(09/07/20): IROS 2020 will not be held in-person. The safety and well-being of our participants is our priority. With the continued resurgence of COVID-19 within the State of Nevada, in particular Las Vegas, along with the city’s prohibiting large or even moderately-sized public gatherings, it makes it an impossibility to hold the in-person event as originally planned.


Instead, IROS On-Demand will be the platform to enjoyable and easily access IROS content anywhere, anytime, with any device. IROS On-Demand will be released Oct 25, 2020 and content will be available for 1-month minimum. More details will be released over the next few weeks.


IROS 2020 Tutorials

(last updated 09/30/20: Latest General FAQs (URL) here)


Registration for IROS 2020 On-Demand is free.


For those who already registered on Cvent, you have already been automatically signed-up for On-Demand; you have nothing else to do. All others must sign-up:

Visit IROS On-Demand Sign-Up to register for free access


IROS On-Demand (goes live Oct 25, 2020) is the platform to view pre-recorded videos of the 1400+ Technical Talks, Plenaries and Keynotes, and 60+ Workshops and Tutorials.

 

Full Day Tutorials

 

Title Organizers
From perception to planning and intelligence: A hands-on course on robotics design and development using MATLAB and Simulink    Roberto G. Valenti*, Anastasia Mavrommati*
 Space Robotics for In-Situ Resource Utilization: Needs, Challenges, and Approaches    William Beksi*, Joe Cloud, Rob Mueller, Kurt Leucht, Michael DuPuis, Issa Nesnas
 Introduction to Autonomous Racing with F1TENTH     Venkat Krovi*, Rahul Mangharam, Houssam Abbas, Matthew O'Kelly, Madhur Behl, Hongrui Zheng, Kim Luong

 

 

Title Organizers
Introduction to Bio-Inspired Robotics
Hamid Marvi*, Hosain Bagheri, Robert J. Full, Ronald S. Fearing, Elliot W. Hawkes, Michael T. Tolley, Sangbae Kim, Daniel Goldman, David Hu, Rolf Mueller
Next Generation of Haptic Devices for XR
Marcia O'Malley*, Zane Zook
F1/10 Competition at IROS2020
Venkat Krovi*, Rahul Mangharam, Houssam Abbas, Matthew O'Kelly, Madhur Behl, Hongrui Zheng, Kim Luong
Constraint programming for mobile robotics
Simon Rohou*, Raphael Voges
Introduction to Robotics Actuation and Motion Control
Andrew Willig*

 


Half-Day Tutorials

 

Title Organizers
Introduction to Deep Learning with AWS DeepRacer
William Harrison, Sahika Genc, Blake Hament
Deep Representation and Estimation of State for Robotics
Fei-Fei Li*, Danfei Xu, Kuan Fang, De-An Huang, Jeannette Bohg, Erwin Coumans, and Yunfei Bai
Tutorial on Deep Probabilistic Generative Models for Robotics
Takayuki Nagai*, Tadahiro Taniguchi, Takato Horii, Chie Hieida, Kaede Hayashi
Benchmarking Interactive Behavior Planning for Self-Driving Vehicles using BARK
Julian Bernhard*, Klemens Esterle, Patrick Christopher Hart, Tobias Kessler, Alois Knoll

 

 

Title Organizers
Mixed Reality and Robotics
Marc Pollefeys, Helen Oleynikova, Jeff Delmerico
REVIEW ON SCREW THEORY & GEOMETRIC ROBOT DYNAMICS A State-of-the-art Differential Geometry Algorithms for Robotics
Jose M. Pardos-Gotor*, Alberto Jardon, Volkert van der Wijk, Taeyoon Lee, Peter Corke, Patrick Wensing, Bruno Siciliano

 

 

 

Full Abstracts of Tutorials

 

Full Day Tutorials


From perception to planning and intelligence: A hands-on course on robotics design and development using MATLAB and Simulink

 

Roberto G. Valenti*, Anastasia Mavrommati*

 

During the tutorial, participants will have the opportunity to learn about the enabling technologies of a robotics system from a series of lectures around prototyping and deployment using MathWorks tools. We will present how Model-Based Design aligns with the development of advanced robotics applications. The tutorial will be comprised of four sessions: Perception, Planning, Machine Learning for Controls, and Implementation and Deployment. The first three sessions will have two parts: one lecture by an esteemed professor to introduce the main concepts of the topic and one hands-on tutorial by a MathWorks engineer who will show the use of MATLAB and Simulink towards that topic in a robotic system application. Finally, the fourth session will be dedicated to automatic code generation and hardware deployment. Moreover, during the same session, ROS connectivity through MATLAB and Simulink will also be explored.



