Curated index of free + open educational material covering the ground OpenLoco builds on: dynamics, control, planning, perception, RL, simulators, actuators, and open-hardware reference designs.

Inclusion criteria: fully free (open-access textbook, open license, free PDF, public lecture series), or canonical reference whose documentation is free even if the underlying tool is proprietary. Paywalled books like Probabilistic Robotics (Thrun et al.), Robotics: Modelling Planning Control (Siciliano), and Principles of Robot Motion (Choset et al.) are excluded.

PRs welcome to extend.

Foundational textbooks (theory + control + planning)

• Underactuated Robotics — Russ Tedrake’s interactive MIT textbook on dynamics and control of underactuated systems. Free online (CC, code BSD).
• Robotic Manipulation — Tedrake’s interactive textbook on perception, planning, control for manipulation. Free online (CC, code BSD).
• Planning Algorithms — Steven LaValle’s comprehensive planning textbook (configuration space, sampling-based, decision-theoretic). Free PDF, full book.
• Reinforcement Learning: An Introduction (2nd ed.) — Sutton & Barto canonical RL textbook. Free PDF (author posted).
• Deep Learning Book — Goodfellow, Bengio, Courville. Free online HTML.
• Modern Robotics: Mechanics, Planning, and Control — Lynch & Park (Northwestern) free wiki + lecture videos. Book paywalled, code MIT-licensed at NxRLab/ModernRobotics.
• Probabilistic Machine Learning, Vol. 1 and Vol. 2 — Kevin Murphy. Free PDF drafts.
• Data-Driven Science and Engineering — Brunton & Kutz (UW) on dynamical systems, control, ML. Free PDF.

University courses with public materials

• MIT 6.832 Underactuated Robotics (OCW 2009) — Tedrake’s lecture notes, problem sets, videos. CC-BY-NC-SA.
• MIT 6.4210 Robotic Manipulation (OCW Fall 2022) — Tedrake’s manipulation course, with notebooks. CC-BY-NC-SA.
• Underactuated Robotics lecture playlist — Tedrake YouTube lectures. Free.
• Stanford CS223A — Introduction to Robotics — Khatib’s course on kinematics, dynamics, control.
• Berkeley CS287 — Advanced Robotics — Pieter Abbeel; lectures, notes.
• CMU Motion Planning (16-735) — Choset et al. course site with full lecture slides.
• ETH Robot Dynamics — Hutter / Siegwart RSL course; PDF script, ~400 pages.
• MIT 6.S191 — Intro to Deep Learning — open course, video + slides.
• Stanford CS231n — CNNs for Visual Recognition — canonical CV course notes.
• MIT 6.5940 — TinyML and Efficient Deep Learning — Han Lab; quantization, pruning, distillation.
• F1TENTH Course Kit — autonomous racing curriculum (MIT/UPenn). CC-BY.
• Duckietown documentation/courses — open self-driving educational platform.

Reinforcement learning / learning-based control

• Spinning Up in Deep RL — OpenAI’s pedagogical RL intro with code. MIT.
• Berkeley CS285 — Deep RL (Sergey Levine) — full lectures, slides, homework. Lecture playlist.
• Deep RL Bootcamp (Berkeley 2017) — Abbeel et al. lectures.
• Hugging Face Deep RL Course — practical RL with stable-baselines3 and CleanRL.
• Gymnasium — Farama Foundation’s maintained fork of OpenAI Gym; standard RL environment API. MIT.
• Stable-Baselines3 — reliable PyTorch RL algorithm implementations. MIT.
• PettingZoo — multi-agent RL environment standard. MIT.
• Minari — offline RL dataset standard (Farama). MIT.
• rsl_rl — fast PPO implementation tuned for legged-robot training. BSD-3.
• legged_gym — Isaac Gym training pipeline for ANYmal, Cassie, etc. BSD-3.
• walk-these-ways — MIT learning-based locomotion controller, full training code. MIT.
• gym-pybullet-drones — quadrotor RL environments (UTIAS DSL). MIT.

