CS CS391R: Robot Learnin ing Perception, Decision Making, and General-Purpose Autonomy Prof. Yuke Zhu Fall 2020 CS391R: Robot Learning (Fall 2020) 1
Robotics and COVID-19 Photos from the Internet CS391R: Robot Learning (Fall 2020) 2
Today’s Agenda ● Overview of general-purpose robot autonomy ● Why studying Robot Learning now? ● Research topics of Robot Learning ● Course content and logistics ● Student introduction CS391R: Robot Learning (Fall 2020) 3
Special-Purpose Robot Automation Structured Environments Fixed Set of Tasks Pre-programmed Procedures CS391R: Robot Learning (Fall 2020) 4
General-Purpose Robot Autonomy … in the Wild Unstructured Environments Ever-changing Tasks Human Involvement CS391R: Robot Learning (Fall 2020) 5
Special-Purpose Robot Automation custom-built human expert special-purpose robots programming behaviors General-Purpose Robot Autonomy ? general-purpose general-purpose robots behaviors CS391R: Robot Learning (Fall 2020) 6
Special-Purpose Robot Automation custom-built human expert special-purpose robots programming behaviors General-Purpose Robot Autonomy Robot Learning general-purpose general-purpose robots behaviors CS391R: Robot Learning (Fall 2020) 7
General-Purpose Robot Autonomy: Im Imagi ginati tions Un Unim imate - The First st Indust strial Robot British TV (1968) CS391R: Robot Learning (Fall 2020) 8
General-Purpose Robot Autonomy: Ch Challe llenges DARPA Robotics s Challenge (2015) “The Moravec's paradox” CS391R: Robot Learning (Fall 2020) 9
General-Purpose Robot Autonomy: Progress ss We will learn the algorithms and techniques behind the latest progress. Grasp sping (DexNet 4.0; 2019) Locomot Locomotion on (ANYmal; 2019) Mani Manipul ulat ation on (OpenAI; 2019) CS391R: Robot Learning (Fall 2020) 10
What Makes Ro Robot Learning Special? Robots are physi ysically y embodied and envi vironmentally y si situated . [Sa et al. IROS 2014] [Levine et al. JMLR 2016] [Bohg et al. ICRA 2018] CS391R: Robot Learning (Fall 2020) 11
What Makes Ro Robot Learning Special? Robots are physi ysically y embodied and envi vironmentally y si situated . Perceive Perceive Act Act Act Perceive [Sa et al. IROS 2014] [Levine et al. JMLR 2016] [Bohg et al. ICRA 2018] CS391R: Robot Learning (Fall 2020) 12
A key challenge in Ro bot Learning is to close the pe oop . Robo percept ption-act action on loop Perceive Perceive Act Act Act Perceive [Sa et al. IROS 2014] [Levine et al. JMLR 2016] [Bohg et al. ICRA 2018] CS391R: Robot Learning (Fall 2020) 13
Now is the best time to study and work on Robot Learning. Ar Artificial Intelligence Computin Co ing Powe wer Ro Robot Hardware Recent breakthroughs in machine Your smartphone is millions of times More reliable and affordable learning and computer vision, e.g., deep more powerful than all of NASA’s cobot hardware that costs around learning (Turing awards 2018) combined computing in 1969. annual salary of American workers CS391R: Robot Learning (Fall 2020) 14
Now is the best time to study and work on Robot Learning. Positive and negative so societal impacts of robot learning research is an important part of our in-class discussions. https://www.therobotreport.com/tag/coronavirus/ CS391R: Robot Learning (Fall 2020) 15
Robot Learning as a Growing Research Community 60 Number of Papers (k) 50 40 6x 30 20 10 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Year Conference on Robot Learning is 3 years old. Growth of “ Ro Robot Learni Learning ng ” Publications [Source: Google Scholar] CS391R: Robot Learning (Fall 2020) 16
When NOT to Make Robots Learn? Learning is not for every problem in robotics. Harnessing the priors and structures of a Learning is most effective when used in problem goes a long way… conjunction with modeling. CS391R: Robot Learning (Fall 2020) 17
When to Make Robots Learn? Learning is critical for getting robots to work in the real world. object variation environment uncertainty adaptation CS391R: Robot Learning (Fall 2020) 18
You learn CS391R: Robot Learning so that Robots learn faster and better . CS391R: Robot Learning (Fall 2020) 19
Key Ingredients of General-Purpose Robot Autonomy Perception Decision Making Intelligence Real-World Systems seeing and understanding planning and control for cognitive reasoning & fast physical embodiment & 3D environments long-term interactions adaptation to new situations hardware constraints CS391R: Robot Learning (Fall 2020) 20
