Connect with us
Tesla-fsd-10.3-release Tesla-fsd-10.3-release

News

Tesla FSD Beta 10.69 release notes highlight better left turns, smoother driving

(Credit: Tesla)

Published

on

Tesla released FSD Beta 10.69 to the first round of testers over the weekend. Read v.10.69’s release notes below to check out the latest improvements. 

Stay in your Lanes

  • Added a new “deep lane guidance” module to the Vector Lanes neural network which fuses features extracted from the video streams with coarse map data, i.e. lane counts and lane connectivites. This architecture achieves a 44% lower error rate on lane topology compared to the previous model, enabling smoother control before lanes and their connectivities becomes visually apparent. This provides a way to make every Autopilot drive as good as someone driving their own commute, yet in a sufficiently general way that adapts for road changes.

 Nothing Like Smooth Driving

  • Improved overall driving smoothness, without sacrificing latency, through better modeling of system and actuation latency in trajectory planning. Trajectory planner now independently accounts for latency from steering commands to actual steering actuation, as well as acceleration and brake commands to actuation. This results in a trajectory that is a more accurate model of how the vehicle would drive. This allows better downstream controller tracking and smoothness while also allowing a more accurate response during harsh manevuers.
  • Increased smoothness for protected right turns by improving the association of traffic lights with slip lanes vs yield signs with slip lanes. This reduces false slowdowns when there are no relevant objects present and also improves yielding position when they are present.
  • Reduced false slowdowns near crosswalks. This was done with improved understanding of pedestrian and bicyclist intent based on their motion.
  • Enabled creeping for visibility at any intersection where objects might cross ego’s path, regardless of presence of traffic controls.
  • Improved accuracy of stopping position in critical scenarios with crossing objects, by allowing dynamic resolution in trajectory optimization to focus more on areas where finer control is essential.
  • Reduced latency when starting from a stop by accounting for lead vehicle jerk.

Chuck’s Left Turn

  • Improved unprotected left turns with more appropriate speed profile when approaching and exiting median crossover regions, in the presence of high speed cross traffic (“Chuck Cook style” unprotected left turns). This was done by allowing optimizable initial jerk, to mimic the harsh pedal press by a human, when required to go in front of high speed objects. Also improved lateral profile approaching such safety regions to allow for better pose that aligns well for exiting the region. Finally, improved interaction with objects that are entering or waiting inside the median crossover region with better modeling of their future intent.

Safety is Number 1

  • Added control for arbitrary low-speed moving volumes from Occupancy Network. This also enables finer control for more precise object shapes that cannot be easily represented by a cuboid primitive. This required predicting velocity at every 3D voxel. We may now control for slow-moving UFOs.
  • Made speed profile more comfortable when creeping for visibility, to allow for smoother stops when protecting for potentially occluded objects.
  • Improved speed when entering highway by better handling of upcoming map speed changes, which increases the confidence of merging onto the highway.
  • Enabled faster identification of red light runners by evaluating their current kinematic state against their expected braking profile.

Tesla FSD “Brain” Improvements

  • Upgraded Occupancy Network to use video instead of images from single time step. This temporal context allows the network to be robust to temporary occlusions and enables prediction of occupancy flow. Also, improved ground truth with semantics-driven outlier rejection, hard example mining, and increasing the dataset size by 2.4x.
  • Upgraded to a new two-stage architecture to produce object kinematics (e.g. velocity, acceleration, yaw rate) where network compute is allocated O(objects) instead of O(space). This improved velocity estimates for far away crossing vehicles by 20%, while using one tenth of the compute.
  • Improved geometry error of ego-relevant lanes by 34% and crossing lanes by 21% with a full Vector Lanes neural network update. Information bottlenecks in the network architecture were eliminated by increasing the size of the per-camera feature extractors, video modules, internals of the autoregressive decoder, and by adding a hard attention mechanism which greatly improved the fine position of lanes.
  • Improved recall of animals by 34% by doubling the size of the auto-labeled training set.
  • Increased recall of forking lanes by 36% by having topological tokens participate in the attention operations of the autoregressive decoder and by increasing the loss applied to fork tokens during training.
  • Improved velocity error for pedestrians and bicyclists by 17%, especially when ego is making a turn, by improving the onboard trajectory estimation used as input to the neural network.
  • Improved recall of object detection, eliminating 26% of missing detections for far away crossing vehicles by tuning the loss function used during training and improving label quality.
  • Improved object future path prediction in scenarios with high yaw rate by incorporating yaw rate and lateral motion into the likelihood estimation. This helps with objects turning into or away from ego’s lane, especially in intersections or cut-in scenarios.

Tesla is rolling out FSD Beta v.10.69 in phases, starting with ~1,000 testers over the weekend. Once the update is rolled out for wide release, the price of FSD Beta will increase.

The Teslarati team would appreciate hearing from you. If you have any tips, contact me at maria@teslarati.com or via Twitter @Writer_01001101.

Maria--aka "M"-- is an experienced writer and book editor. She's written about several topics including health, tech, and politics. As a book editor, she's worked with authors who write Sci-Fi, Romance, and Dark Fantasy. M loves hearing from TESLARATI readers. If you have any tips or article ideas, contact her at maria@teslarati.com or via X, @Writer_01001101.

Advertisement
Comments

Elon Musk

Tesla says it denied Musk CEO replacement report before it was published

Tesla says it responded to the WSJ’s request for comment, denying that it was in search of a new CEO to replace Elon Musk.

Published

on

tesla
(Credit: Tesla)

Tesla said that it denied seeking a replacement for CEO Elon Musk before a report was published claiming the company was considering a new frontman.

