Teaching Staff
- All
- General
- Project 1
- Project 2
- Project 3
- Website & Forum
- AWS & PyTorch
Announcements
10.02.2020
Lecture times were extended until 16:00, so instead of having 2 x 45min blocks we will now have 3 x 45min blocks.
As a result of this change, we have moved the exercise session to the morning.
They will now take place from 10:00 to 12:00 in
ETZ
D 61.1
and
D 61.2.
If you can't or don't want to use your personal laptop for the exercises, both rooms together offer around 60
linux workstations.
13.02.2020
There will be no exercises on 21.02.2020, exercises will start on 28.02.2020.
19.02.2020
We will be using
Piazza as a class discussion forum, all class participants will get an email with the access code soon.
The system is highly catered to getting you help fast and efficiently from fellow classmates, TAs and instructors.
Rather than emailing questions to the TAs or instructors, we encourage you to post your questions directly on Piazza - you can even do so anonymously.
From now on, all course relevant announcements will only be posted there.
Lectures
Date
Time
Room
Slides
Video
Topic
21.02.2020
13:15 - 16:00
Fundamentals of a Self-Driving Car
28.02.2020
13:15 - 16:00
Fundamentals of Deep Learning
06.03.2020
13:15 - 16:00
Fundamentals of Deep Learning (continued)
13.03.2020
13:15 - 16:00
Semantic Segmentation and Inertial Navigation System
20.03.2020
13:15 - 16:00
Depth Estimation
27.03.2020
13:15 - 16:00
Multi-tasking and 2D Object Detection
03.04.2020
13:15 - 16:00
3D Object Detection
24.04.2020
13:15 - 16:00
Localization
08.05.2020
13:15 - 16:00
Path Planning
15.05.2020
13:15 - 16:00
Motion Planning and Vehicle Control
22.05.2020
13:15 - 16:00
Imitation Learning and Reinforcement Learning
29.05.2020
13:15 - 16:00
Lane Detection and Maps
Exercises
Date
Time
Room
Slides
Video
Topic
28.02.2020
10:15 - 12:00
Getting Started with Amazon Web Services (AWS)
06.03.2020
10:15 - 12:00
at home
Project 1: Understanding Multimodal Driving Data
13.03.2020
10:15 - 12:00
Pytorch Tutorial and Q&A for Project 1
20.03.2020
10:15 - 11:00
Q&A for Project 1
23.03.2020
18:00 - 18:45
Q&A for Project 1
27.03.2020
10:15 - 11:00
Project 2: Multi-task learning for semantics and depth
31.03.2020
Tutorial: AWS, Git, Training Instances
03.04.2020
10:15 - 12:00
Q&A for Project 2
24.04.2020
10:15 - 12:00
Q&A for Project 2
08.05.2020
10:15 - 12:00
Q&A for Project 2
15.05.2020
10:15 - 12:00
Q&A for Project 2
22.05.2020
10:15 - 12:00
Q&A for Project 2
29.05.2020
10:15 - 12:00
Q&A for Project 2
Abstract
Autonomous driving has moved from the realm of science fiction to a very real possibility during the past twenty years,
largely due to rapid developments of deep learning approaches, automotive sensors, and microprocessor capacity.
This course covers the core techniques required for building a self-driving car, especially the practical use of deep learning through this theme.
Objective
Students will learn about the fundamental aspects of a self-driving car.
They will also learn to use modern automotive sensors and HD navigational maps, and to implement, train and debug their own deep neural networks
in order to gain a deep understanding of cutting-edge research in autonomous driving tasks, including perception, localization and control.
After attending this course, students will:
- understand the core technologies of building a self-driving car,
- have a good overview over the current state of the art in self-driving cars,
- be able to critically analyze and evaluate current research in this area,
- be able to implement basic systems for multiple autonomous driving tasks.
Content
We will focus on teaching the following topics centered on autonomous driving:
deep learning, automotive sensors, multimodal driving datasets, road scene perception, ego-vehicle localization, path planning, and control.
The course covers the following main areas:
- Foundation
- Fundamentals of Deep Learning
- Fundamentals of a Self-Driving Car
- Perception
- Semantic and Instance Segmentation
- Depth Estimation with Images and Sparse LiDAR Data
- 3D Object Detection with Images and LiDAR Data
- Object Tracking and Motion Prediction
- Localization
- GPS-Based Localization
- Visual Localization and LiDAR-Based Localization
- Path Planning and Control
- Path Planning
- Motion Planning and Vehicle Control
- Imitation Learning and Reinforcement Learning
Exercises
The exercise projects will involve training complex neural networks and applying them on real-world, multimodal driving datasets.
In particular, students should be able to develop systems that deal with the following problems:
- Sensor calibration and synchronization to obtain multimodal driving data,
- Semantic segmentation and depth estimation with deep neural networks,
- Learning to drive with images and map data directly (a.k.a. end-to-end driving).
Prerequisites
This is an advanced grad-level course. Students must have taken courses on machine learning and computer vision or have acquired equivalent knowledge.
Students are expected to have a solid mathematical foundation, in particular in linear algebra, multivariate calculus, and probability.
All practical exercises will require basic knowledge of Python and will use libraries such as PyTorch, scikit-learn and scikit-image.
Notice
Registration for this class requires the permission of the instructors.
Preference is given to EEIT, INF and RSC students.
Exam
Examiners:
Dengxin Dai, Alex Liniger
The grade is based on
- the realization of two projects (15% and 30%), and
- a 30 minutes oral exam during the session examination period (55%).
Successfully completing the projects is compulsory for attending the exam.
The projects will be group based but we assess the contribution of each student individually.
The examination is based on the contents of the lectures, the associated reading materials and exercises.
The performance assessment is only offered in the session after the course unit. Repetition only possible after re-enrolling.
Acknowledgement
We thank Amazon AWS and HESAI for sponsoring our education efforts,
and Toyota Motor Europe for sponsoring our autonomous driving research via the project TRACE Zurich.