Meet Paul Newman, Founder, Oxbotica & Professor at University of Oxford

Loading...

From the course by University of Toronto

Introduction to Self-Driving Cars

266 ratings

University of Toronto

Introduction to Self-Driving Cars

266 ratings

Course 1 of 4 in the Specialization Self-Driving Cars

Welcome to Introduction to Self-Driving Cars, the first course in University of Toronto’s Self-Driving Cars Specialization. This course will introduce you to the terminology, design considerations and safety assessment of self-driving cars. By the end of this course, you will be able to: - Understand commonly used hardware used for self-driving cars - Identify the main components of the self-driving software stack - Program vehicle modelling and control - Analyze the safety frameworks and current industry practices for vehicle development For the final project in this course, you will develop control code to navigate a self-driving car around a racetrack in the CARLA simulation environment. You will construct longitudinal and lateral dynamic models for a vehicle and create controllers that regulate speed and path tracking performance using Python. You’ll test the limits of your control design and learn the challenges inherent in driving at the limit of vehicle performance. This is an advanced course, intended for learners with a background in mechanical engineering, computer and electrical engineering, or robotics. To succeed in this course, you should have programming experience in Python 3.0, familiarity with Linear Algebra (matrices, vectors, matrix multiplication, rank, Eigenvalues and vectors and inverses), Statistics (Gaussian probability distributions), Calculus and Physics (forces, moments, inertia, Newton's Laws). You will also need certain hardware and software specifications in order to effectively run the CARLA simulator: Windows 7 64-bit (or later) or Ubuntu 16.04 (or later), Quad-core Intel or AMD processor (2.5 GHz or faster), NVIDIA GeForce 470 GTX or AMD Radeon 6870 HD series card or higher, 8 GB RAM, and OpenGL 3 or greater (for Linux computers).

From the lesson

Module 0: Welcome to the Self-Driving Cars Specialization!

This module will introduce you to the main concepts and layout of the specialization and discusses the major advances made in the field over the last two decades, highlighting the most recent progress made by major players in terms of safety and performance metrics, where available.

Meet the Instructor, Steven Waslander5:33

Meet the Instructor, Jonathan Kelly2:23

Meet Diana, Firmware Engineer2:54

Meet Winston, Software Engineer3:36

Meet Andy, Autonomous Systems Architect2:02

Meet Paul Newman, Founder, Oxbotica & Professor at University of Oxford5:05

Why Should You Take This Course?2:36

Meet the Instructors

Steven Waslander
Associate Professor
Aerospace Studies
Jonathan Kelly
Assistant Professor
Aerospace Studies

[MUSIC]

Hi, I'm Paul Newman.

I'm CTO and founder of Oxbotica, and

Professor of Information Engineering at the University of Oxford.

So I can remember the day that I got interested in self-driving vehicles and

autonomy, and it was actually in Sydney.

And I was talking to a PhD student, I'd just started my PhD,

and he introduced this problem called the SLAM problem.

So can a vehicle figure out where it is in the world without having been given a map,

and can it build a map itself?

And I knew then that was the mother of all problems I wanted to work on.

So I can remember, and it was the 19th of March 1996,

I really discovered that there was a big problem here.

And since then, it's never got old.

We know how to solve those problems now, but

I was always an engineer trying to build things.

And then I discovered that the most exciting way to build things was

a combination of software and mechanical things, and

that's almost by definition robotics, then I was in heaven.

So I've never really lost that from probably about nine, when I started

trying to program very basic computers through to what I do everyday now.

It's a thrill that's never left me.

Okay, so what do we do here at Oxbotica?

Well, we build the software that makes autonomous vehicles autonomous.

We go from the lowest level sensors all the way up, through the stack,

through planning, through localization, through control,

through perception up to fleet management systems.

So the whole software environment that makes fleets of vehicles autonomous

is what we build and we develop.

Now, down at the university I'm the Director of the Autorobotics Institute,

I've got great life there too.

So there I work on what are the hard, hard problems of autonomy and

robotics that are perhaps ten years down the line?

What are those really sort of nascent research questions of machines and

intelligence and

machine learning that we would want in our vehicles in the five to ten-year period?

So I've got a very privileged life.

So why do I care about self-driving cars so much?

Because transport's dumb, it's broken at the moment.

80% of all accidents in the road are caused by inattention.

The one thing that machines won't do is get bored.

So we don't deal with that in any way.

The way in which we drive is inefficient.

As we get older, as we get more infirm,

the one thing we need is to stay in contact with our species.

And if we say, yes, I'm sorry, you're eyesight's fading, can't drive.

If you had an epileptic fit, you can't drive.

So the whole edifice of how we drive ourselves and move our atoms,

be they human atoms or goods atoms, is just nuts.

It's derived from horses and carts.

We wouldn't do that if we had our technology now, so that's nuts.

And software and algorithms are going to change that.

I'm absolutely sure that we'll have more atoms move better and

safer, because of AV technology.

So there's a question about should the vehicles themselves be

independently smart or should they leverage smart infrastructure?

So what's smart infrastructure?

A traffic light that says, I'm about to turn red, and this is my location.

Or a roundabout, a rotary, that says,

this is where the vehicles are, your slot's coming up.

There's any number of ways we could think about instrumenting our cities.

So my view is that the vehicles themselves need to be smart

enough to deal without a dependency on infrastructure.

And if smart infrastructure is provided,

then sure, you can use it, if you can trust it.

And there's questions about cybersecurity, and that's throughout the whole stack,

really, the cybersecurity question.

So my view is the vehicle itself must be smart enough to deal,

just like you are as a driver.

You can look at traffic lights and you can look at lane boundaries, and you don't

need to be told by the car in front that there's a car in front, you can see it.

Because if that infrastructure is taken away,

you still want to be able to have an autonomous vehicle.

It could be augmented by infrastructure, but

I believe you shouldn't be dependent on it.

because if you solve that problem,

then you really do have all of transport in any place in grasp.

Might take some time, but you're in that trajectory.

And you're not dependent on expensive bespoken infrastructure.

And my view of expensive, bespoke infrastructure for

self-driving cars is called city.

[MUSIC]

Explore our Catalog

Join for free and get personalized recommendations, updates and offers.

Coursera provides universal access to the world’s best education, partnering with top universities and organizations to offer courses online.

© 2019 Coursera Inc. All rights reserved.

Download on the App Store

Get it on Google Play