Algorithms and Systems for Autonomous Vehicles, Umeå University, Spring 2021

Zonghua Gu and Kalle Prorok

Lec Date	Lecture Notes	Topics Discussed	Additional Readings (Optional)
24/03	L1 Introduction and Overview PPTX, PDF	Background, history, AD processing pipeline	2020 Autonomous Vehicle Technology Report by Wevolver Self-Driving Cars Specialization at Coursera Artificial Intelligence for Robotics at Udacity, YouTube playlist Lecture: Self-Driving Cars (Winter 2020/21, Uni Bonn) (YouTube) ETHx DT-01x Self-Driving Cars with Duckietown
29/03	L1 Introduction and Overview cont.	Sensors and perception, HD maps, HW platforms, SW platforms, ethical issues, V2X communication
31/03	L2 Functional Safety PPTX, PDF	ISO 26262
07/04	L3 Intro to Machine Learning PPTX, PDF	Activation functions, Cross-Entropy Loss, SoftMax, ROC and AUC, NN training issues	A collection of the latest machine learning and deep learning courses Netron for NN topology visualization
12/04, 14/04, 19/04	L4.1 CNN and RNN PPTX, PDF	Intro to CNN, case studies (LeNet, AlexNet, VGGNet, GooLeNet, ResNet...)	UC Berkeley CS231n Convolutional Neural Networks for Visual Recognition A Comprehensive Guide to Convolutional Neural Networks: the ELI5 way Convolutional Neural Networks by DeepLearning.AI at Coursera
26/04, 28/04	L4.2 Object Detection and Segmentation PPTX, PDF	Object detection (R-CNN, Fast R-CNN, Faster R-CNN, Single-Stage Detectors), object segmentation	Deep Learning for Computer Vision, U Michigan, YouTube playlist
19/04, 21/04	L4.3 Adversarial Robustness PPTX, PDF	Adversarial attacks and defenses	NeurIPS 2018 tutorial, Adversarial Robustness: Theory and Practice, by Zico Kolter and Aleksander Madry PhDOpen: Aleksander Mądry, Machine Learning: A Robustness Perspective (YouTube)
05/05	L5 Planning PPTX, PDF	Route planning, behavior planning, local planning, Responsibility Sensitive Safety (RSS)	Self-Driving Cars: Planning (Benedikt Mersch 2020) (YouTube)
03/05	L6 Control Theory PPTX, PDF	PID, MPC, Udacity racetrack control, PID tuning with twiddle()	Lesson 15 PID Control, AI for Robotics at Udacity Controlling Self Driving Cars PID Control tutorial PIDs Simplified Understanding PID Control, Part 1: What Is PID Control? By MATLAB
10/05, 17/05, 19/05	L7.1 MDP Planning, PPTX, PDF	Markov Decision Process (MDP), Bellman Equations, Policy Iteration, Value Iteration	Reinforcement Learning: An Introduction, Richard S. Sutton and Andrew G. Barto Reinforcement Learning Specialization at Coursera (based on Sutton & Barto book) Reinforcement Learning MOOC by Chris G. Willcocks (based on Sutton & Barto book) UC Berkeley CS285 Deep Reinforcement Learning Reinforcement Learning Lecture Series 2018 by DeepMind x UCL
24/05, 26/05	L7.2 Value-based RL, PPTX, PDF L7.2.X Worked Examples, PPTX, PDF	Monte Carlo Methods, TD learning, Sarsa, Q learning, Dyna-Q, worked examples	Q-Learning, let us create an autonomous Taxi
26/05	L7.3 Policy-based RL, PPTX, PDF (not covered in exam)	Value-based RL with Function Approximation, Policy-based RL: Policy Gradient Theorem, MC REINFORCE, Actor-Critic	A friendly introduction to deep reinforcement learning, Q-networks and policy gradients (YouTube)
N/A	L8 Autonomous Driving with IL&RL, PPTX, PDF (not covered in exam)	Discussion of selected research papers	Kiran et al. Deep Reinforcement Learning for Autonomous Driving: A Survey, 2021
	Lecture Videos on YouTube

---

*Slides subject to change. Please download the latest version after class.

*Lectures on Mon & Wed 14:00-16:00 on Zoom. Link: https://umu.zoom.us/s/61403558067

*For your reference, you may look at the lecture videos from 2020 (contents are updated in 2021).

Lab Sections

	Assign Date	Assignment	Due Date	Additional Readings (Optional)
1	07/04	Lab1: Adversarial Attacks on Traffic Sign Classification (available in Canvas)	30/04	Transfer Learning for Image Classification with PyTorch & Python Tutorial | Traffic Sign Recognition, YouTube video, Blog UC Berkeley CS231n Python Numpy Tutorial (with Colab) T81 558:Applications of Deep Neural Networks by Jeff Heaton, YouTube playlist (Good intro to Python, CoLab, Tensorflow, Keras) Deep Neural Networks with PyTorch at Coursera PyTorch Notebooks/Tutorials Official tutorials: Colab, PyTorch Free Python Books
3	03/05	Lab2 PID Control https://github.com/guzonghua/saavlabs	17/05	Parameter Optimization w. twiddle() from Udacity Racetrack Control from Udacity
4	10/05	Lab3 DQN for Highway Driving, PPTX, PDF https://github.com/guzonghua/saavlabs	30/05	highway-env Script to Stop Google Colab From Disconnecting
5	04/06	Optional bonus lab	19/08

---

Anonymous feedback form (valid all semester)