Module 06-13520 (2019)

Artificial Intelligence is concerned with mechanisms for generating intelligent behaviour. When this behaviour occurs in the everyday physical world, with its uncertainty and rapid change, we find that all kinds of new problems and opportunities arise. We will try to understand some of these in the context of robotics. In a series of lectures we will look at some theories of how to sense the real world, and act intelligently in it. In a series of labs you will build your own robots to see how well (or badly) these theories actually work.

• give an appreciation of the issues that arise when designing complete, physically embodied autonomous agents
• introduce some of the most popular methods for controlling autonomous mobile robots
• give hands on experience of engineering design
• encourage independent thought on possible cognitive architectures for autonomous agents

1. design, build and program simple autonomous robots
2. implement standard signal processing and control algorithms
3. describe and analyse robot processes using appropriate methods
4. write a detailed report on a robot project
5. carry out and write up investigations using appropriate experimental methods

• Note 1 There is a limit on the number of students allowed to take this module.

Available only to School of Computer Science students; registration limited to approx. 42 in combination with 06-15267 (Intelligent Robotics (Extended)). May not be taken by anyone who has taken or is taking 06-15267 (Intelligent Robotics (Extended)).

Approximately 18 lectures and 24 laboratory sessions

Contact Hours: 42

Sessional: continuous assessment (100%).

Supplementary (where allowed).

The continuous assessment consists of a writeup and demonstration of an autonomous robot project designed and constructed by each team.

1. Introduction
   • What is robotics?
   • Robotics and AI
   • Embedded Systems
   • Agent-Task-Environment model
   • Embodied Systems
   • Synthetic approaches to science
2. Sensors and signal processing
   • Common sensors and their properties
   • 1D signal processing
   • Vision
3. Planning approaches to robot control
   • STRIPS and SHAKEY
   • Robot kinematics
   • Limitations of planning approaches
4. Control Theory
   • Feedback, feedforward and open loop control
   • Linear first order lag processes
   • Limitations of control theory
5. Probability Based Approaches
   • Markov Decision Processes (MDPs)
   • Partially Observable Markov Decision Processes
   • Navigation using POMDPs
6. Kinematics and Motion Planning
   • Kinematics of differential drive robots
   • Probabilistic road map planning
7. Behaviour-Based Control
   • The subsumption architecture
   • Hybrid architectures
   • Formalising behaviour based control
8. Adaptive approaches to robot control
   • Reinforcement learning for control
   • Model based learning approaches to control
   • Learning maps
   • Evolutionary approaches
9. Architectures for control
   • CAS
   • ROS
10. Current research topics
    • Learning for manipulation
    • Gaze control
    • Planning visual search