Last night, The Wall Street Journal reported that Tesla’s Board of Directors was looking for Musk’s replacement after he had devoted too much time to his role within the government. The publication revised its headline to the report no fewer than five times, initially stating the company was still seeking a replacement.

By the time the headline revisions were complete, it had outlined that Tesla had looked for a replacement a month ago, but had stopped its search following Musk’s commitment to Tesla during the company’s earnings call last month.

Shortly after the report surfaced, Board of Directors chairwoman Robyn Denholm officially issued a statement on behalf of Tesla:

“Earlier today, there was a media report erroneously claiming that the Tesla Board had contacted recruitment firms to initiate a CEO search at the company. This is absolutely false (and this was communicated to the media before the report was published). The CEO of Tesla is Elon Musk and the Board is highly confident in his ability to continue executing on the exciting growth plan ahead.  – Robyn Denholm.”

Tesla Board Chair slams Wall Street Journal over alleged CEO search report

Interestingly, Denholm’s statement indicates it had responded to a request for comment from the Wall Street Journal before the report was published. This is especially interesting because Tesla does not typically respond to media outreach, as it dissolved its media department several years ago.

Tesla typically makes its statements publicly on X.

Musk also responded to the report, indicating that the WSJ had committed an “extremely bad breach of ethics” by publishing a “deliberately false article” that did not include Tesla’s “unequivocal denial beforehand.”

Continue Reading

News

Robotaxis are already making roads safer, Waymo report reveals

Waymo Driver is already reducing severe crashes and enhancing the safety of vulnerable road users.

Published

on

Waymo hires former Tesla Executive 
(Credit: Waymo)

Industry leaders such as Elon Musk have always maintained that autonomous robotaxis will make roads safer. A recent blog post from Waymo about the safety of its self-driving cars suggests that Musk’s sentiments are on point.

Way More Safety

Waymo Driver is already reducing severe crashes and enhancing the safety of vulnerable road users. As per a new research paper set for publication in the Traffic Injury Prevention Journal, Waymo Driver had outperformed human drivers in safety, particularly for vulnerable road users (VRUs).

Over 56.7 million miles, compared to human drivers, Waymo Driver achieved a 92% reduction in pedestrian injury crashes. It also saw 82% fewer crashes with injuries with cyclists and 82% fewer crashes with injuries with motorcyclists. Waymo Driver also slashed injury-involving intersection crashes by 96%, which are a leading cause of severe road harm for human drivers. Waymo Driver saw 85% fewer crashes with suspected serious or worse injuries as well.

What They Are Saying

Mauricio Peña, Waymo’s Chief Safety Officer, was optimistic about Waymo Driver’s results so far. “It’s exciting to see the real positive impact that Waymo is making on the streets of America as we continue to expand. This research reinforces the growing evidence that the Waymo Driver is playing a crucial role in reducing serious crashes and protecting all road users,” the Chief Safety Officer noted.

Jonathan Adkins, Chief Executive Officer at Governors Highway Safety Association, also noted that Waymo’s results are very encouraging. “It’s encouraging to see real-world data showing Waymo outperforming human drivers when it comes to safety. Fewer crashes and fewer injuries — especially for people walking and biking — is exactly the kind of progress we want to see from autonomous vehicles,” Adkins stated.

Advertisement
Continue Reading

Elon Musk

Tesla hints at June 1 launch of Robotaxi platform in Austin

Tesla has hinted at a potential launch date for the Robotaxi service in Austin, Texas.

Published

on

tesla robotaxi app on phone
Credit: Tesla

Tesla just dropped its biggest hint yet about the potential launch date of its Robotaxi ride-hailing platform in Austin, Texas, shedding more light on when to expect it to take off.

In preparation for the ride-hailing service to launch, Tesla has been in talks with the City of Austin for months. It has also spent recent months bolstering its Full Self-Driving suite, aiming for it to handle initially supervised rides with the use of teleoperators to keep things safe and dependable, at least early on.

The company has also said that it expects the Robotaxi service, which will drive passengers in Tesla Model Y vehicles to start, to launch in Austin in June. However, Tesla has not given an exact date.

Now, Tesla is hinting that Robotaxi could launch on June 1, based on a very vague X post it published on May 1:

Of course, this is extremely speculative. However, it’s the first time Tesla has made any suggestions about a potential launch date, so it’s worth taking it seriously.

While the automaker has often missed timelines in the past, most notably the launch of a “feature-complete” Full Self-Driving platform, this is the first time we’ve seen Tesla be so adamant and truly reiterate a target date.

Tesla has not shied away from this June date for the Robotaxi launch yet, something that is worth noting as we move closer to June. All signs point toward Tesla being able to come through on this timeline, and it could be one of its biggest accomplishments yet on the grand scheme of things. The Robotaxi rollout will be controlled and small to start, the company noted on its most recent Earnings Call.

CEO Elon Musk said:

“The team and I are laser-focused on bringing robotaxi to Austin in June. Unsupervised autonomy will first be solved for the Model Y in Austin.”

At first, it also seems as if the first Robotaxi rides will be available to a select group, as Musk said the ability to order one will not be available to the general public until later in the month. He also said the initial fleet will be between 10 and 20 vehicles:

“Yeah. We’re still debating the exact number to start off on day one, but it’s, like, I don’t know, maybe 10 or 20 vehicles on day one. And watch it carefully. They scale it up rapidly after that. So, we want to make sure that you’re paying very close attention the first time this happens. But, yeah, you will be able to — end of end of June or July, just go to Austin and order a Tesla for autonomous drive.”

While the June 1st date of the Robotaxi launch is extremely speculative, Tesla seems convinced that its vehicles could already handle this task. It would be something to see them come through on this date, especially on the first day of the month.

Continue Reading

Trending