|
|
[Announcements] [Syllabus] [Project teams] [Homeworks] [Grading] [Class Schedule/Lecture Notes] |
Instructor Dr. Sethu Vijayakumar 2107F, JCMB, The King's Buildings, School of Informatics, Univ. of Edinburgh |
Announcements (Spring 2007)
|
Description Control of complex, compliant, multi degree of freedom (DOF) sensorimotor systems like humanoid robots or autonomous vehicles have been pushing the limits of traditional control theoretic methods. This course aims at introducing adaptive and learning control as a viable alternative. The course will take the students through various aspects involved in motor planning, control, estimation, prediction and learning with an emphasis on the computational perspective. We will learn about statistical machine learning tools and methodologies particularly geared towards problems of real-time, online learning for sensorimotor control. Issues and possible approaches for multimodal sensor integration, sensorimotor transformations and learning in high dimensions will be discussed. This will be put in context through exposure to topics in human motor control, experimental paradigms and the use of computational methods in understanding biological sensorimotor mechanisms |
Syllabus |
|
|
Fundamental Control Theory Classical Control: PD, PID Model based vs Direct control Limitations of traditional control (Why learning control?) |
|
Machine Learning Tools for
Learning Control Intrinsic Dimensionality & Dimensionality Reduction Multiple Model Learning Synergistic Control and Activation Coordinate Transformations: Body centric, Retinotopic, Object Centered |
|
Adaptive and
Learning Control Trajectory Planning, Inverse Kinematics and Inverse Dynamics Nonparametric methods for learning Real time and online learning Distal Learning Problem |
|
Predictive Control Kalman filters, Extended Kalman Filters Particle Filters |
|
Movement Primitives Dynamical Systems as Movement Policies Extracting, Tuning and Learning Movement Primitives Rhythmic vs Point-to-point primitives |
|
Biological & Human Motor Control Force Field Hypothesis, Equilibrium Point Hypothesis Internal Models Tuning Curves and Force Adaptation |
|
Sensorimotor Integration Bayesian Cue Integration Multimodal sensor fusion Explaining away and Cue reliability based integration |
Class Schedule & Lecture Notes
(Check out
this page)
Class Format The course consists of lectures
with discussions, reading assignments, and homework assignments. One of
these assignments might involve a group mini-project. There
will be a final exam covering the basics of the course.
Textbook & Reading Materials
Prerequisites All students are required
to abide by the regulations and guidelines laid down by the University of
Edinburgh. Various documents relating to the guidelines can be viewed at
the UoE
regulations & guidelines page for students. If you have any questions about the responsibilities
of either students, faculty, or graders under this policy, contact
the instructor or the
ITO.
|