Hacker finds Tesla is working on a neighborhood-friendly Autopilot

Tesla’s Autopilot suite of semiautonomous technology is a work in progress, and the company is putting a lot of effort into making it better and smarter every year. Autopilot-equipped Tesla models are about to learn how to recognize stop signs and traffic lights, according to a hacker known as Green who cracks open the automaker’s secret files as a hobby.

Posting on Twitter, Green explained Tesla expanded its repertoire of 3D assets with a stop sign on a pole, and several traffic lights. If you need a brief refresher course, the 3D assets are used to show the driver what the car is doing while it’s traveling on Autopilot mode. For example, if your Model S is in a construction zone, the hardware that powers the system detects traffic cones, and the software displays them on the instrument cluster. The technology shows lane markings, too.

The 3D renderings are ways for Tesla to quell anxiety surrounding the use of semi-autonomous technology, like Autopilot. By seeing their surroundings on a screen, motorists know the car is perfectly aware of the environment it’s operating in, and it will behave accordingly. The software is capable of telling the difference between a pedestrian and an Audi RS 5.

While Tesla hasn’t commented on the report, Green’s findings suggest engineers are adding more functions to Autopilot in a bid to improve how it navigates urban environments. Making semiautonomous technology for highway use is relatively easy; the car needs to stay in its lane, it needs to maintain a constant speed, and it needs to brake when the vehicle in front of it slows down. It’s like combining lane-keeping assist, cruise control, and collision avoidance. Making it suitable for urban environments is more difficult, however, because intersections can be tricky to navigate, and a number of other factors come into play, like kids running across the road. Waymo has mostly figured it out, but it’s ahead of the pack.

Editors' Recommendations