6.881  Underactuated Robotics

Robots today move far too conservatively, using control systems that attempt to maintain full control authority at all times. Humans and animals move much more aggressively by routinely executing motions which involve a loss of instantaneous control authority. Controlling systems without full authority requires a different style of control, and requires that we rely on and exploit the natural dynamics of our machines.

This course introduces nonlinear dynamics and control in the context of underactuated mechanical systems, with an emphasis on machine learning control methods. Topics will include nonlinear dynamics of passive robots (walkers, swimmers, flyers), motion planning, partial feedback linearization, energy-shaping control, analytical optimal control, reinforcement learning/approximate optimal control, and the influence of mechanical design on control. Discussions will include examples from biology and applications to legged locomotion, compliant manipulation, underwater robots, and flying machines.

Prerequisites:
Introductory class on robotics (6.141, 2.12, 2.165 or equivalent), or permission of instructor. 6.302, 6.867 recommended. Familiarity with MATLAB is also recommended.

Workload:
Coursework will involve problem sets and a final project. There will not be a final exam.