Home

October 5, 2020, 8:30 am-5:30 pm Room 330B
Huntington Place, Detroit, USA

The workshop is Detroit Time GMT-4

Time Talk
8:20 Welcome
8:30 Talk 1 Dimitra Panagou, University of Michigan, “Prediction and Adaptation for Certified Constrained Control”
9:00 Talk 2 H. Jin Kim, Seoul National University, “Learning toward complex tasks with curriculum generation and automated resets”
9:30 Talk 3 Elena Arcari and Melanie Zeillinger, ETH Zurich,”Data efficient learning for model predictive control”
10:00 Coffee break and Poster Overview
10:30 Session 1 Contributed papers 5 minutes each 6 papers, check full list below
11:00 Talk 4: Giuseppe Loianno, NYU, “Learning and Physics Synergies for Agile, Safe, Resilient, and Collaborative Aerial Robots”
11:30 Talk 5 Kostas Alexis, NTNU, “Modular Navigation Learning for Aerial Robots”
12:00 Talk 6 Davide Scaramuzza, University of Zurich, “Champion Level Drone Racing using Deep Reinforcement Learning”
12:30 Session 2 Contributed papers 5 minutes each 6 papers, check full list below
13:00 Lunch Break
14:00 Talk 9 Kushal Jaligama, Shield AI: Bulding the World’s Best AI Pilot
14:30 Talk 10 Hayak Martirios, Skydio, “The Future of Aerial Autonomy in Industry”
15:00 Talk 11 R. Tapia and Anibal Ollero, University of Seville, “Event-based Vision for Ornithopter Perception and Autonomy”
15:30 Coffee Break and Poster Session
16:00 Talk 12 Marija Popovic, University of Bonn, “Learning for Active Robotic Perception”
16:30 Talk 13 Necmiye Ozay, University of Michigan, “Learning-in-the-loop correct-by-construction control”
17:00 Panel Discussion and closing remarks
17:30 End of the workshop and reception/refreshments

Contributed papers Session 1

  1. A. La Rocca, G. Lunardi, M. Saveriano, and A. Del Prete, “Safe Nonlinear Model Predictive Control using a Learned Approximate Control-Invariant Set”, pdf
  2. O. Guerrero Rosado, I. T. Freire, A. F. Amil, and P. Verschure, “Enabling robot autonomy through biomimetic self-regulatory dynamics”, pdf
  3. D. de Tinguy, P. Mazzaglia, B. Dhoedt, T. Verbelen, S. Remmery, “Learning to Navigate from Scratch using World Models and Curiosity: the Good, the Bad, and the Ugly”, pdf
  4. G. Cioffi, L. Bauersfeld, and D. Scaramuzza, “HDVIO: Improving Localization and Disturbance Estimation with Hybrid Dynamics VIO”, pdf
  5. R. Banerjee, P. Ray, M. Campbell, “Improving Environment Robustness of Deep Reinforcement Learning Approaches for Autonomous Racing Using Bayesian Optimization-based Curriculum Learning”, pdf

Contributed papers Session 2

  1. Y. Song and D. Scaramuzza, “Reinforcement Learning for Agile Flight: From Perception to Action”, pdf
  2. K. Singhal, M. Yaghouti, P. Jamshidi, “Multi-Sense-Rescuer: Multi-Target Audio-Visual Learning and Navigation in Search and Rescue Scenarios”, pdf
  3. J. Xiao, D. Toertei, E. Roura, and G. Loianno, “Long-range UAV Thermal Geo-localization with Satellite Imagery”, pdf
  4. N. Simon and A. Majumdar, “MonoNav: MAV Navigation via Monocular Depth Estimation and Reconstruction”, pdf
  5. C. Pan, A. Datar, M. Nazeri, Mohammad, and X. Xiao, “Toward Wheeled Mobility on Vertically Challenging Terrain: Platforms, Datasets, and Algorithms”, pdf
  6. N. Dashora, S. Jung, D. Shah, V. Ibars, O. Lerner, C. Jung,  R. Thakker, N. Rhinehart, A. Agha-mohammadi, “Imitative Models for Passenger-Scale Autonomous Off-Road Driving”, pdf

Overview

Autonomous aerial and ground robots have the potential to assist humans in complex, time-sensitive, and dangerous tasks such as search and rescue or monitoring in indoor and outdoor environments. This requires single and multiple robots to autonomously navigate in a coordinated, agile, and collaborative manner in uncertain, dynamic, cluttered, and extreme environments. The focus of this workshop is to study and analyze the role and benefits of data-driven techniques in the sense and act problem for achieving robot super autonomy. Robot super autonomy refers to unmanned, agile, resilient, and collaborative machines that can make decisions without human intervention, and can outperform current autonomous vehicles in uncertain, complex, dynamic, extreme, and cluttered environments. Learning-based solutions are becoming essential in complementing or replacing physics-based techniques to boost online navigation performances, robots’ resilience, and algorithms’ scalability in terms of the number and types of robots. This workshop will bring together researchers, industry experts, and practitioners from the aerial, ground, space, and legged robotics communities. The audience will be heterogeneous in terms of expertise and interests, involving people from academia, industry, and government agencies, as well as practitioners focusing on the broad areas of robotics, AI, and machine learning for autonomous robots working across air, space, ground, and off-terrain domains. The workshop will be extremely useful for researchers to foster ideas in the areas of robotics, AI, and related disciplines, learn about the challenges and limitations of current approaches for robot super autonomy, and explore potential solutions. Finally, this event includes several activities tailored for early-career researchers, giving them an opportunity to learn about the latest advances in autonomous robot technology.

Organizers

Giuseppe Loianno
New York University		 Davide Scaramuzza
University of Zurich

This workshop is endorsed by the IEEE RAS TC on Aerial Robotics and Unmanned Aerial Vehicles, IEEE Technical Committee on Computer and Robot Vision, and the IEEE Technical Committee on Algorithms for Planning and Control of Robot Motion.