New York City is a crowded place. But, it is also hard on Tesla's 'Full-Self Driving' (FSD) software.
If you are driving in Manhattan, it might seem tricky. But, it is not as tough as it appears. However, if you are relying on Tesla's FSD software, it will become a challenge for you to navigate the city.
Previously, we saw FSD Teslas trying to navigate Manhattan. A new video shows the progress the FSD software has made but also highlights how much more it needs to improve. The FSD handles much of the city reasonably well. But it is a bit slow. The software seems to be struggling with certain aspects.
Tesla’s FSD car seems to be particularly confused by two things: taxis and pedestrians.
In the video, Tesla car struggles to understand the movements of taxis. A human can easily predict a taxi's actions. You see cabs, people on the sidewalk, and hazard lights.
On a crowded road, you know when to change lanes and go around. However, the Tesla does not seem advanced enough to be able to piece these clues together correctly.
Crosswalks are another challenge for Tesla cars. New Yorkers stand just off the sidewalk while waiting to cross streets. But the car interprets it as people actively crossing the road. It backs off to give them space to move. But it disrupts the flow for both drivers and pedestrians.
It seems the Tesla FSD still has a lot to learn to cope with the challenges of urban driving, especially in bustling cities like New York.