Tesla's fully self-driving version can recognize more common objects

There has been a lot of good news about Tesla recently. First, Tesla's stock price broke through $420, and recently Tesla's autonomous driving technology has taken another step forward.

According to foreign media reports, a Tesla owner in the United States broke the news on December 24, saying that after upgrading to the latest 2019.40.50 system, Tesla can already recognize traffic lights and parking signs (currently only available for models equipped with HW3.0 hardware), which also means that Tesla is technically capable of realizing assisted driving on urban roads.

Previously, Musk had announced that he would launch a preview version of fully autonomous driving (FSD sneak preview).

Some Tesla owners have received a preview version of the "FSD (Full Self-Driving) Kit", which shows that the electric car maker's neural network has made great improvements and can detect and recognize more common objects, such as traffic lights, trash cans, and lane markings.

Tesla owner and enthusiast Steve Hamel recorded some of the new features added to the Model 3 after receiving the latest 2019.40.50 software update and shared multiple photos.

The photos show that the neural networks on Tesla cars are now able to recognize trash cans, lane lines, arrows painted on the road, and the changing colors of traffic lights. Stop signs and road markings are also accurately displayed in recent updates. In terms of performance, Hamel said on Twitter that features such as Navigate on Autopilot have been improved.

Unlike self-driving companies such as Waymo and Cruise, Tesla does not intend to use LiDAR. Musk explained that LiDAR is useful in space, but it is quite stupid to use it on ordinary cars. In his opinion, only a set of cameras and a neural network are enough to teach a fleet of cars how to drive autonomously.

Tesla has been working to improve its Autopilot and Full Self-Driving suites through data collected by the company's fleet of vehicles. This data is then fed into the company's neural networks, which improve based on the input data.

Tesla has remained mum on the official capabilities of its neural network in the past, until a few days ago when Tesla CEO Elon Musk said on Twitter that a preview version of the "Full Self-Driving Kit" would be included in the upcoming software update version 2019.40.50.

The update adds many new features, such as Driver Profiles, Tesla Arcade Games, Trax, and Tesla Theater. But in addition to these improvements, the update also gives people a sneak peek at the fully autonomous driving function.

Of course, Tesla’s new visualizations only show the tip of the iceberg for the company’s complete self-driving solution, but the release of the latest visualizations to the company’s fleet shows how much progress Tesla has made in improving its full self-driving capabilities over the past few months.

However, seeing Tesla's electric cars recognize objects that are otherwise ubiquitous on the road shows that Autopilot and Full Self-Driving functions are on the right path to true self-driving. After all, these visible things are already impressive, and the back-end technology supporting this system may be even more advanced.