6205ICT: Advanced Topics in Information Technology A

Robotics

(SEMESTER TWO, 2016)

Identifying Information

 

Course catalogue no:

 6117CIT

Course title:

 Adv Topics in Information Technology

Field of Education Code

 Computer Science /Robotics

Program/s

2011Bachelor of Information Technology with Honours

 

 

 

Status of Course within program/s or academic plan/s

Elective, honours 

 

Credit point value

 10

Prerequisites:

 Enrolment in Honours Program

Year and semester:

Semester 2, 2014

Course convenor

Prof. Vladimir Estivill-Castro


Email:
V.Estivill-Castro@griffith.edu.au

Teaching team members:

Same as Course convenor

 

 

Objectives

 

The major aims and objectives of the subject are to introduce the student to the dynamic research in the area of software systems for mobile robotics.

Summary

  Developing software for a mobile robot is exciting and challenging. The software is controlling behaviour that has an impact on the real worlds, and should bridge in an intelligent manner the signals perceived in sensors to commands for the actuators. Thus, robotics brings together software architecture
concepts to integrate advances from artificial intelligence, agent technology, computer vision and pattern analysis and its developments are applicable to many scenarios including system control or computer games.  This course brings you to the opportunity of starting frontier research in mobile robotics by engaging in exciting projects in mobile robots.  We will cover essential concepts of robotics, but the focus would be in the software architecture to integrate intelligent capabilities like vision, motion, localisation, and planning.  By the end of the course you will understand basic software architectures for robotic control like feedback-loop control, open loop control, reactive architectures and behaviour based architectures. You have command of integration technologies like white-board architectures (or publisher subscriber services like in ROS) as well as finite-state-machines and its modalities. You would be exposed to mobile robotic challenges like localisation, simultaneous localisation and mapping (SLAM) as well as integration of planning and learning.

 

 

Content

 

Lecture 1A,- Introduction, research topics, challenges, logic-labelled finite-state machines

Lecture 1B.- The key quesitons in mobile robotics, Sense Act Cycle, What is localization

Lecture 2.- Middleware, Locomotion, Emergent Behavior

Lecture 3.- Kinematic Concepts, Feedback-Loop Control

Lecture 4.- Perception concepts, sensors (not vision)

Lecture 5.- Introduction to Computer Vision and Image segmentation

Lecture 6.- Features in Vision (correaltion,convolution), Edge Detection, Harris Corners SIFT points.

Lecture 7.- Vision, Object Recognition (bag of words approach)

Lecture 8.- Representingand propagating uncertanity, Vision:line and shape (circel, rectangle) recognition

Lecture 9.- Localization: Basic Bayesian Inference, Sensor Model. Analytical motion models, odometry.

Lecture 9B.- Maps, Behavior-Based Control and Toto. Maps and representations of the beleif regarding position/posture

Lecture 10.- Probabilistic Localization

Lecture 11.- Kalman Filter (a demo of one dimensional Kalman filter )

Lecture 12.- Intro to SLAM

 

 

 

 

 

 Prerequisites

Or any equivalent course that provides you a background in Object-Oriented Programming, with perhaps some emphasis into C++ (memory management).

Generic Skills Development

 

Emphasis will be placed in generic research skills. However, it will also palce emphasis on research topics in the development of software systems for mobile robots.

 

 

Assessment (confrim details in the near future.)

 

1.      5 laboratory challenges (20% each)

 

CHALLENGE 1 (pdf)

CHALLENGE 2 (pdf).

CHALLENGE 3 (pdf) Resources include.

  1. Original paper (M.J. Mataric. Integration of representation into goal-driven behavior-based robots. Robotics and Automation, IEEE Transactions on, 8(3):304-312, jun 1992),
  2. textbook with Chapter 16 (Think the Way You Act Behavior-Based Control) and
  3. ROS packages that include llfsms for stroll and avoid.

 

 

 

Texts and Supporting Materials

 

 

Videos for discussion