Tesla Expands FSD Beta v12.3; Autopilot Director Confirms Ability to Reverse is Coming Soon

FSD Beta v12 is living up to hype

Not a Tesla App

Tesla's latest Full Self-Driving (FSD) Beta v12 is reaching users after an extended rollout over the weekend. This update is not just another iteration; it represents a leap in self-driving technology, primarily due to its integration of end-to-end neural networks for vehicle control. So far, it seems to be living up to the hype.

A Neural Network Driven Approach

At the core of FSD v12 is a shift from traditional programming to neural network-based decision-making. This allows the vehicle to process raw camera footage and vehicle kinematics directly into driving actions, mimicking human cognitive processes more closely than ever before. Ashok Elluswamy, Director of Autopilot Software at Tesla, highlighted the monumental effort to surpass the capabilities of the previous v11, setting a new standard for FSD's future.

Ability to Reverse Is Coming

The release has garnered widespread acclaim, with tech leaders like Michael Dell praising its capabilities and likening the car to human-like driving proficiency. Many in the Tesla community have been posting their weekend drivers, including Chuck Cook, who was amazed during one point of his drive, referring to a move as “Robo-taxi navigation.” Cook believed he was in too tight of a spot to pull a U-turn, but his Tesla pulled it off. Elluswamy commented: “Reverse coming soon when the Actually Smart Summon and the FSD models merge together over the next few releases.”

This clip blew my mind, next level for FSDBeta v12.3 in finding its way out of a bad situation. Nice work @Tesla_AI pic.twitter.com/9cDccpr2uO

— Chuck Cook (@chazman) March 16, 2024

There were numerous examples of v12.3 navigating complex driving scenarios easily, showcasing significant improvements over earlier versions.

“Reverse coming soon when the Actually Smart Summon and the FSD models merge together over the next few releases.” - Ashok Elluswamy

Next-Level Capabilities on the Horizon

Elon Musk has teased that v12.4 will introduce even more advanced features, emphasizing the continuous improvement in training compute constraints. In fact, he was so thrilled with the next update that he said it could be called version 13. He posted that “V12.4 is another big jump in capabilities. Our constraint in training compute is much improved.”

Tesla’s FSD trajectory seems to be hitting a new level, as it appears to be headed toward approval of use on roadways in Europe. While it is taking longer than initially believed, the dream of autonomous driving seems to be getting closer to reality.

Follow

Follow

Follow

Tesla Vehicles Spotted With LiDAR: What Do They Use It For?

Not a Tesla App

Tesla recently hit the news for purchasing approximately $2M in LiDAR sensors from Luminar, one of Tesla’s long-term suppliers. You’ve probably seen photos of Tesla’s Semi and various Tesla models, including the Model 3 and Model Y sporting LIDAR equipment on the roof. These cars drive around with manufacturer plates scanning streets and highways.

However, many people confuse Tesla’s purpose in purchasing LiDAR equipment with using it for FSD versus testing. So, let’s look at what LiDAR is, and why Tesla uses it on its Fleet Validation Vehicles.

What is LiDAR?

LiDAR stands for Light Detecting and Ranging – essentially using lasers to measure distances. A laser pulse is sent out, and the time it takes to return is measured – providing extremely accurate distance measurements.

Some companies working on self-driving vehicles, including Waymo and BYD, use LiDAR as part of their self-driving suites, but Tesla is one of the few stand-outs that does not. Even Rimac’s “Verne” Robotaxi – which uses self-driving technology from Mobileye, also uses LiDAR.

While LiDAR can produce extremely accurate and high-quality 3D environments, it comes with its downsides as well. Not only is LiDAR costly and requires large gear strapped to a vehicle, but it also can not be used in bad weather and can have interference issues if there are other strong light sources present.

Why Does Tesla Use LiDAR?

A LiDAR rig mounted on a Tesla Semi for testing FSD.

Not a Tesla App

At Autonomy Day in 2019, Elon Musk mentioned that LiDAR isn’t the solution for self-driving cars – it's just a crutch. Thus, Tesla hasn’t used LiDAR for any production self-driving software.

Instead, Tesla uses it exactly how it's described – they use it to gather ground-truth data. This data is then used to feed Tesla’s Full Self Driving system – which helps validate its vision-only system's accuracy. LiDAR provides very accurate measurements to help ensure that FSD’s perception of space is accurate – and is only used by Tesla to ensure that its AI technology which is the brains of FSD is capable of accurately interpreting depth from just visual data.

Tesla’s vision-only system has been seen to be extremely accurate, with Vision-only Autopark being able to park in even narrower and tighter spaces faster than the previous version that relied on ultrasonic sensors.

We’ll likely continue to see Tesla purchase LiDAR systems, as well as use them for validation well into the future.

Follow

Follow

Follow

Tesla's Upcoming Robotaxi Event in August Delayed, According to Bloomberg

Sugar Design

In a report from Bloomberg, it is claimed that Tesla will be delaying its much-anticipated 8/8 Robotaxi event by two months to October 2024.

While sources other than Bloomberg haven't confirmed this report, Bloomberg has a positive track record of reporting on financial decisions. We’ll be sure to update the article if there is confirmation on X from Elon Musk or another Tesla senior official.

Tesla’s stock has dropped nearly 8.5% over the day, ending back-to-back gains over the last two weeks. It closed yesterday at $ 241 after hitting a peak of $270 earlier in the day before the news broke.

Why the Delay?

The delay – of approximately two months – has been communicated internally, but not publicly announced just yet. Bloomberg goes on to mention that the design team was told to rework certain elements of the Cybercab, necessitating the delay.

If Bloomberg’s report is correct, it sounds like Tesla’s unveil event will be largely focused on showing off the vehicle, instead of demoing how it will work. Of course, it could still be both, but given past events, Tesla has always shown off the vehicle years before it hits production.

Rimac recently showed off their version of robotaxi vehicle named Verne, and surprisingly, it could almost pass for Tesla’s own robotaxi. A lot of design cues in Rimac’s version are elements we have already seen or expect to see in Tesla’s autonomous taxi.

A recent Tesla patent revealed that Tesla is incorporating a sanitation system into their robotaxi that will be responsible for analyzing and cleaning the vehicle’s interior, although the delay itself is likely tied more to a physical feature rather than software.

Another element we know almost nothing about is how Tesla plans to charge these robotic taxis. Will they rely on the existing charge port and adapt a solution like the robotic charging arm (video below) we saw almost eight years ago, or will wireless charging or a dock finally become realized?

While the delay for Tesla’s event appears to be related to the vehicle’s design itself and not further development of FSD, Tesla is wasting no time in getting FSD working for the upcoming vehicle. Model 3 vehicles have already been spotted with camera locations that resemble a robotaxi.

Is the Delay Accurate?

We expect that this delay might actually be true – Elon Musk usually takes to X within hours of such news breaking if it's false to refute it and hasn’t done so yet.

Tesla has delayed several of their events in the past, and a delay of a couple of months seems plausible. We should hear from Musk himself soon on whether this report is accurate.

Follow

Follow

Follow