RoboGenesis 101 | RoboGenesis 101: Practical Robot Learning with Genesis

Genesis foundations

Learn environment diagnostics, scenes, rigid-body physics, robot control, inverse kinematics, cameras, and parallel simulation.

From control to grasping

Build a grasping scene and understand scripted experts and demonstration acquisition.

From data to policies

Continue through data recording, imitation learning, domain randomization, policy training, and closed-loop evaluation.

Course status

L01–L07 are cpu-verified after their clean-kernel CPU verification; their additional AMD paths also passed their respective contracts on the reference R9700. L08 is gpu-verified after its bilingual CPU numerical paths, CPU+EGL path, and scripted-expert path on a reference R9700 all passed. L09 is also gpu-verified: its bilingual CPU diagnostics, CPU+EGL path, and dual-camera recording and readback on a reference R9700 all passed. The renumbered L12 training content is gpu-verified after ACT and SmolVLA GPU smoke, checkpoint audit, and reload checks. L10 is cpu-verified after its bilingual CPU-only data-readback and inspection paths passed. L11 is gpu-verified after its bilingual diagnostics, CPU+EGL path, and bounded-DR recording, readback, and provenance path on a reference R9700 passed. L13 remains planned.

Lesson	Topic	Planned duration	Hardware	Status
L01	Introduction, Runtime Platforms, and Environment Diagnostics	30 min	`cpu-ok`	`cpu-verified`
L02	Scenes, Entities, and the Simulation Lifecycle	90 min	`cpu-ok`	`cpu-verified`
L03	Rigid-Body Physics and Stable Simulation	90 min	`cpu-ok`	`cpu-verified`
L04	Robot Models, Degrees of Freedom, and Joint Control	90 min	`cpu-ok`	`cpu-verified`
L05	Inverse Kinematics, End-Effector Poses, and Cameras	120 min	`cpu-ok`	`cpu-verified`
L06	Parallel Simulation and Batched Franka Control	90 min	`cpu-ok`	`cpu-verified`
L07	Building a Grasping Task Scene	90 min	`cpu-ok`	`cpu-verified`
L08	Demonstration Acquisition and Scripted Experts	120 min	`gpu-recommended`	`gpu-verified`
L09	Synthetic Data Recording and Collection Throughput	120 min	`gpu-recommended`	`gpu-verified`
L10	Dataset Anatomy and Imitation Learning 101	90 min	`gpu-recommended`	`cpu-verified`
L11	Domain Randomization	90 min	`gpu-recommended`	`gpu-verified`
L12	Training ACT and SmolVLA Policies	150 min	`gpu-required`	`gpu-verified`
L13	Closed-Loop Evaluation and Capstone	120 min	`gpu-required`	`planned`

Lesson

Topic

Planned duration

Hardware

Status

L01

Introduction, Runtime Platforms, and Environment Diagnostics

30 min

cpu-ok

cpu-verified

L02

Scenes, Entities, and the Simulation Lifecycle