Simulators with open documentation

• MuJoCo — DeepMind physics engine for robotics/RL, now open source. Apache 2.0. User manual.
• MJX — JAX/GPU port of MuJoCo for massively parallel RL. Apache 2.0.
• MuJoCo Menagerie — high-quality URDF/MJCF assets for many robots (Spot, ANYmal, Franka, etc.). Apache 2.0.
• MuJoCo Playground — DeepMind RL training environments on MJX. Apache 2.0.
• dm_control — DeepMind Control Suite + MJCF Python bindings. Apache 2.0.
• Drake — Tedrake / MIT model-based design and verification toolbox. BSD-3. Source, C++ API.
• Brax — Google JAX-native rigid-body physics for accelerated RL. Apache 2.0.
• Genesis — multi-physics differentiable simulator (CMU/Tsinghua/MIT). Apache 2.0.
• Habitat — Meta AI 3D simulator for embodied agents. MIT.
• Isaac Lab — NVIDIA’s GPU-accelerated robot learning framework on Isaac Sim. BSD-3.
• Isaac Sim docs — NVIDIA full Omniverse robotics simulator docs. Free use, proprietary engine.
• Gazebo — successor to classic Gazebo, used with ROS 2. Apache 2.0. Docs (Harmonic). Classic Gazebo.
• Bullet / PyBullet — Coumans’ physics engine widely used in RL benchmarks. zlib.
• ARGoS — multi-robot swarm simulator (IRIDIA Brussels). MIT.
• SimBenchmark — comparison suite for MuJoCo/Bullet/ODE/RaiSim accuracy.

Actuator + motor hardware references

• mjbots Moteus — open brushless servo controller; full hardware/firmware/protocol. Apache 2.0. Reference manual.
• SimpleFOC — Arduino-grade FOC library + extensive theory docs. MIT.
• ODrive Robotics — high-performance BLDC controller with deep tuning docs. MIT (firmware).
• ODRI Open Robot Actuator Hardware — full open-source quasi-direct-drive actuator (motor + driver + transmission). BSD-3.
• Trinamic / Analog Devices — motor driver landing page (datasheets/app notes free).
• Robotis Dynamixel e-Manual — protocol docs, control table, tutorials.
• Maxon Academy — free training pages on DC/BLDC motor sizing.
• T-Motor — drone/legged motor specs and datasheets.
• Field-Oriented Control (Wikipedia) — theory primer with citations. CC-BY-SA.

Open hardware reference designs (robots)

• Stanford Pupper / StanfordQuadruped — open quadruped reference (Stanford Robotics Club). MIT.
• ODRI Solo — Open Dynamic Robot Initiative quadruped (MPI / NYU / MPI-IS). BSD-3.
• MIT Mini Cheetah / Cheetah-Software — original Cheetah-3 / mini-cheetah codebase. MIT.
• CHAMP — quadruped controller framework supporting Mini Cheetah and others. BSD-3.
• ANYmal C URDF — ANYbotics-released ANYmal C model. BSD-3.
• Spot in MuJoCo Menagerie — Boston Dynamics Spot model. Apache 2.0 (model files).
• Upkie — open-source wheeled-biped (Stéphane Caron). Apache 2.0.
• LeRobot — Hugging Face end-to-end robot learning stack with hardware integrations. Apache 2.0.
• Bitcraze Crazyflie firmware — open nano-quadrotor flight stack. GPL-3.
• Trossen Interbotix — ROS packages for Interbotix arms (ViperX, etc.). BSD-3.
• Pollen Reachy 2 SDK — open SDK for the Reachy 2 humanoid arm/torso. Apache 2.0.

ROS / middleware / tooling

• ROS 2 documentation (Humble) — official current LTS docs. CC-BY.
• MoveIt 2 — motion planning framework for ROS 2 manipulators. BSD-3.
• Nav2 — ROS 2 navigation stack documentation. Apache 2.0.
• urdf_tutorial — official ROS URDF tutorial repo. BSD.
• URDF XML spec — formal URDF tag reference. CC-BY.
• robot_localization — EKF/UKF state estimation package for ROS. BSD.

Probabilistic / state estimation / SLAM

• Cyrill Stachniss YouTube — Bonn professor’s full SLAM/photogrammetry/robotics lecture series. SLAM playlist.
• IPB Bonn teaching page — Stachniss group lecture materials and slides.
• UZH Robotics & Perception Group teaching — Scaramuzza’s VO/VIO/SLAM courses (slides + videos).
• ORB-SLAM3 — state-of-the-art open visual-inertial SLAM. GPL-3.
• Google Cartographer — real-time 2D/3D LIDAR SLAM. Apache 2.0. Docs.
• Kimera-VIO — MIT-SPARK visual-inertial odometry with loop closure. BSD-2. Full Kimera stack.
• A-LOAM — advanced implementation of LOAM LIDAR odometry (HKUST). BSD.
• OpenSLAM — historical archive of canonical SLAM algorithm implementations (gmapping, GraphSLAM, etc.).

Algorithm libraries + canonical reference implementations

• PythonRobotics — Atsushi Sakai’s huge collection of clean Python implementations of canonical robotics algorithms. MIT. Algorithm explanations.
• Peter Corke Robotics Toolbox (Python) — kinematics/dynamics/trajectory toolkit. MIT.
• spatialmath-python — SE(3) / SO(3) math companion library. MIT.
• mpc.pytorch — differentiable MPC reference implementation (Kolter lab). MIT.

Lecture series

• Steve Brunton (UW) — control theory, Koopman, SINDy. Control bootcamp playlist.