ProjectFeatured
Isaac Lab
NVIDIA robot learning framework on Isaac Sim for large-scale parallel training and sim-to-real.
ProjectAdvanced
Drake
MIT robotics toolbox for modeling, planning, optimization-based control, and simulation.
Overview
Drake is MIT’s robotics toolbox for modeling, planning, optimization-based control, and simulation. It shines when you need mathematical rigor: mathematical programs, systems framework, and careful continuous dynamics.
Less of a “train a VLA overnight” tool, more of a “design a correct controller and hybrid system” tool.
Who it is for
Planning/control researchers; hybrid systems; teaching advanced robotics.
Key highlights
- Optimization-centric control
- Systems framework
- Strong academic pedigree
When to use
- Trajectory optimization, formal control design, teaching
When not to use
- Quick imitation learning on consumer GPUs only
Getting started
- 1Install Drake
- 2Run a manipulation or pendulum tutorial
- 3Formulate a small MathematicalProgram
How it compares
★ 7.8k
Caveats & pitfalls
- Heavier than pinocchio alone for simple FK needs.
Content reviewed 2026-07-23