Tesla Starts Rolling Out Full Self-Driving v15 With Reality View

Written by

in

Tesla has started rolling out Full Self-Driving version 15, its newest driver-assistance software, the company announced on 21 November. Tesla claims the update delivers “25x better handling of safety critical cases” and adds a new live driving display called Tesla Reality View.

Tesla says v15 upgrades every part of its end-to-end driving network. The release notes credit a new vision encoder, shared between Full Self-Driving, Robotaxi and the Optimus robot and trained on millions of internet videos, with improving the critical intervention rate by 25 times. That is Tesla’s own claim, tied to the critical intervention rate — the company has not shared the raw numbers behind it, so it should be read as a company claim until independent testing backs it up.

Reality View is the most visible change. It is a real-time generative visualisation that streams a view from a virtual camera above the vehicle, showing the driver what the system is planning to do. Tesla says it lets drivers see and control what Full Self-Driving is planning.

The update also improves detection of road signs, small animals, road debris, chains, potholes and deep standing water, and improves speed handling in school zones, at speed limits and in bad weather. It can now slow down for speed cameras, and a new Bike Rack Mode allows Full Self-Driving to be used with a bike rack or similar hitch accessory. Navigation is now learned from fleet route preferences, covering the start and end of trips for now.

What it means

Version 15 is Tesla’s biggest Full Self-Driving update yet, and its safety claim is the one to watch — a 25-times improvement would be a major step towards unsupervised driving, but only real-world use will prove it. Reality View could also make the system easier for drivers to understand and supervise, which matters while the driver is still legally responsible.

Sources: