Tesla Rolls Out Full Self-Driving Beta v11 Update to More Vehicles
Drivers who have experienced the v11 Beta have reported several improvements, including better highway driving and more accurate inner-city street navigation. This update marks a step in the right direction for Tesla and its FSD Beta program.
 |
| Tesla Full Self Driven Car |
Drivers Report Improvements in Highway Driving and Inner-City Navigation
Enabled FSD Beta on highways: This feature unifies the vision and planning stack on and off highways and replaces the over four-year-old legacy highway stack that relies on several single-camera and single-frame networks. The FSD Beta's multi-camera video networks and next-gen planner allow for more intelligent behaviors, smoother control, and better decision-making.
FSD Beta Update Includes Improved Recall and Smoother Maneuvers
Improved recall for close-by cut-in cases: This improvement was achieved through an additional 30k auto-labeled clips mined from the fleet. This applies particularly to large trucks and high-yaw rate scenarios. Additionally, dedicated speed control for cut-in objects has been expanded and tuned.
Tesla's FSD Beta Program Takes a Step Forward with v11 Update
Improved the position of ego in wide lanes: This improvement biases the ego in the direction of the upcoming turn to enable other cars to maneuver around it.
Improved handling during scenarios with high curvature or large trucks: This improvement offsets the ego in the lane to maintain safe distances to other vehicles on the road and increase comfort.
Improved behavior for path blockage lane changes in dense traffic: Ego will now maintain more headway in blocked lanes to hedge for possible gaps in dense traffic.
Improved lane changes in dense traffic scenarios: This improvement allows for higher acceleration during the alignment phase, resulting in more natural gap selection to overtake adjacent lane vehicles close to ego.
Made turns smoother: This improvement was accomplished by integrating the latest version of the lane-guidance module into the road edge and lines network.
Enhanced Features of the FSD Beta v11 Update
Improved accuracy for detecting other vehicles' moving semantics: Precision was improved by 23% for cases where other vehicles transition to driving, and error was reduced by 12% for cases where Autopilot incorrectly detects its lead vehicle as parked. These improvements were achieved by increasing video context in the network, adding more data of these scenarios, and increasing the loss penalty for control-relevant vehicles.
Extended maximum trajectory optimization horizon: This improvement results in smoother control for high-curvature roads and faraway vehicles when driving at highway speeds.
Improved driving behavior next to a row of parked cars in narrow lanes: This improvement prefers to offset and stay within the lane instead of unnecessarily lane-changing away or slowing down.
Improved back-to-back lane change maneuvers: This improvement was achieved through better fusion between vision-based localization and coarse map lane counts.
Added text blurbs in the user interface: The text blurbs communicate upcoming maneuvers that the FSD Beta plans to make. The visualization of upcoming slowdowns along the vehicle's path has also been improved. Chevrons render at varying opacity and speed to indicate the slowdown intensity, and a solid line appears at locations where the car will come to a stop.
Improved the recall and precision of object detection: This improvement notably reduces the position error of semi-trucks by 10%, increases the recall and precision of crossing vehicles over 100m away by 3% and 7%, respectively, and increases the recall of motorbikes by 5%.
Keywords: Tesla, Full Self-Driving Beta, v11 update, Tesla Software Update 2022.45.11, highway driving, inner-city navigation, autonomous driving, Elon Musk, update improvements.
Tags: Tesla, autonomous driving, Full Self-Driving Beta, v11 update, software update, Elon Musk, self-driving cars.
No comments:
Post a Comment