Description

Have you ever driven a car? If not, have you ever sat in a car? Tried to avoid getting hit by a car while walking across the street? Tried to design a car?

This course will explore some of the issues around this classic and familiar example. (although this material and the underlying theory can be applied to many domains, including medical devices, aerospace engineering, biologicial systems, and others)

It is not your typical course on automatic controls. While we will achieve a high level of rigour in the theory and analysis of classical and modern feedback control, we will focus on design issues centered around the interaction between humans and autonomy and/or the interactions between autonomous systems and other autonomous systems. The development of, for instance, linear quadratic regulation (or estimation) will be grounded in familiar case studies.

Intended Learning Objectives

1. Design planners and controllers using state-space, search, and optimization methods

   1. Indicate the robustness of design

   2. Indicated generality of algorithms

2. Design automated planning and perception systems

   1. Centralized planning

   2. Decentralized planning

   3. Sharing of information

3. Understand impact of implementation issues

   1. nonlinearity, delay, noise

   2. human usage and interactions

   3. interactions with physical systems and other autonomy

Expectations and Policies

• Graduate students will do all the work of the undergraduates,

  • plus additional problems on homework sets and

  • short class project due at the end of the semester

• Honor code; some assignments may be collaborative

Schedule