Course Content We review the Robot Learning literature in these topics. Part I Part II Part III Part IV Perception Decision Making Intelligence Real-World Systems seeing and understanding planning and control for cognitive reasoning & fast physical embodiment & 3D environments long-term interactions adaptation to new situations hardware constraints Prerequisi site: coursework / experience in AI and Machine Learning CS391R: Robot Learning (Fall 2020) 21
Course Content Part I Part II Part III Part IV Perception Decision Making Intelligence Real-World Systems seeing and understanding planning and control for cognitive reasoning & fast physical embodiment & 3D environments long-term interactions adaptation to new situations hardware constraints CS391R: Robot Learning (Fall 2020) 22
Course Content: Pe Percepti ption unsupervised visual learning object detection 3d point cloud visual tracking multimodal understanding recursive state estimation pose estimation interactive perception CS391R: Robot Learning (Fall 2020) 23
Course Content Part I Part II Part III Part IV Perception Decision Making Intelligence Real-World Systems seeing and understanding planning and control for cognitive reasoning & fast physical embodiment & 3D environments long-term interactions adaptation to new situations hardware constraints CS391R: Robot Learning (Fall 2020) 24
Course Content: Decisi sion Maki king model-free reinforcement learning model-based reinforcement learning imitation as supervised learning adversarial imitation learning inverse reinforcement learning CS391R: Robot Learning (Fall 2020) 25
Course Content Part I Part II Part III Part IV Perception Decision Making Intelligence Real-World System seeing and understanding planning and control for cognitive reasoning & fast physical embodiment & 3D environments long-term interactions adaptation to new situations hardware constraints CS391R: Robot Learning (Fall 2020) 26
Course Content: In Inte tellige gence compositionality: hierarchy learning to learn: learning to learn: meta-learning lifelong learning compositionality: task and motion causal reasoning CS391R: Robot Learning (Fall 2020) 27
Course Content Part I Part II Part III Part IV Perception Decision Making Intelligence Real-World Systems seeing and understanding planning and control for cognitive reasoning & fast physical embodiment & 3D environments long-term interactions adapting to new situations hardware constraints CS391R: Robot Learning (Fall 2020) 28
Course Content: Syst ystems simulation-reality gap data-driven robotic grasping building robotic systems CS391R: Robot Learning (Fall 2020) 29
Learning Objectives ● understand the potential and societal impact of general-purpose robot autonomy in the real world, the technical challenges arising from building it, and the role of machine learning and AI in addressing these challenges; ● get familiar with a variety of model-driven and data-driven principles and algorithms on robot perception and decision making; ● be able to evaluate, communicate, and apply advanced AI-based techniques to robotics problems. … through literature reviews , research presentations , and course projects CS391R: Robot Learning (Fall 2020) 30
Learning Objectives Get a taste of Robot Learning research in the full circle CS391R: Robot Learning (Fall 2020) 31
Logistics Lectures Time: 2:00-3:30pm CT, Tuesdays and Thursdays Location: Online (Zoom links on Canvas) Office Hours Instructor: TBA next week Fill in the time zone survey on Canvas! TA: TBA next week CS391R: Robot Learning (Fall 2020) 32
https://www.cs.utexas.edu/~yukez/cs391r_fall2020/ Logistics Instructor Lectures overview of research topics Student Presentations presentation of research papers Final Project Spotlights spotlight talks of course projects CS391R: Robot Learning (Fall 2020) 33
Logistics Required Readings (No Review) overview or survey papers with lectures Required Readings key papers that will be discussed in class Optional Readings recommended papers for in-depth reviews CS391R: Robot Learning (Fall 2020) 34
Recommend
More recommend
