Tesla’s Traffic Light and Stop Sign Control is arguably the company’s most advanced driver-assist feature yet. But if Elon Musk’s recent comments are any indication, the function is actually more capable than its current functions would suggest.
Musk covered several key developments at Tesla during his opening remarks in the recently-held Q1 2020 earnings call. The CEO discussed Autopilot developments at length, particularly the release of traffic Light and Stop Sign Control. Interestingly enough, Musk stated that Tesla is only exposing a part of its vehicles’ true capabilities today, just to be on the safe side.
“Regarding Autopilot, we released a new software update for traffic lights and stop signs to early access users in March and to all US customers with full self-driving package just last week. Our cars will now automatically stop at each stop sign or traffic light until the driver gets a confirmation to proceed. I should say that the car is actually capable of much more than this, but we are only exposing functionality that we feel quite good about and where we feel it is probably a safety improvement,” Musk said.
Musk did not provide more details about the other capabilities that Teslas already have today. That being said, his comments may suggest that Tesla is making headway in the development of inner-city Full Self-Driving features. Both inner-city driving and traffic light control are crucial parts of the electric car maker’s FSD suite, which, in turn, is a key component needed to introduce Elon Musk’s Robotaxi program. As noted by Tesla AI Director Andrej Karpathy, the development of FSD functions is really a matter of neural net training. Fortunately, Musk stated that Tesla’s neural network training is moving along very well.
“We are collecting data from over 1 million intersections every month at this point. This number will grow exponentially as more people get the update and as more people start driving again. Soon, we will be collecting data from over 1 billion intersections per month. All of those confirmations are training on neural net, essentially, the driver when driving and taking action is effectively labeling — the labeling reality as they drive, and making the neural net better and better. I think this is an advantage that no one else has, and we’re quite literally orders of magnitude more than everyone else combined,” Musk said.
A similar strategy was adopted by Tesla when it released Navigate on Autopilot. In its initial release, Navigate on Autopilot required drivers to confirm lane changes by using their vehicles’ turn signals. But as the data was compiled and used to train the neural network, the feature was refined until it reached a point where it could safely take lane changes on its own. Tesla then released Navigate on Autopilot with automatic lane changes, which remains as one of the company’s most advanced driver-assist features today. As noted by Elon Musk in the recently-held earnings call, Tesla intends to do the same with Traffic Light and Stop Sign Control.
“So, all those confirmations are training on neural net, and soon cars will be able to drive through an intersection without a confirmation, as well as to make turns. And we really feel we’re — I feel extremely confident that it will be possible to do a drive from your home to your office, and most of the time with no interventions by the end of the year,” Musk said.