surveys perception, cognition, and movement in humans, humanoid robots,
and humanoid graphical characters.
Application areas include more human-like robots, videogame characters,
and interactive movie characters.
Kinematics: forward and inverse kinematics (rotation, how to automatically
generate forward kinematics (mathematica), how to use optimization
to solve inverse kinematics (matlab))
Trajectory formation in humans and robots (splines, minimum jerk)
Dynamics and simulation (how to automatically generate inverse dynamics
equations (mathematica), forward dynamics, numerical integration,
ODE, sdfast, ...)
Control (springs, dampers, PD control, integral control,
LQR regulation (matlab))
Actuators: Muscles and motors, equilibrium position
Planning (A*, RRT)
Legged locomotion: walking and running (PDW, ZMP, foot placement)
Manipulation and hand control
Learning, including imitation learning, reinforcement learning, supervised learning (kin + dyn),
function approximation, neural nets, LWR,
policy search, DP, Q learning,
imitation learning as RL,
Hearing, Speech production and recognition
Tactile sensing, proprioception
Taste and smell?
Cognition, Reasoning, Making Decisions
Humanoids in literature
Philosophical and ethical issues
Goals of the course:
Introduce students to the wide range of research involved in humanoids.
Get students interested and involved in humanoids research.
No formal prerequisites.
Knowing a programming language such as C, Java, or Matlab
will be useful. Knowing what a vector and matrix are is useful.
Readings will be made available electronically.
The course will start with a series of assignments that everyone does, and
then you will do a longer term project.
There will be no exams. Your
grade will be based on assignments and your project.
There will be more doing than reading
in this course.
Some homeworks will consist of working with the elements of a
humanoid simulator, similar to programming a videogame.
We are still working out the details
of the homeworks.
To be described.