Space Robotics for In-Situ Resource Utilization: Needs, Challenges, and Approaches

 

William Beksi*, Joe Cloud, Rob Mueller, Kurt Leucht, Michael DuPuis, Issa Nesnas

 

Experts from NASA's Kennedy Space Center and Jet Propulsion Lab will guide participants through concepts and operations of In-Situ Resource Utilization (ISRU) space robotics by way of presentations, demonstrations, and mini-tutorials. ISRU is the practice of extracting, processing, and storing useful resources found at a given location. On terrestrial planetary bodies such as the Moon and Mars, ISRU will enable the affordable establishment of space exploration and operations by minimizing the materials carried from Earth and by developing advanced, autonomous devices to optimize the benefits of available in-situ resources such as regolith. Regolith can be used as a building material to construct habitats and landing pads, or processed to extract Oxygen for breathing and rocket propellant, for example. The tutorials follow a theme discussing the needs for ISRU (what it is and proposed life-cycle), mobility (challenges and current approaches), excavation (challenges of digging regolith), and finally autonomy (communication issues and challenges unique to resource-denied environments). The session features a Gazebo-based simulation of a lunar excavator that the participants will install as part of the tutorials.



Introduction to Autonomous Racing with F1TENTH

 

Venkat Krovi*, Rahul Mangharam, Houssam Abbas, Matthew O'Kelly, Madhur Behl, Hongrui Zheng, Kim Luong

 

The deployment of new algorithms on autonomous vehicles is expensive, slow, and potentially unsafe. This tutorial presents the F1TENTH platform, a low-cost open-source 1/10th scale racecar and validated simulator which enables fast, safe, and rapid experimentation suitable for laboratory research settings. Participants will engage with the organizers in interactive sessions, learning how to use the simulator, write motion planning algorithms, and deploy them on the vehicle. Time will be allocated during the second half of the day for participants to prepare for a short race using the elements of the hands-on lab sessions.


Introduction to Bio-Inspired Robotics

 

Hamid Marvi*, Hosain Bagheri, Robert J. Full, Ronald S. Fearing, Elliot W. Hawkes, Michael T. Tolley,Sangbae Kim, Daniel Goldman, David Hu, Rolf Mueller

 

The Bio-inspired Robotics session will deliver an overview of this rapidly evolving field with an emphasis on recent developments. Its goal is to provide alternative solutions to existing real-world problems by looking at engineering and robotics from the perspective of biology and nature. With a panel of well renowned experts in the field, the session will cultivate points of inspirations and solutions for individuals of different backgrounds and discipline in both science and engineering. Get ready to be inspired, bioinspired!



Next Generation of Haptic Devices for XR

 

Marcia O'Malley*, Zane Zook

 

This session showcases the efficacy of dual haptic and and XR technologies for several potential applications. We combine hands on work with a set of overview lectures in order to teach participants about the basics of haptic interactions and of simple tools they can use to get started. The full tutorial will present haptics to new users to be able to incorporate in to robotic, XR or other applications. The lecture series will give participants an overview of the basic and the cutting edge in the field. The hands-on session will then give participants a cheap and easy method to incorporate haptics into the current research projects. Students with basic programming experience should be able to walk out of this session confident in their ability to design simple haptic interactions.



Introduction to Autonomous Racing with F1TENTH

 

Venkat Krovi*, Rahul Mangharam, Houssam Abbas, Matthew O'Kelly, Madhur Behl, Hongrui Zheng, Kim Luong

 

The deployment of new algorithms on autonomous vehicles is expensive, slow, and potentially unsafe. This tutorial presents the F1TENTH platform, a low-cost open-source 1/10th scale racecar and validated simulator which enables fast, safe, and rapid experimentation suitable for laboratory research settings. Participants will engage with the organizers in interactive sessions, learning how to use the simulator, write motion planning algorithms, and deploy them on the vehicle. Time will be allocated during the second half of the day for participants to prepare for a short race using the elements of the hands-on lab sessions.



Constraint programming for mobile robotics

 

Simon Rohou*, Raphael Voges

 

There are several ways to deal with state estimation in mobile robotics. The constraint programming approach consists in defining a problem as a set of rules and letting a solver perform the estimation. For mobile robotics, rules are constraints coming from state equations. The tutorial will stand on the Tubex library, that provides tools for computations over sets of reals and trajectories. It has been designed to deal with dynamical systems defined by non-linear differential equations and involving constraints such as trajectory evaluations, time uncertainties or delays. These computations stand on interval analysis, a well suited tool that reliably propagates uncertainties. In a nutshell, the tutorial brings the opportunity to apply new approaches based on interval analysis and constraint programming on classical problems of robot localization. It aims to be pedagogical in order to provide new concepts useful for the resolution of a wide class of problems.



