Hit enter to search or ESC to close

Watch a Self-Driving Vehicle Negotiate with Oncoming Cars on a Narrow Road

An Argo AI test vehicle on Wednesday October 14, 2020 in Pittsburgh, Pennsylvania. (Photo by Jared Wickerham/For Argo AI)

Toggle Filter Camera

An Argo AI test vehicle on Wednesday October 14, 2020 in Pittsburgh, Pennsylvania. (Photo by Jared Wickerham/For Argo AI)

Driving down a narrow, two-way road can be a lot like dancing — you either pull over or you take the lead. For self-driving vehicles, the same analogy holds true; these kinds of road negotiations require the self-driving system (SDS) to interpret the intent of other drivers, clearly communicate its own intentions, and identify gaps where it can safely pull over for oncoming traffic.

To handle these narrow road situations in the most natural way possible — what engineers call “naturalistic driving” — a self-driving system must understand and smoothly react to the give and take of social interaction. This involves interpreting oncoming drivers’ intentions and anticipating how the SDS’ actions may affect the actions of those drivers. All of these factors must be computed in fractions of a second.

If the SDS is designed to be too cautious — for example, on a narrow road, requiring the vehicle to avoid any area typically travelled by oncoming vehicles — then it will remain stopped forever. On the other hand, if the SDS is designed to exercise too much confidence — for instance, speeding up to take the right of way with the expectation that the other driver will pull over to let the self-driving vehicle pass — that approach may assume too much about the anticipated behavior of other vehicles.

In the video above, you can see an Argo self-driving vehicle navigating a narrow, two-way street in Pittsburgh. Watch how it responds with two separate, opposite actions.

  • In the first instance, at 0:06, the SDS detects an oncoming car and stops to allow the car to pass. This action is proper because there are no gaps in the parked cars at right that are wide enough for the self-driving vehicle to safely enter. 
  • Once the oncoming car passes at 0:12, the self-driving vehicle begins to inch forward before pausing again when another oncoming car is detected.
  • At this point in the video, the SDS responds differently than in the previous encounter. Here, at 0:15, it detects the other driver stopping completely, giving right of way to the self-driving vehicle. 
  • The SDS computes the other driver’s action, detects a wide-enough gap between the parked cars immediately ahead, and proceeds to drive safely forward into the “pocket” of space left open by the oncoming driver. 
  • It steers gently into the pocket in order to pass the other car, then re-centers itself on the street to proceed in its intended direction.

Taken together, the actions captured in this video demonstrate the sophistication and nuance needed in order to drive naturalistically. In one instance, it stopped. In another, it took the lead. It’s a dance that Argo self-driving vehicles are rapidly learning to master.

Choose your lane

How Autonomous Vehicles Distinguish Between Bicycles and People Who Ride Them

How Autonomous Vehicles Distinguish Between Bikes and People

When it comes to how autonomous vehicles see the world, humans come first, literally. Autonomous vehicles (AVs), like the kind operated by Pittsburgh-based Argo AI, use Machine Learning to detect and classify the objects in their surroundings, identifying people...
Why The League of American Bicyclists is optimistic about autonomous vehicles

Why a Leading Cycling Advocacy Group Is Optimistic About Autonomous Vehicles

As autonomous vehicle use grows, AV companies and the League of American Bicyclists are collaborating on how to ensure cyclists and motorists can share the roads safely, even if the “motorist” is artificial intelligence software. As part of the...

Self-Driving Is Arriving Right On Time. Just Like Ice Cream Did

Seven years ago, I was a self-driving skeptic. Not of the technology. Of all the “experts” promising autonomous vehicles would be everywhere by 2020. You didn’t need to be Nostradamus to know that was ridiculous. All you needed was...
Illustration of a futuristic parking deck turned into a mixed-use space, with AVs driving by

How Autonomous Vehicles Could Help Transform Parking Lots

Researchers say it’s likely that autonomous vehicles (AVs) can help reduce the need for parking lots, opening more room for grass and trees and other elements of nature. It may not seem like it when you’re circling the block...
An illustration of an Argo autonomous vehicle in teal against a backdrop of buildings, a bicyclist, and research papers

7 Big Breakthroughs From Argo Research at CVPR 2022

The 2022 Conference on Computer Vision and Pattern Recognition (CVPR 2022) is nearly here. Thousands of computer scientists, software engineers, and researchers from around the globe will gather in New Orleans to review and discuss their latest work in...

Researchers Predict the Future With Lidar Data

Researchers at Carnegie Mellon University’s Argo AI Center for Autonomous Vehicle Research, a private-public partnership funded by Argo for advancing the autonomous-vehicle (AV) field, say they have come up with a way to use lidar data to visualize not...

Must Reads