Tesla FSD Beta V10.11 with Critical Improvements May Lower Safety Score Requirement to 95

Photo: Kim Paquette/YouTube

Tesla has begun distributing the Full Self-Driving (FSD) Beta V10.11 update, and the release notes show several critical improvements. If the testing is successful, the manufacturer will consider lowering the minimum safety score of testers to 95, Elon Musk said.

Tesla has started rolling out the FSD Beta V10.11 update to users, and the release notes include several important improvements. Among them are more accurate predictions of where vehicles are turning or merging, which should reduce unnecessary slowdowns, and an improved right-of-way understanding if the map is inaccurate or the car cannot follow the navigation, which is critical.

In addition, the precision of vulnerable road users (VRU) detections was improved by 44.9%, dramatically reducing spurious false positive pedestrians and bicycles (especially around tar seams, skid marks, and rain drops), this was achieved thanks to increasing the data size of the next-gen autolabeler, training network parameters that were previously frozen, and modifying the network loss functions.

In addition, in V10.11 there was improved control for nearby obstacles by predicting continuous distance to static geometry with the general static obstacle network and reduced vehicle “parked” attribute error rate by 17%, achieved by increasing the dataset size by 14%.

Such significant improvements make driving even smoother and more accurate. Some of the testing participants, numbering over 60,000 as of Q4 2021, share their positive driving experience, claiming that the car has driven from one place to another with zero intervention. However, do not forget that the driver must never lose vigilance and must be ready to intervene at any moment.

Tesla CEO Elon Musk said that if the test of FSD Beta V10.11 goes well, the company will likely lower the minimum safety score of testers to 95. This will mean that a wider range of owners will have access to FSD Beta. However, the most important thing to note here is that lowering the requirements for testers could mean that the safety of the feature has increased significantly.

If this version performs well, we can probably lower min safety score to 95

— Elon Musk (@elonmusk) March 14, 2022

The release notes:

• Upgraded modeling of lane geometry from dense rasters (“bag of points”) to an autoregressive decoder that directly predicts and connects “vector space” lanes point by point using a transformer neural network. This enables us to predict crossing lanes, allows computationally cheaper and less error prone post-processing, and paves the way for predicting many other signals and their relationships jointly and end-to-end. Use more accurate predictions of where vehicles are turning or merging to reduce unnecessary slowdowns for vehicles that will not cross our path.
• Improved right-of-way understanding if the map is inaccurate or the car cannot follow the navigation. In particular, modeling intersection extents is now entirely based on network predictions and no longer uses map-based heuristics.
• Improved the precision of VRU detections by 44.9%, dramatically reducing spurious false positive pedestrians and bicycles (especially around tar seams, skid marks, and rain drops). This was accomplished by increasing the data size of the next-gen autolabeler, training network parameters that were previously frozen, and modifying the network loss functions. We find that this decreases the incidence of VRU-related false slowdowns.
• Reduced the predicted velocity error of very close-by motorcycles, scooters, wheelchairs, and pedestrians by 63.6%. To do this, we introduced a new dataset of simulated adversarial high speed VRU interactions. This update improves autopilot control around fast-moving and cutting-in VRUs.
• Improved creeping profile with higher jerk when creeping starts and ends.
• Improved control for nearby obstacles by predicting continuous distance to static geometry with the general static obstacle network.
• Reduced vehicle “parked” attribute error rate by 17%, achieved by increasing the dataset size by 14%. Also improved brake light accuracy.
• Improved clear-to-go scenario velocity error by 5% and highway scenario velocity error by 10%, achieved by tuning loss function targeted at improving performance in difficult scenarios.
• Improved detection and control for open car doors.
• Improved smoothness through turns 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 stability of the FSD Ul visualizations by optimizing the ethernet data transfer pipeline by 15%.

© 2022, Eva Fox | Tesmanian. All rights reserved.

_____________________________

We appreciate your readership! Please share your thoughts in the comment section below. 

Article edited by @SmokeyShorts, you can follow him on Twitter

Follow @EvaFoxU