Robotics & Physical AI Learning Ecosystem

Official USB-to-UART bridge driver executable for Windows 10/11

Signed driver package for Apple Silicon (M1/M2/M3/M4) & Intel Macs

Official IDE with language server auto-complete and integrated serial plotter

package_esp32_index.json index URL for Arduino IDE preferences

Official hardware user guide, pinout diagrams, and power rail specifications

Complete electrical characteristics, strapping pin definitions, and RF ratings

Comprehensive API reference for WiFi, BLE, GPIO, and peripherals on ESP32

Official Espressif Arduino core source code, libraries, and hardware definitions

700+ pages of tutorials, C/C++ sketches, and Python scripts for S3 kit

Standard library for controlling WS2812 / WS2812B addressable RGB LED on GPIO 48

Official MicroPython firmware source and build toolchains

Official precompiled MicroPython firmware builds with SPIRAM support

Step-by-step pinout reference, deep sleep examples, and board setup guide

Virtual COM Port drivers for boards using CP2102/CP2104 bridge chips

Path A recommended complete kit with board, camera, display, sensors, breadboard

Direct manufacturer product page for ESP32-S3 Ultimate Starter Kit

Official Espressif dual-USB development board (Product ID 5364)

Budget sourcing for dual-USB ESP32-S3 N8R8 / N16R8 boards ($4 to $8 each)

Standard 830 tie-point prototyping breadboards

Assorted breadboard jumper wires (Male-Male, Male-Female, Female-Female)

Complete resistor kit for current-limiting LEDs and voltage dividers

Community server for DIY robotics, physical AI, and hardware troubleshooting

Open-source robotics codebases, CAD models, and course materials

Prof. Stephen Boyd — 18.5h on vectors, matrices, least squares, and QR factorization (M1)

Prof. Oussama Khatib — 19h on spatial descriptions, forward/inverse kinematics, and Jacobians (M2)

Prof. Fei-Fei Li & Justin Johnson — CNNs, visual recognition, and spatial transformers (M3)

Prof. Mac Schwager — Cameras, IMUs, lidar, Kalman filters, and robot state estimation (M3)

Prof. Marco Pavone — Trajectory optimization, MPC, dynamic programming, and LQR (M4)

Prof. Chelsea Finn — Policy gradients, actor-critic, offline RL, and robot learning (M5)

Prof. Emma Brunskill — Formal theoretical foundation of MDPs, value iteration, and RL theory (M5)

Building backprop, micrograd, makemore, and GPT architectures from scratch (M3)

Prof. Sergey Levine — Advanced policy optimization and robotics applications (M5)

Prof. Russ Tedrake — Manipulation, geometric perception, grasping, and contact mechanics (M6)

Prof. Russ Tedrake — Walking, running, swimming, and passive non-linear dynamics (M4)

Curated research talks and seminars from humanoid and foundation model roboticists (M6)

Lecture 17: Interactive robot perception, multisensory learning, and physical embodiment

Lecture 17: Generalization, cross-embodiment data sharing, and foundational imitation

Lecture 18: Open challenges in physical AI, scaling laws, and tactile reasoning

Action Chunking with Transformers (ACT) and Moravec's paradox in manipulation

Ψ0 architecture for whole-body loco-manipulation on real humanoid hardware

Structuring simulation, teleoperation, and passive human video data for policy training

GEAR Lab: Scaling laws for physical agents and universal humanoid controllers

Bridging passive human video observation to active robot policy execution

Ocean OneK and the evolution from classical mechanics to real-time physical AI

Official documentation tutorial covering language syntax, data structures, and modules (M0)

Complete downloadable handouts, problem sets, solutions, and exam papers for CS223A (M2)

Spring 2026 course schedule, lecture slide decks, and homework code for AA203 (M4)

Course syllabus, reading list, and project guidelines for Deep RL in Robotics (M5)

Stephen Boyd & Lieven Vandenberghe — Free complete textbook PDF, Julia, and Python code (M1)

Marco Pavone & Mac Schwager — Free open-access textbook on autonomous robotic systems (M4)

Russ Tedrake — Modern manipulation algorithms with runnable interactive Drake Python notebooks (M6)

Russ Tedrake — Nonlinear control, stabilization, and dynamic locomotion algorithms (M4)

High-performance multi-joint rigid-body contact dynamics engine (M2)

Curated, high-fidelity validated models of modern robots, quadrupeds, and humanoids (M2)

Official interactive Colab notebook walking through MuJoCo's Python bindings (M0)

Validated MJCF simulation model of the 3D-printable SO-ARM100 arm (M2)

End-to-end PyTorch library for real-world robotics AI and simulation (M6)

