23 episodios

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 nonlinear systems without complete control authority requires methods that can reason about and exploit the natural dynamics of our machines.

This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods. Topics 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 include examples from biology and applications to legged locomotion, compliant manipulation, underwater robots, and flying machines.

License: Creative Commons BY-NC-SA

Underactuated Robotics MIT

    • Tecnología

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 nonlinear systems without complete control authority requires methods that can reason about and exploit the natural dynamics of our machines.

This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods. Topics 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 include examples from biology and applications to legged locomotion, compliant manipulation, underwater robots, and flying machines.

License: Creative Commons BY-NC-SA

    • video
    Lecture 1: Introduction

    Lecture 1: Introduction

    This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

    • 1h 14 min
    • video
    Lecture 2: The Simple Pendulum

    Lecture 2: The Simple Pendulum

    This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

    • 1h 8 min
    • video
    Lecture 3: Optimal control of the double integrator

    Lecture 3: Optimal control of the double integrator

    This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

    • 1h 16 min
    • video
    Lecture 4: Optimal control of the double integrator (continued)

    Lecture 4: Optimal control of the double integrator (continued)

    This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

    • 1h 24 min
    • video
    Lecture 5: Numerical optimal control (dynamic programming)

    Lecture 5: Numerical optimal control (dynamic programming)

    This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

    • 1h 12 min
    • video
    Lecture 6: Acrobot and cart-pole

    Lecture 6: Acrobot and cart-pole

    This course discusses nonlinear dynamics and control of underactuated mechanical systems, with an emphasis on machine learning methods.

    • 1h 20 min

Top podcasts en Tecnología

Otros usuarios también se han suscrito a

Más de MIT