Connect with us

Electric Cars

What the Tesla Fully Featured Autopilot Beta Can Do



On October 20, Tesla, as you know, began beta testing the next major Full Self-Driving (FSD) update, which should pump the proprietary Autopilot system from the current second to the highest fifth level, thereby providing the company’s cars with the long-promised ability to move independently without any or driver assistance. While the new Autopilot is available only to individual Tesla electric car owners in the US, you can already find a lot of rave reviews and videos demonstrating the system’s capabilities on social networks.

Recall that the hardware of Autopilot, in addition to the FSD computer, which is responsible for processing all the data, includes eight cameras that create a panoramic picture around the car (here is a 360-degree video of the trip “through the eyes of the autopilot” of a Tesla electric car), 12 ultrasonic sensors that calculate the distance to obstacles, and a radar installed in the bow. Since 2016, all Tesla vehicles, including the junior Model 3, have been equipped with the above components. Earlier, the head of Tesla, Elon Musk, repeatedly said that lidars (3D laser scanners that detect objects and determine the distance to them) are an expensive and unnecessary technology for autonomous cars. Unlike Waymo and others, which rely on lidars and HD mapping for this, Tesla relies on computer vision. We talked about this in more detail here. In short: Tesla’s system is easier to scale to a large fleet of vehicles, while the mapping approach, which requires constant tracking of the smallest changes, is expensive and “non-scalable”.

We have selected several videos showing how detailed Tesla’s new autopilot “sees” everything around (visible on the center display) and reacts to various driving situations. The videos demonstrate the operation of the new generation system in the daytime and at night – in these videos you can see how Tesla cars independently move along expressways and roads within the city, automatically slowing down and accelerating, changing from one lane to another to make a turn / exit and pass intersections stopping at a red traffic light. And although in some cases the system can still make mistakes (for example, as in this situation), in general it performs well enough. Which is impressive considering it’s beta. Actually, a few illustrative examples:


By the way, from October 27, the Full Self-Driving package will rise in price again. The last time the price increased in July by $ 1000 (up to $ 8000). This time the rise in price will be 25% ($ 2000) – so far only for US residents. As the beta spreads to other countries, a similar increase will occur there (at the same weekly interval).

A broader beta test of the new autopilot is due to kick off in the coming weeks, with a final launch in the US slated for mid-December.

Recent leaks indicated that Tesla is developing a processor for the next-generation HW 4.0 platform, and is also preparing to move to an exascale-class Dojo supercomputer capable of handling huge streams of visual data that is growing rapidly with its fleet of vehicles and teaching cars more efficiently. By the way, to date, the company’s cars with the autopilot on have driven over 3 billion miles on public roads. In addition, just the other day there were reports that Tesla plans to update the current radar, replacing it with an improved 4D model with a doubled detection range. Apparently, this is the very Tesla Autopilot 4D update that Musk talked about in July.

As we have repeatedly noted, the launch of the promised Tesla Network robotic taxi service with 1 million self-driving people will depend on the implementation of full Full Self-Driving (we are talking about the “feature complete” Full Self-Driving package, we talked about it in more detail in this article). cars, which the company originally planned to launch in 2020.

According to the initial idea, at the first stage, the Robotaxi service will work with insuring drivers. Based on the results of testing, Tesla plans to obtain the necessary data to develop a model of a fully autonomous service and contact the regulator to obtain an appropriate license. Musk has previously said cautiously that he expects to achieve some of the goals in 2021.

It remains to be hoped that in the end, Tesla will succeed, and incidents with the mistaken mistake of flags at a gas station for a traffic light, as well as fatal errors leading to a collision with a bump stop in problem areas (analysis below), will remain far in the past.


Popular on Gadget Tendency