Introduction to Robotics Actuation and Motion Control

 

Andrew Willig*

 

This tutorial will present technical approaches and underlying philosophies that have been developed over years of building and working with real robotics hardware but are rarely presented in textbooks, lectures, or research papers. In particular, we will present hardware and software tools that focus on providing flexible solutions to technical bottlenecks such as robust low-latency communications, motion control, and robot kinematics.

 

Another unique aspect of this tutorial is that presenters from both research and industrial applications will give short talks about how they have applied these tools in their work. These universities and companies have leveraged using physical hardware in order to advance research efforts, create systems that have real-world impact, and educate the next generation of roboticists.



 


Half-Day Tutorials



Introduction to Deep Learning with AWS DeepRacer

 

William Harrison, Sahika Genc, Blake Hament

 

This tutorial features an introductory talk on Deep Learning by William Harrison of the National Institute of Standards, “How I went from Zero to almost Hero with Deep Neural Networks.” From this introduction to Deep Learning, we will pivot to explore Reinforcement Learning. Dr. Sahika Genc of AWS will present the DeepRacer platform with a review of important machine learning concepts relevant to Reinforcement Learning and a hands-on introduction to training and racing with the DeepRacer.



Deep Representation and Estimation of State for Robotics

 

Fei-Fei Li*, Danfei Xu, Kuan Fang, De-An Huang, Jeannette Bohg, Erwin Coumans, and Yunfei Bai

 

Abstract coming soon!



Tutorial on Deep Probabilistic Generative Models for Robotics

 

Takayuki Nagai*, Tadahiro Taniguchi, Takato Horii, Chie Hieida, Kaede Hayash

 

Abstract coming soon!



Benchmarking Interactive Behavior Planning for Self-Driving Vehicles using BARK

 

Julian Bernhard*, Klemens Esterle, Patrick Christopher Hart, Tobias Kessler, Alois Knoll

 

BARK is an open-source benchmarking platform to foster a comparison of approaches in the field of interactive behavior planning for self-driving vehicles. BARK offers a unique approach with behavior models being used exchangeably for prediction, planning, and simulation. To grow the community around a common benchmarking platform, we want to enable participants to employ BARK within research and industry projects. In the first part of the tutorial, the attendees will get an introduction into the principles and pros and cons of state-of-the-art approaches in interactive behavior planning and understand the requirements on benchmarking platforms to systematically develop interactive behavior planners. In the second half, they will gain hands-on-experience with BARK and implement and benchmark their own interactive-behavior planner.



Mixed Reality and Robotics

 

Marc Pollefeys, Helen Oleynikova, Jeff Delmerico

 

Mixed, Augmented, and Virtual Reality offer exciting new frontiers in communication, entertainment, and productivity. A primary feature of Mixed Reality (MR) is the ability to register the digital world with the physical one, opening the door to a wide variety of robotics applications. This capability enables more natural human-robot interaction: instead of a user interfacing with a robot through a computer screen, we envision a future in which the user interacts with a robot in the same environment through MR, to see what it sees, to see its intentions, and seamlessly control it in its own representation of the world.

 

The purpose of this tutorial is to introduce the audience to both the high-level concepts of Mixed Reality and practically demonstrate how these concepts can be used to interact with a robot through an MR device. We will discuss how various hardware devices (mobile phones, AR/MR/VR headsets, and robots’ on-board sensors) can integrate with cloud services to create a digital representation of the physical world, and how such a representation can be used for co-localization. All participants will get a chance to create an iOS, Android, or Microsoft HoloLens 2 app to control and interact with a small mobile robot.



REVIEW ON SCREW THEORY & GEOMETRIC ROBOT DYNAMICS A State-of-the-art Differential Geometry Algorithms for Robotics

 

Jose M. Pardos-Gotor*, Alberto Jardon, Volkert van der Wijk, Taeyoon Lee, Peter Corke, Patrick Wensing, Bruno Siciliano

 

This tutorial will show how many robotics problems addressed today only with numerical iterative solutions, are solved much better with closed-form Geometric and Screw Theory algorithms. The approach is truly suitable for robot dynamics real time applications for their efficiency and effectiveness. The participants will get software, simulations and examples useful for implementations with many robotics structures and configurations.

 

The advantages of Screw Theory and Geometric approach for robot dynamics have been recognised for quite some time by practitioners, but in practice, these tools have been largely inaccessible because they require an entirely new language of notations (e.g., screws, twists, wrenches, conjugacy, spatial vector algebra, etc.). With this tutorial, the participants will increase their knowledge on these topics.