Geometry and control foundations
Describe positions and orientations with named frames, then build toward verified simulated motion.
What you’ll explore
- Coordinate frames
- Forward kinematics
- Units
- Reachability
- 3D rotations
- Quaternions
Your lessons
0 / 21 completeDescribe where the robot is
Move a two-link arm, express points in named frames and distinguish ideal reach from verified motion.
Represent motion in three dimensions
Compare fixed and moving axes, matrix and quaternion descriptions, and the limits of a projected 3D picture.
Use a Jacobian
Predict small tool movements, inspect missing directions and compare inverse velocity proposals.
Turn a target into motion
Find alternative target-reaching postures and separate geometric reach, joint limits and obstacle contact.
Correct errors using feedback
Tune an ideal cart controller and inspect overshoot, velocity damping and delayed measurements.
Choose continuous actions
Train a small force controller, then compare continuous outputs, discrete choices, noise and saturation.
Stop when the controller leaves its limits
Inspect runtime rejection, delayed observations and bounded fallback without equating a stopped task with success.
Turn observations into estimates
Distinguish noisy sensor readings, delayed information, filtered estimates and simulator truth.
Fuse noisy measurements
Predict motion, correct with sensors and inspect the uncertainty assumptions of a linear Kalman filter.
Reason about forces and contact
Compare force and position commands, friction limits and the forces needed to hold a load.
Plan a collision-free trajectory
Search joint-space waypoints, try shortcuts and independently verify the full motion.
Enforce operating limits
Keep learned proposals separate from action checks, watchdogs, collision prevention and latched braking.
Collect useful robot demonstrations
Define observation/action timing, resets and whole-trajectory splits.
Predict actions from observations
Train an imitation policy and diagnose closed-loop distribution shift.
Predict several actions at once
Compare predicted action chunks with fresh-feedback execution.
Generate more than one valid action plan
Train conditional action denoising and independently verify sampled routes.
Connect vision language and action
Trace finite instruction features through a learned action policy.
Separate task planning from motor control
Use explicit skill preconditions, termination receipts and bounded recovery.
Vary the world during training
Compare randomized collection with separately changed evaluation conditions.
Evaluate a robotic system
Freeze policies and account for complete episodes, failures and exposure.
Transfer a policy to a richer embodiment
Compare planar velocity and 3D-engine torque contracts with explicit missing evidence.