Gary Zhou 
The release notes for Tesla’s Full Self-Driving Beta v10.11 hint at a number of critical improvements for the advanced driver-assist software. Tesla FSD Beta 10.11 is rolling out to Tesla employees for the time being. However, if the system performs well, external users should receive the update within the coming days.
There are several notable improvements outlined in FSD Beta v10.11’s release notes. Tesla stated that V10.11 utilizes more accurate predictions of where other vehicles are turning or merging, reducing unnecessary slowdowns. The company also stated that V10.11 should improve vehicles’ right-of-way understanding, which should be invaluable in scenarios when maps turn out to be inaccurate.
More importantly, FSD Beta V10.11 featured specific improvements for vulnerable road users (VRU). Tesla notes that the most recent version of FSD Beta should improve VRU detection by 44.9%, allowing the system to dramatically reduce “spurious false positive pedestrians and bicycles.” The company was able to accomplish these VRU improvements by increasing the size of its next-generation labelers.
Following are FSD Beta v10.11’s release notes.
Early Access Program | FSD Beta 10.11
– 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 auto-labeler, 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.
– 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%.
– 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%.
Tesla FSD Beta v10.11 will likely be released as software version number 2022.4.5.15, as per reports from the online electric vehicle community. Tests of v10.11’s performance in real-world roads are typically shared by members of the company’s FSD Beta program within hours of the system’s wide release.
