Tesla starts releasing a big, full self-driving beta update

Tesla has started releasing a big new Full Self-Driving (FSD) beta software update with lots of improvements.

The automaker could expand access to the beta if this update “works well”, according to CEO Elon Musk.

Fully Autonomous Tesla in Beta

Since October 2020, Tesla has been slowly rolling out what it calls the “Full Self-Driving Beta” (FSD Beta), which is an early version of its self-driving software currently being tested by a fleet of company-selected Tesla owners. and through its “security test score”.

The software allows the vehicle to navigate autonomously to a destination entered into the car’s navigation system, but the driver must remain alert and ready to take control at all times.

Since the responsibility lies with the driver and not with Tesla’s system, it is still considered a level two driver assistance system despite its name. It was something of a “two steps forward, one step back” program, as some updates saw regressions in terms of driving capabilities.

Tesla frequently released new software updates for the FSD Beta program and added more owners to it.

The last major update was FSD Beta 10.10 in early February.

In the fourth quarter of 2021, the automaker said it had nearly 60,000 owners in the FSD Beta program.

Tesla FSD beta 10.11

Now, Tesla has started rolling out a new FSD Beta 10.11 update to its early access program, and it’s a big update based on the release notes.

Here are the release notes:

  • Improved modeling of lane geometry from dense (“bag of points”) rasters to an autoregressive decoder that directly predicts and connects “vector space” lanes point by point using a transformative neural network . This allows us to predict crossover pathways, enables less expensive and less error-prone post-processing, and paves the way for the prediction of many other signals and their relationships jointly and end-to-end. Use more accurate predictions of where vehicles turn or merge to reduce unnecessary slowdowns for vehicles that won’t cross our path.
  • Better understanding of right-of-way if map is inaccurate or car cannot follow navigation. In particular, intersection extent modeling is now entirely based on network predictions and no longer uses map-based heuristics.
  • Improved accuracy of VRU detections by 44.9%, significantly reducing false positives for pedestrians and bicycles (especially around tar seams, skid marks and raindrops). This was accomplished by increasing the data size of the next-generation automatic labeler, training network parameters that were previously frozen, and modifying network loss functions. We find that this decreases the incidence of VRU-related false slowdowns.
  • 63.6% reduction in predicted speed error of motorcycles, scooters, wheelchairs and very close pedestrians. To do so, we introduced a new dataset of simulated adversarial high-speed VRU interactions. This update improves autopilot control around fast moving and activating VRUs.
  • Improved crawling profile with higher jerk at the start and end of the crawling.
  • Improved close obstacle control by predicting the continuous distance to static geometry with the general static obstacle network.
  • Reduced “parked” vehicle attribute error rate by 17%, achieved by increasing dataset size by 14%. Also improved the accuracy of the brake lights.
  • Improved out-of-the-box scenario speed error by 5% and highway scenario speed error by 10%, achieved by adjusting the loss function to improve performance in difficult scenarios.
  • Improved detection and control for open car doors.
  • Improved smoothness in corners by using an optimization-based approach to decide which road lines are irrelevant for control given lateral and longitudinal acceleration and jerk limits as well as vehicle kinematics.
  • Improved the stability of FSD Ul visualizations by optimizing the Ethernet data transfer pipeline by 15%.

CEO Elon Musk says if this update “works well”, Tesla will “likely” reduce access to drivers who scored ’95’ on the Driver Safety Score. This should allow many more FSD buyers to access the beta.

As for Canada, Tesla owners who purchased FSD have had access to the Driver Safety Score for over a week now. The automaker should soon start rolling out the beta to those with the highest scores.

FTC: We use revenue-generating automatic affiliate links. Continued.

Subscribe to Electrek on YouTube for exclusive videos and subscribe to the podcast.