Tesla has started deploying the next version of its Full Self-Driving (FSD) Beta software to testers in the United States.
The first downloads of 2022.4.5.15 started appearing at around 10:30am this morning.
— Zack (@BLKMDL3) March 14, 2022
The release notes of this version first leaked yesterday, showing it will contain a number of substantial improvements.
But most importantly, this is the version of FSD Beta that is supposed to be released in Canada for the first time. As of Elon Musk’s most recent update, Canada was supposed to get this version 3 days after our friends in the US, however since it was delayed we don’t know if that timeline is still true.
There have been no reported downloads of 2022.4.5.15 in Canada at the time of publication, but we will update you as soon as one appears.
2022.4.5.15 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 VI interactions. This update improves autopilot control around fast-moving and cutting-in VRUs.
Improved creeping profile with higher jerk when creeping stand 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%, achieve 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 functions 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 c ? given lateral and longitudinal acceleration and jerk limits as w? vehicle kinematics.
Improved stability of the FSD Ul visualizations by optimizing ethernet data transfer pipeline by 15%.
Improved recall for vehicles directly behind ego, and improved precision for vehicle detection network.