Skip to main content

Dive into Embodied AIHome Introduction

Decision making

Reinforcement learning for decisions

Build intuition for sequential decisions, from MDPs to PPO and SAC.

Vision-language-action models (VLA)

RT, OpenVLA, ACT, Diffusion Policy, and the π family.

Explore how world models can be applied to embodied tasks.

Motion control

Reinforcement learning for control

Connect policy learning to continuous control and robotics tasks.

ControlPolicy training

A structured course from PID and LQR to MPC and impedance control.

Motion planning

From models and frame trees to closed-loop planning with MoveIt 2.

Motion PlanningMoveIt 2

Perception

Visual perception and VLMs

Transformers, ViT, vision encoders, and multimodal fusion.

Localization and tactile sensing

SLAM, foot contact, tactile sensing, and multisensor fusion.

Engineering foundations

Simulation tools

Get started with MuJoCo, Isaac Sim, Gymnasium, and PyBullet.

SimulationMuJoCo

Coordinate transforms, FK/IK, tf2, URDF, and MoveIt 2.

CAN and MCU communication

Low-level communication, actuator protocols, and host–controller links.

Mechanical design

Links, joints, motors, gearboxes, and end effectors.

HardwareRobot body

Data engineering and imitation learning

From teleoperation data to imitation learning and policy training.

Use the knowledge map to find gaps, then explore a topic.Explore the foundations

AMD projects

AUP Learning Cloud

Access Ryzen AI APUs in your browser with ROCm, JupyterHub, and Code Server.

AMDAUPRyzen AIROCmCloud compute

ACT bimanual training

ACT BF16 training on Radeon GPUs, closed-loop evaluation, and successful rollouts.

AMDROCmACTALOHALeRobot

Explore the Pupper quadruped

An AMD flagship project: reinforcement learning for locomotion and VLA experiments.

AMDPupperRL LocomotionVLA

Simulation projects

Build a quadruped from scratch

Follow eight chapters, from PD control and kinematics to policy training and perception.

CS123MuJoCo8 chapters

MicroDuck RL biped

Parallel PPO training on GPUs and stable gaits with mjlab and MuJoCo Warp.

MicroDuckmjlabPPO

MuJoCo robot arms and DDPG

From environment setup to continuous control with InvertedPendulum, Reacher, and Pusher.

MuJoCoDDPGRobot arms

ACT bimanual training

Train ACT on ALOHA simulation data and evaluate it over multiple episodes.

ACTALOHAImitation learning

Flamingo wheeled biped · Preview

Preview the course on Isaac Lab training and cross-simulator validation.

Isaac LabPreview

Real robot projects

LeRobot course notes in Chinese

Prepare for robot learning with datasets, toolchains, and classical robotics.

LeRobotChinese notes

SO-101 + LeRobot hardware tutorial

From hardware connectivity and safety tests to action playback on a real robot.

SO-101LeRobotReal robots

Validate a simulation policy before deploying it to a real robot.

Sim2RealDeploymentValidation

Choose projects by platform, simulation experiments, or real robot deployment.View projects

中文
English
Español
Italiano

双足 / 人形机器人

双足 / 人形机器人
task-planning-demo
任务规划 Demo

View as Markdown

Page content converted to Markdown. Use the original page link at the end to explore interactive graphics.

This page is currently available in Chinese. Explore the English introduction, or read the Chinese version.

任务规划 Demo

施工中。

Last updated on May 2, 2026 by Bo Kang

Explore

Introduction
Foundations
Projects

Community

GitHub Discussions
Datawhale

More

GitHub

Copyright © 2026 Datawhale. Built with Docusaurus.