AI & Robotics: Research Course SS 20 TU Berlin

Wednesdays, 10-12
Fridays, 14-16
Zoom invite: this invite link
Teaching Assistants
Jung-Su Ha
Ingmar Schubert
Papers, slides & exercises
topics material due on
Introduction 01-introduction
Reading: The TossingBot paper
e01-tossingBot April 29
Lecture: Differentiable AI 02-differentiableAI
Technical Exercise e02-endToEnd May 6
Reading: LeTS-Drive paper
e03-LeTS-Drive May 20
Review Exercise e04-LeTS-Drive-Review May 29
Lecture: Tree Search & POMDPs 03-TreeSearchPOMDPs
Technical Exercise: Tree Search & POMDPs e05-TreeSearchPOMDPs June 5
Reading: PoseRBPF paper
e06-PoseRBPF June 17
Lecture: Probabilities \& Filtering 04-probabilities
Technical Exercise: Filtering e07-filtering June 24
Review Exercise e08-PoseRBPF-Review July 1
Reading: GVF paper
e09-GVF July 8
Lecture: Model-based RL 06-RL
For your information, this is my top choice of conferences. But this course might focus on papers of RSS only.
Paper candidates (prelim)
  • First pass through RSS 2019:
    • Improvisation through Physical Understanding: Using Novel Objects As Tools with Visual Foresight Annie Xie, Frederik Ebert, Sergey Levine, Chelsea Finn
    • TossingBot: Learning to Throw Arbitrary Objects with Residual Physics Andy Zeng, Shuran Song, Johnny Lee, Alberto Rodriquez, Thomas A.Funkouser
    • DESPOT-Alpha: Online POMDP Planning with Large State and Observation Spaces Neha Priyadarshini Garg, David Hsu, Wee Sun Lee
    • Autonomous Tool Construction Using Part Shape and Attachment Prediction Lakshmi Velayudhan Nair, Nithin Shrivatsav Srikanth, Zackory Erikson, Sonia Chernova
    • A Differentiable Augmented Lagrangian Method for Bilevel Nonlinear Optimization Benoit Landry, Zachary Manchester, Marco Pavone
    • LeTS-Drive: Driving in a Crowd by Learning from Tree Search Panpan Cai, Yuanfu Luo, Aseem Saxena, David Hsu, Wee Sun Lee
    • BayesSim: Adaptive Domain Randomization Via Probabilistic Inference for Robotics Simulators Fabio Ramos, Rafael Possas, Dieter Fox
    • Differentiable Algorithm Networks for Composable Robot Learning Peter Karkus, Xiao Ma, David Hsu, Leslie Kaelbling, Wee Sun Lee, Tomas Lozano-Perez
    • DensePhysNet: Learning Dense Physical Object Representations Via Multi-Step Dynamic Interactions Zhenjia Xu, Jiajun Wu, Andy Zeng, Joshua Tenenbaum, Shuran Song
    • PoseRBPF: A Rao-Blackwellized Particle Filter for6D Object Pose Estimation Xinke Deng, Arsalan Mousavian, Yu Xiang, Fei Xia, Timothy Bretl, Dieter Fox
    • Leveraging Experience in Lazy Search Mohak Bhardwaj, Sanjiban Choudhury, Byron Boots, Siddhartha Srinivasa
    • Toward Asymptotically-Optimal Inspection Planning Via Efficient Near-Optimal Graph Search Mengyu Fu, Alan Kuntz, Oren Salzman, Ron Alterovitz
    • Learning to Plan with Logical Automata Brandon Arak, Kiran Vodrahalli, Thomas Leech, Cristian Ioan Vasile, Mark Donahue, Daniela Rus
    • End-To-End Robotic Reinforcement Learning without Reward Engineering Avi Singh, Larry Yang, Chelsea Finn, Sergey Levine
    • Learning Robotic Manipulation through Visual Planning and Acting Angelina Wang, Thanard Kurutach, Kara Liu, Pieter Abbeel, Aviv Tamar
  • Differentiable Physics and Stable Modes for Tool-Use and Manipulation Planning (RSS 2018)