible scenarios: 1) if a local device was placed in advance of the work zone to send out a signal to approaching traffic, or 2) if maps are continuously updated with real-time information…in other words if they have perfect knowledge of all work zones.
That was the subject discussed later in a session developed by Ross Sheckler of iCone. I was asked to deliver that presentation and it was well received. Ross pointed out both the need to place work zones on the digital map, and the fact it can be done today. We are already sharing real time work zone locations and related data with services including Waze and HERE. But we could do so much more if we can just get more sensors out into devices like arrow boards or flagger paddles.
Once we have enough of these in service, motorists and traffic management centers will both come to depend on this accurate, real-time data. And safety will be much improved when even utility workers appear automatically on the map when they turn on their flashing lights.