Self-driving cars are first and foremost being developed to increase the safety of people – both in and outside the car. One place cars and people interact the most is at a crosswalk.

Our system components include perception and behavior prediction – we classify pedestrians as a different entity than cars or motorcycles, and our autonomous software predicts that people will move more slowly and act differently than vehicles. Our software must also “know” to follow established traffic laws, yielding to people crossing in designated areas. It must recognize the different types of crosswalks, some occur at intersections or have designated signs, and some are simply marked by lines on the road.

We’ve previously shown the Torc self-driving car safely slowing down for a pedestrian crossing outside of a marked crosswalk, and it is equally important for the car to yield at designated pedestrian travel areas. In this video, you can see several non-signalized crosswalks on the road. The Torc car performs an autonomous lane change, and then slows to a stop for a pedestrian crossing. Once the pedestrian moves out of our lane of travel, the car slowly starts moving forward, then continues driving down the road.

We’re showcasing the capabilities of our autonomous car system on real public roads with real scenarios the car has encountered. To get updates on the latest videos, check back on our blog or subscribe to our YouTube channel.