# CMPE-633C-Spring2012

### From CYPHYNETS

(→Schedule) |
(→Schedule) |
||

Line 90: | Line 90: | ||

| align ="left" | Week 3. Feb 6 | | align ="left" | Week 3. Feb 6 | ||

| align ="left" | '''Lec 2.''' Largrangian mechnaics; variational principles; Hamilton's principle; Euler-Lagrange Equations; Legendre transformation; Hamiltonian mechanics; | | align ="left" | '''Lec 2.''' Largrangian mechnaics; variational principles; Hamilton's principle; Euler-Lagrange Equations; Legendre transformation; Hamiltonian mechanics; | ||

- | '''Lec 3.''' Holonomic and nonholonomic constraints; Lagrange D'Alembert's principle; external forces; dynamical and variational nonholonomic equations of motion; optimal control problem; rolling | + | '''Lec 3.''' Holonomic and nonholonomic constraints; Lagrange D'Alembert's principle; external forces; dynamical and variational nonholonomic equations of motion; optimal control problem; vertical rolling disk example; |

| align ="left" | | | align ="left" | |

## Revision as of 08:21, 13 February 2012

CMPE-633C / PHYS-401: Topics in Robotics and Control | |
---|---|

Spring 2012: Geometric Mechanics and Control |

## Instructor

Dr Abubakr Muhammad, Assistant Professor of Electrical Engineering

Email: abubakr [at] lums.edu.pk

Office: Room 9-311A, 3rd Floor, SSE Bldg

## Course Details

Year: 2011-12

Semester: Spring

Category: Grad

Credits: 3

Elective course for electrical engineering, computer engineering and physics majors

Course Website: http://cyphynets.lums.edu.pk/index.php/CMPE-633c-Spring2012

### Course Description

The aim of this course is to provide a unified treatment of nonlinear control theory and mechanical systems subject to motion constraints. Both Lagrangian and Hamiltonian formulations will be developed with an emphasis on solving control problems for nonholonomic and under-actuated systems. Important topics covered in this course include the center manifold theorem for stability, Frobenius and Chow’s theorems, Poisson geometry of nonholonomic systems, symplectic geometry of Hamiltonian flows etc. The students will also learn basic concepts of differential forms, Lie algebras, distributions and Riemannian geometry to formulate these results. Applications of this formulation span controller design in robotics, quantum information, aerospace and other systems that do not yield to non-geometric methods in control.

### Objectives

- Use of geometric frameworks to understand nonholonomic and under-actuated systems
- Lagrangian and Hamiltonian formulation of analytical mechanics
- Differential geometry methods in control problems
- Applications in robotics, systems theory and mathematical physics

### Learning Outcomes

- Link control theory with a geometric view-point of classical mechanics.
- Identify holonomic & nonholonomic constraints in physical systems.
- Understand control limitations due to under-actuation and motion constraints.
- Appreciate the value of abstract differential geometry methods in solving real-world problems.

### Pre-requisites

- CMPE-633b (Robot Dynamics & Control) OR
- PHYS-310 (Classical Mechanics) OR
- MATH-361 (Dynamical systems) OR
- By Permission of Instructor

**Self check.** Before start of class you should be familiar with configuration spaces, Euler-Lagrange equations, phase-plane analysis and fluent with use of vector calculus, linear algebra and differential equations.

**Who should take it?** EE/CMPE majors seriously interested in robotics and control theory; SSE students interested in pursuing mathematical physics.

### Text book

**Main text book.**

- Nonholonomic Mechanics and Control by Anthony Bloch, P. Crouch, J. Baillieul, J. Marsden. Interdisciplinary Applied Mathematics Springer-Verlag NY, 2003.

**Reference texts.**

- An Introduction to Differentiable Manifolds and Riemannian Geometry, Revised, Volume 120, Second Edition (Pure and Applied Mathematics) by William M. Boothby.
- Introduction to Mechanics and Symmetry: A Basic Exposition of Classical Mechanical Systems (Texts in Applied Mathematics) by Jerrold E. Marsden, Tudor S. Ratiu.
- Mathematical Methods of Classical Mechanics (Graduate Texts in Mathematics) by V.I. Arnold [ geometric mechanics]
- Riemannian Geometry by Manfredo P. do Carmo , Francis Flaherty.
- Geometric Control of Mechanical Systems: Modeling, Analysis, and Design for Simple Mechanical Control Systems by Francesco Bullo, Andrew D. Lewis.

### Grading Scheme

- Home-works (weekly): 25%
- Class Participation: 10%
- Paper Project: 40%
- Midterm Examination: 25%

## Schedule

WEEK | TOPICS | READINGS/REFERENCES |
---|---|---|

Week 1. Jan 23 | Lec 1. Configuration spaces; topology and geometry of important configuration spaces; visualization methods for S^n, RP^n, SO(n), T^n;
| |

Week 2. Jan 30 | No classes | |

Week 3. Feb 6 | Lec 2. Largrangian mechnaics; variational principles; Hamilton's principle; Euler-Lagrange Equations; Legendre transformation; Hamiltonian mechanics;
| |

Week 4. Feb 13 | ||

Week 5. Feb 20 |
| |

Week 6. Feb 27 |
| |

Week 7. Mar 5 | Lec 8.
| |

Week 8. Mar 12 |
| |

Week 9. Mar 19 |
| |

Week 10. Mar 26 |
| |

Week 11. April 2 |
| |

Week 12. April 9 |
| |

Week 13. April 16 |
| |

Week 14. April 23 | Lec 25.
| |

Week 15. April 30 |
| |

Week 16. May 7 | Lec 28.
| |

Week 17. May 14 |

### Project Ideas

- Parallel-parking a car: From Chow’s theorem to sub-Riemannian geometries.
- Quantum feedback control: Steering trajectories over SU(2^n).
- Discrete geometric mechanics for computer simulation.
- Berry’s phase in Foucault’s pendulum, cyclic adiabatic processes and other applications.
- Rigid body attitude stabilization: Time-optimal control on SO(n).
- Robotic grasping: Contact kinematics in rolling bodies.
- Cars, trailers and roller racers: Motion planning algorithms with nonholonomic constraints.