Setup guides for teleoperation, imitation learning pipelines, and dataset recording (M6)

GPU-accelerated modular framework for robot learning built on Omniverse and Isaac Sim

Model-based design, trajectory optimization, and simulation suite for manipulation

Standard RL benchmark tasks (Ant, HalfCheetah, Hopper, Humanoid, Walker2d) (M4)

Production-grade PyTorch implementations of PPO, SAC, TD3, and A2C algorithms (M5)

Open-source robotics foundation models and control policies (M6)

Python package to load URDF and MJCF descriptions of 70+ robots in one line (M2)

High-performance multimodal visualization engine for robotics and sensor streams

World's largest open robot dataset: 1M+ real demonstrations across 22 robot types (M6)

Action diffusion for visuomotor robot policy learning (Chi, Song et al.) (M6)

$122 low-cost 3D-printable 6-DOF robotic manipulator arm (M6)

$399 open-source desktop interactive robot arm

Consumer quadruped robot platform with 4D lidar, camera, and obstacle avoidance ($1,600)

Open-source 3D-printed bipedal walking humanoid robot legs (~$3,000)

Commercial research humanoid robot with 23–43 degrees of freedom ($13,500)

Associate Professor at Stanford; CS224R instructor and co-founder of Physical Intelligence

Associate Professor at UC Berkeley; pioneer in deep reinforcement learning and robotics

Senior Research Scientist & Head of GEAR Lab at NVIDIA, leading embodied AI

Assistant Professor at Columbia; research in physical dynamics and loco-manipulation

Professor at UC Berkeley and pioneer in robotic deep reinforcement learning

Creator of FreeRobotics. Built the curated 6-month Stanford robotics curriculum

Official 3D CAD STL models and assembly guide for the Robonine 76mm parallel gripper designed for SO-ARM100 and SO-ARM101 robot arms. GPL-3.0 licensed.

Browser-based text-to-CAD prompt pattern generator with replicad and manifold kernels. MIT licensed.

Official documentation, complete schematics, pinout diagrams, and C++/Python libraries for the Freenove ESP32-S3 starter kit used in RobotForge courses.

Complete URDF model of the 6-DOF SO-ARM100/101 manipulator with Robonine 76mm parallel gripper prismatic fingers, joint limits, visual/collision geometry, and inertias.

Complete 8-DOF quadruped robot description with trunk chassis, 4 hip revolute joints, 4 knee revolute joints, and foot contacts.

Differential drive rover model with chassis body, two continuous drive wheels, front caster sphere, and 2D LiDAR cylinder sensor.

Minimal 2-DOF planar manipulator URDF for learning forward kinematics, DH parameters, Jacobian matrices, and inverse kinematics.

Dual-core ESP32 task creation, core pinning (xTaskCreatePinnedToCore), task handles, priorities, and stack allocation in Arduino IDE.

Binary semaphores, counting semaphores, and interrupt synchronization with xSemaphoreGiveFromISR.

Thread-safe FIFO queues for inter-task communication, telemetry passing, and struct message serialization.

Mutual exclusion locking for shared serial ports, I2C buses, and global state to prevent race conditions and priority inversion.

Architectural breakdown of FreeRTOS task lifecycle states (Ready, Running, Blocked, Suspended) on ESP32.

Real-time operating system concepts: preemptive round-robin scheduling, cooperative scheduling, and heap memory management.

26 progressive robotics projects categorized from beginner Bristlebots and ArtBots to BlueBot line followers and underwater ROVs.

Curricular pathways: Student Kit, CTC GO! modular physical computing, Explore IoT Cloud, and Engineering Kit Rev 2 with MATLAB & Simulink.

FIRST Tech Challenge (FTC) & FRC curriculum covering drivetrains (Tank, Mecanum), UltraPlanetary gearboxes, Control Hub, and Java autonomous code.

5-DOF Arduino robot arm with smartphone app, SCARA pick-and-place, Mecanum wheel rover, Ant hexapod, and 3D printed harmonic drive gearboxes.

6-course specialization on configuration space, twists, screws, wrenches, Product of Exponentials (PoE) kinematics, dynamics, and KUKA youBot capstone.

Official code library supporting the Modern Robotics textbook in Python, MATLAB, and Mathematica. All algorithms for kinematics, dynamics, and trajectory generation.

20k+ star repository containing clean Python implementations of EKF/UKF Localization, Particle Filters, FastSLAM, A*, RRT*, Informed RRT*, Pure Pursuit, and MPC.

Curated repository of top community robot builds: self-balancing bots with PID, flex-sensor bionic hands, LiDAR maze rovers, and spider hexapods.