WHAT IT IS
Full Self-Driving is Tesla’s advanced driver-assistance system, pushed to some owners as a test version from October 2020. It uses cameras alone as its sensory input, with a neural network producing driving decisions directly from images. Versions around 2024 replaced much of the hand-written control logic with an end-to-end trained network, iterated on data from a large vehicle fleet.
Why it matters
By insisting on a camera-only approach and training an end-to-end network on large-scale fleet data, it represents the technical path in autonomous driving opposite to multi-sensor systems.
Key specs
- Form
- Vehicle-mounted driver-assistance system
- Sensors
- Cameras only (vision-only)
- Method
- An end-to-end trained neural network
Related concepts
Object Detection
From “what is in the image” to “what, where, and how many”
Semantic Segmentation
Colouring every pixel: not a box around the object, but a colouring book
Deep Reinforcement Learning
Let a neural network decide straight from pixels — then hold it steady with decades-old tricks
Safety, Alignment & Prompt Injection
A model optimises the proxy we wrote into the loss, never the thing we actually want — the gap between them is the whole alignment problem