Control, Robotics, and Autonomous Systems¶

Audience¶

Learners integrating dynamics, estimation, planning, and manipulation into autonomous systems

Final outcome¶

Close the perception–planning–control loop in simulation or on safe hardware, reporting stability, error, and failure modes.

Stages¶

Mathematics and dynamics¶

Selection rule: Complete all 4 required courses.

• Differential Equations — Required; MIT; Mainline; S
• Linear Algebra — Required; MIT; Mainline; S
• Classical Mechanics — Required; MIT; Mainline; S
• Introduction to Linear Dynamical Systems — Required; Stanford University; Mainline; S

Stage exit criterion: Derive a multistate system from physical assumptions, identify its parameters, and validate on a trajectory excluded from fitting; normalized state-prediction error must be below 10%, with controllability and observability rank checks completed.

Feedback and optimal control¶

Selection rule: Complete all 1 required course and choose 1 of 2 elective options. The other course is an optional supplement and does not count toward the elective requirement.

• Feedback Systems — Required; MIT; Mainline; A
• Dynamic Systems and Control — Elective option; MIT; Alternative; A
• Feedback Systems: An Introduction for Scientists and Engineers — Optional supplement; Caltech; Mainline; S; Audit review
• Dynamic Programming and Stochastic Control — Elective option; MIT; Supplement; A

Stage exit criterion: Implement a classical or optimal controller for one plant, recording gain/phase margins, overshoot, and settling time; stability must hold across at least 100 parameter perturbations, with the worst-performing sample explained.

Robotic systems¶

Selection rule: choose 1 of the 2 complete paths below and finish every course in the selected path in the listed order.

Complete path option — MIT Robotics path (complete in the listed order)

1. Robotic Manipulation — Course in selected path; MIT; Mainline; S
2. Underactuated Robotics — Course in selected path; MIT; Mainline; S

Complete path option — Complete Modern Robotics path (Courses 1–6 in order; full platform access may be paid) (complete in the listed order)

1. Modern Robotics, Course 1: Foundations of Robot Motion — Course in selected path; Northwestern University; Alternative; A
2. Modern Robotics, Course 2: Robot Kinematics — Course in selected path; Northwestern University; Alternative; A
3. Modern Robotics, Course 3: Robot Dynamics — Course in selected path; Northwestern University; Alternative; A
4. Modern Robotics, Course 4: Robot Motion Planning and Control — Course in selected path; Northwestern University; Alternative; A
5. Modern Robotics, Course 5: Robot Manipulation and Wheeled Mobile Robots — Course in selected path; Northwestern University; Alternative; A
6. Modern Robotics, Course 6: Capstone Project, Mobile Manipulation — Course in selected path; Northwestern University; Supplement; A

Stage exit criterion: Close a perception-planning-control loop in simulation or on a safe platform, achieving at least 90% task success and zero collisions over twenty perturbed trials; submit the trajectory-error distribution and a failure-mode review.

Execution rules¶

• Follow each stage's selection rule: complete every required course and the stated number of electives; when complete path options are provided, choose one and finish every course in its listed order; use optional supplements only to close a specific gap.
• Produce at least one reproducible artifact per stage and include failed attempts in the retrospective.
• Work involving mains voltage, high voltage, RF exposure, lasers, chemicals, or fabrication equipment requires local-law compliance and qualified supervision.