Supervised Learning — the robot imitates you¶
You label camera photos with the right steering; a CNN learns to do what you did.
Collect & label¶
- PhysiCar panel → Racing → pill Machine Learning → Train page → Learning Algorithm Supervised.
- The labeling UI is embedded right there: for each camera frame, press the button for the correct action (left / straight / right ...). Photos land in your labeling dataset.
- Label a few hundred frames — cover curves, straights, and both directions. Bad or lazy labels teach bad driving: the model can only be as good as your labels.
Train¶
- Pick the Deep Learning Base Model (or continue from an earlier model).
- Click Run — training progress, loss chart, and phase show in the monitoring modal. Every run adds a model to the store; nothing is overwritten.
Drive¶
- Switch to the Inference page, pick your model (default: newest), Run.
- Watch the live view. Wobbly on curves? Label more curve frames and train again — this loop (collect → train → drive → improve) is the daily life of every ML engineer.
You can even switch models while driving — pick another model mid-run and the car follows it between frames.
How a CNN "sees" → Learn: deep learning · Why labels matter → Learn: supervised learning
