RobotForge

The curriculum

Robotics, end to end

The study resource we wish existed when we started. Thirteen tracks — from the math you actually need to the latest VLA models — mapped against Modern Robotics, UPenn, Columbia, Thrun, and the modern toolchain. Parity where it earns its place, depth where it matters, and the topics nobody else teaches.

136 published0 planned13 tracks100% free, no paywall, forever
01

Foundations

The math, code, and tooling every roboticist uses. Calibrated for builders, not mathematicians.

10 / 10 lessons
10 ready
02

ROS 2

The de-facto robotics middleware. Nodes, topics, services, launch, TF, URDF, Nav2, MoveIt — the full working knowledge.

12 / 12 lessons
12 ready
03

Kinematics & Dynamics

Where the robot is, how fast it's moving, and what forces it takes to make it move. The mechanical spine of the field.

10 / 10 lessons
10 ready
04

Control

Making the robot do what you want despite physics, uncertainty, and bad luck. From PID to MPC to impedance.

10 / 10 lessons
10 ready
05

Perception & Computer Vision

How the robot sees. Cameras, lidar, classical CV, and the deep models that power modern robot perception.

10 / 10 lessons
10 ready
06

State Estimation & SLAM

Where is the robot, and what does the world around it look like? The probabilistic backbone of autonomous mobility.

10 / 10 lessons
10 ready
07

Motion Planning

Getting the robot from A to B without hitting anything, violating limits, or burning actuators. From A* to trajectory optimization.

10 / 10 lessons
10 ready
08

Manipulation

The hardest open problem in robotics. Grasping, pick-and-place, and the dexterity that still separates robots from humans.

10 / 10 lessons
10 ready
09

Mobile & Legged Robotics

From differential drive to humanoid locomotion. Getting a body through the world on wheels, legs, or wings.

10 / 10 lessons
10 ready
10

Learning for Robotics

Imitation, reinforcement, and the new wave of foundation models. The field that's redefining what robots can do.

12 / 12 lessons
12 ready
11

Simulators

Where most robotics actually happens in 2026. The six simulators worth learning and the patterns they share.

10 / 10 lessons
10 ready
12

Embedded & Hardware

The atoms side. Motors, encoders, microcontrollers, power, and the edge-AI hardware that closes the physical loop.

12 / 12 lessons
12 ready
13

Frontiers

The topics most curricula ignore — and that separate a serious roboticist in 2026. Teleop rigs, VLA fine-tuning, humanoid whole-body, tactile, safety, cost engineering, and careers.

10 / 10 lessons
10 ready

How this curriculum was built

We mapped the curricula of the top robotics learning resources on earth — Coursera Modern Robotics (Northwestern), UPenn and Columbia robotics programs, MIT Underactuated Robotics, Stanford CS223A, The Construct, Duckietown, Peter Corke's Robotics, Vision & Control, Thrun's Probabilistic Robotics, LeRobot, and Isaac Lab — then took the union. Every topic they cover shows up here. We then added a Frontiers track for the topics they miss: humanoid whole-body control, VLA fine-tuning, teleop rigs, tactile sensing, BOM engineering, and career building. Published lessons are full long-form articles. Planned lessons are stubs with target summaries — we ship them one by one, honest about what's ready.