Home

The workshop attendance will be free of charge for everyone.
We will use the zoom platform https://nyu.zoom.us/j/91803779281
The workshop will be held on October 1, 2021 from 9 am-3:30 pm EST.

Speakers

Program October 1, 2021

 

Time Talk
9:00 Welcome
9:10 Talk 1 Raquel Urtasun, Waabi and University of Toronto, “TBD”
9:30 Talk 2 Melanie Zeillinger, ETH Zurich, “Learning-based MPC – Pushing performance under constraints”
9:50 Talk 3 Angela Schoellig, University of Toronto, “Fly Out The Window: Exploiting Discrete-Time Flatness for Fast Vision-Based Multirotor Flight”
10:10 Talk 4 Anibal Ollero, University of Seville, “Perception functionalities of fully autonomous flapping and rotary wings robots in inspection applications”
10:30 Panel Discussion 1
11:00 Break 1
11:10 Talk 6 Sertac Karaman, MIT, “Towards Autonomous Super Vehicles: Integrated Hardware and Algorithms, Simulations and Transfer”
11:30 Talk 7 Giuseppe Loianno, NYU, “Lightweight Perception, Learning and Control for Resilient Agile Flight”
11:50 Talk 8 Davide Scaramuzza, University of Zurich, “Learning to fly in the wild”
12:10 Talk 9 Aleksandra Faust, Google, “Toward Autonomous Real World Robot Navigation”
12:30 Moved at the end

Talk 10 Animashree Anandkumar, Caltech, “Uncertainty Estimation for Safety-Critical Applications”
12:50 Panel Discussion 2
13:10 Break 2
13:20 Talk 11 David S. Bayard and Andrew E. Johnson, NASA JPL, “An Overview of the Mars Helicopter Vision-Based Navigation System” and “Terrain Relative Navigation for Safe Landing of the Perseverance Mars Rover”
14:10 Talk 12 Kostas Alexis NTNU, “Team CERBERUS in the DARPA Subterranean Challenge: Our strategy, robots, capabilities, and limitations”
14:30 Talk 13 Keenan Wyrobek, Zipline, “From airspace architecture to autonomy – lessons learned developing and scaling autonomy at Zipline” 
14:50 Talk 14 Terese Manley, NIST, “First Responder UAS Triple Challenge”
15:00 Talk 10 Animashree Anandkumar, Caltech, “Uncertainty Estimation for Safety-Critical Applications”
15:20 Panel Discussion 3 and closing remarks

Motivation

During the last decade, research on autonomous vehicles has started to transform our society creating novel intelligent machines (drones, ground robots, and autonomous cars) that have the potential to change the human mobility by providing alternative transportation solutions in urban settings. Recently, remotely piloted aerial and ground vehicles/cars navigating at high speed in complex racing courses have inspired many researchers to design autonomy algorithms with the goal to create the so-called super-vehicles, i.e. autonomous vehicles with the ability to execute agile and racing maneuvers with superior performances compared to human controlled vehicles. These capabilities can empower autonomous machines to accomplish complex tasks ranging from search and rescue, to inspection within a limited amount of time and with only a limited risk for people.

Goal

This workshop will focus on how to integrate and jointly design perception, learning, and control within the sense and act paradigm of autonomous vehicles to scale navigation performances to a super level of autonomy, agility, and racing capabilities. We bring together heterogeneous communities working on aerial robots, mobile ground vehicles, racing cars, and autonomous cars to unify and jointly design perception, learning, and control approaches for autonomous vehicles. Most of the previous workshops have attempted to study the basic perception, navigation, and control in a disjoint manner. While this approach currently works, it does not consider the interplay between perception, learning, and control modules which is of fundamental importance to achieve agile navigation and super vehicle performances in extreme navigation conditions. The problem is extremely challenging especially considering the weight, power and size constraints often impose multiple severe algorithm and hardware design constraints.

Topics of interest to this workshop include, but are not necessarily limited to:

  • Visionary ideas for autonomy of ground and aerial vehicles
  • Learning for control
  • Agile autonomous navigation, transportation and manipulation with ground and aerial vehicles
  • Agile visual control and state estimation
  • SLAM
  • Sensor fusion
  • Motion planning
  • Obstacle avoidance
  • Modeling and benchmarking of performances for three-dimensional navigation
  • Cooperative estimation and control with multiple robots
  • Search and rescue robotics

Organizers

Giuseppe Loianno
New York University		 Davide Scaramuzza
University of Zurich		 Sertac Karaman
MIT

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