MOB: Course Materials

flittlet — Tue, 01 Aug 2023 17:05:58 +0000

MOB: Course Materials flittlet Tue, 01/08/2023 - 18:05

Lecture Recordings

All lecture recordings should be accessed via Learn; you will need to log in using your EASE account. (Learn provides you with access to any lecture recordings available for this course. You will need to select the "lecture recording" link once, before you can access any direct links to a lecture recording.)

License

MOB: Introduction to Mobile Robotics

flittlet — Tue, 01 Aug 2023 17:05:57 +0000

MOB: Introduction to Mobile Robotics flittlet Tue, 01/08/2023 - 18:05

Welcome to Introduction to Mobile Robotics

My name is Chris Lu and I am your course organizer for "Introduction to Mobile Robotics". I lead the "Mobile Autonomy Perception and Sensing" lab researching different layers of mobile robotic technology and you can find my homepage here.

Introduction

The past decade witnessed the fast-changing landscape and increasing ubiquity of mobile robots. Mobile robots, in various forms and presence, enter into our everyday life with new use cases unlocked by them every month. A representative example following the trend is autonomous driving, the ultimate system integration of cutting-edge hardware/sensors and unprecedently powerful machine learning and computer vision algorithms. The game has just started and far from being finished. And the question for you is - are you ready to embrace this onging trend and would like to shape the future of mobile robots or autonomous vehicles with your hands?

If you do, then this course will be a good start and suit your needs. During this course, you will learn and gradually develop skills in the analysis of mobile robots, being able to understand the perception, localization and navigation system for a self-driving car. The course is consisted with three main modules at the introductory level:

i. Visual Perception (3 weeks)

ii. Localization and Mapping (3 weeks)

iii. Motion Planning and Control (2 weeks)

Other two weeks in addition to the above will be used for optional drop-in sessions and course introduction/review.

Delivery Methods

The course will be delivered through a combination of: i. weekly lectures in person (will be recorded), ii. four (non-assessed) practical labs iii. two tutorials, and (4) an online discussion forum on Pizzaza.

This year, we will also have guest lecturer from the industry to share their insights of mobile robots in their work. The delivery also features the collabration with our award-winning EUFS team. They will also give us guest lectures and tell your how the methods to be introduced in this course help them build an autonomoous vehicle that won the UK Formula Student competetion in a row.

Assessment

Written Exam 60%, Coursework 40%. The (only) coursework will be released in the begining of November and you will have 4 weeks to finish and submit it.

Learning Outcomes

On successful completion of this course, you should be able to:

Understand the architecture and components of a mobile robot software stack.
Describe the pros and cons for a wide range of sensors used in mobile robotic perception: camera, LIDAR, GPS/INS, wheel odometry and radar.
Implement methods for static and dynamic object detection, segmentation, localization and mapping, behaviour and maneuver planning and robot control.
Demonstrate skills in robot operating systems, simulation tools and build programs with Python.

** NB: To avoid confusion, we want to stress that this is a totally different course compared to an outdated course Introduction to Vision and Robotics (INFR09019) ditched by the school. By saying different, we mean 0% overlapping content slide, coursework, and assessment way. The coursework of this course is also much lighter and has no dependency on ROS. We also do not cut down computer vision but will teach it in the context of deep learning (that's why we call it visual perception module) and autonomous vehicles - in a way the content of CV got more than INFR09019.

Course Outline

Why this course and what topics inside?

i. Visual Perception/ Computer Vision (3 weeks)

ii. Localization and Mapping (3 weeks)

iii. Motion Planning and Control (2 weeks)

More detailed syllabs and covered topics are provided in the timetable section below.

Delivery Method:

The course will be delivered through a combination of: i. weekly lectures in person (will be recorded), ii. four (non-assessed) practical labs iii. two tutorials, and iv. an online discussion forum on Pizzaza.

License