The GOAT Racer course

Teach a real car to drive itself.

Drive the car in your browser. Train a policy on a cloud GPU and watch it learn. Race it. Then put it on the real car.

Drive the car now Race the trained policy Create an account

Hello Racer, the race and one vision lab need no account. The other labs and a GPU run need one: ask your instructor for an invite code. Want your own car? Buy a GOAT Racer at cryptide.com.

The real car, on a track in our office.
The GOAT Racer: a Traxxas rally chassis carrying a Jetson Orin, a depth camera and a lidar
The car A 1/10 rally chassis with a Jetson Orin, a depth camera and a lidar. The simulator is its digital twin, so the policy you train here is the policy it drives. See the cars at cryptide.com →
Sim to real

The same policy, in the sim and on the car

The simulator is a digital twin of the car. A policy you train in the cloud is the file the real car loads, with no change.

In the sim A trained policy on the twin, with the readout you see in the labs.
On the car The real car on a track made of duct hose.

See the full loop on the NVIDIA livestream.

The course

Nine labs, from your first drive to the real car

Start here

Meet the car, then the twin it is trained on.

Train

Hand the wheel over: train a policy and watch it drive.

See

A waypoint from a tag, then the open path from depth; then your code pulls the wheel while you drive.

Real car

Make a track, train the car's own policy on tracks like it, and prove the car can drive it.

See every lab Read the docs

Plans

Start free. Train on a cloud GPU when you are ready.

Coming soon

No car? Send your policy to ours.

Train in the cloud, then run your policy on a real GOAT Racer in our lab. You get the video and the telemetry back. Tell me when it opens.

Questions

Before you start

Do I need a GPU?

No. Training runs on a cloud GPU. Your browser shows the live reward curve and the replay.

Do I need a car?

No. Every lab runs in your browser and on the cloud. The car is the last step, when you want your policy on real hardware.

What do I learn?

How a robot car senses and moves, its digital twin, reinforcement learning, how to read a training curve, vision from a depth camera, and how a policy goes from the simulator to the car.

Where are the docs?

At docs.goatracer.ai: getting started, the labs, and the car.

1. Drive

Hello Racer, in the browser. Steer and throttle: the two numbers every policy sends.

2. See

Depth, AprilTags, the open path: the camera labs run in your browser, on frames from the real car.

3. Train

Reinforcement learning on a cloud GPU. Live curve, replay, and the policy file.

4. Real car

Your policy goes to a GOAT Racer over the air, and its own software drives it. Get a car.