WAFR 2022

The 15th International Workshop on the Algorithmic Foundations of Robotics will be held on June 22-24, 2022 at the University of Maryland, College Park

Registration Live Streaming

Program at a Glance

Wednesday June 22 Thursday June 23 Friday June 24
08:10 Opening
08:30 Session 1 Session 4 Session 7
09:45 Keynote: Russ Tedrake Keynote: Hadas Kress-Gazit Keynote: Oliver Brock
10:45 Coffee Break Coffee Break Coffee Break
11:15 Session 2 Session 5 Lab tours; NSF Q&A
12:30 Lunch Lunch Lunch
14:00 Open Problems Session 6 Session 8
15:00 Coffee Break Coffee Break Coffee Break
15:30 Session 3 Remembering Jean-Paul Laumond: Steve LaValle Session 9
Opening Reception (18:00) Banquet (18:30) Closing and Award Ceremony (16:45)


Edward St. John Learning and Teaching Center

Edward St. John Learning and Teaching Center

Brendan Iribe Center for Computer Science and Engineering

Brendan Iribe Center for Computer Science and Engineering

The Smithsonian Castle

The Smithsonian Castle

Local Map

Local Map

The technical sessions will be held in Room 1224 of the Edward St. John Learning and Teaching Center at the University of Maryland, College Park. The ESJ building a short ~8 minute walk from The Hotel.

The opening reception will be held in the lobby of the Brendan Iribe Center for Computer Science and Engineering. The IRB building is right across the street from The Hotel.

The conference banquet will be held at the Smithsonian Institution Building, popularly known as the "Castle" in Washington DC. Completed in 1855, the Castle today sits in the middle of the national mall that also feature the U.S. Capitol, the Washington Monument, the Lincoln Memorial, and the White House. Shuttles will be provided from the ESJ building (boarding begins at 5PM; departing at 5:15PM) and The Hotel (boarding begins at 5PM; departing at 5:30PM) to reach the Castle. The shuttles will also provide transportation back to The Hotel after the banquet. Participants can also choose to take an Uber or ride on the green/yellow line of the metro to travel to the national mall.

Covid-19 Policies

All in-person atttendees will have to adhere to the University of Maryland, College Park Covid-19 policies for visitors. In particular, this means that all in-person attendees will have to wear a KN95 mask when inside the conference room where the technical sessions are taking place and when on the University shuttle. Masks will be provided at the registration desk. This policy is subject to change based on the county, state, and University of Maryland guidelines.

Detailed Program

Session 1 (Wednesday June 22)
Session Chair: Srinivas Akella
8:30-8:45 Sihui Li and Neil Dantam. Exponential Convergence of Infeasibility Proofs For Kinematic Motion Planning. [supplementary material]
8:45-9:00 Alexander Botros, Nils Wilde, Armin Sadeghi, Javier Alonso-Mora and Stephen Smith. Error-Bounded Approximation of Pareto Fronts in Robot Planning Problems. [supplementary material]
9:00-9:15 Brian Axelrod and Luke Shimanuki. Efficient Motion Planning under Obstacle Uncertainty with Local Dependencies. [supplementary material]
9:15-9:30 Michael Farber and Shmuel Weinberger. Parametrized motion planning and topological complexity.
9:30-9:45 Q&A for Session 1
Keynote 1 (Wednesday June 22, 9:45am)

Motion planning around obstacles with convex optimization

Russ Tedrake

Abstract: In this talk, I'll describe a new approach to planning that strongly leverages both continuous and discrete/combinatorial optimization. The framework is fairly general, but I will focus on a particular application of the framework to planning continuous curves around obstacles. Traditionally, these sort of motion planning problems have either been solved by trajectory optimization approaches, which suffer with local minima in the presence of obstacles, or by sampling-based motion planning algorithms, which can struggle with derivative constraints and in very high dimensions. In the proposed framework, called Graph of Convex Sets (GCS), we can recast the trajectory optimization problem over a parametric class of continuous curves into a problem combining convex optimization formulations for graph search and for motion planning. The result is a non-convex optimization problem whose convex relaxation is very tight — to the point that we can very often solve very complex motion planning problems to global optimality using the convex relaxation plus a cheap rounding strategy. I will describe numerical experiments of GCS applied to a quadrotor flying through buildings and robotic arms moving through confined spaces. On a seven-degree-of-freedom manipulator, GCS can outperform widely-used sampling-based planners by finding higher-quality trajectories in less time, and in 14 dimensions (or more) it can solve problems to global optimality which are hard to approach with sampling-based techniques.

Bio: Russ Tedrake is the Toyota Professor at the Massachusetts Institute of Technology (MIT) in the Department of Electrical Engineering and Computer Science, Mechanical Engineering, and Aero/Astro, and he is a member of MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). He is also the Vice President of Robotics Research at Toyota Research Institute (TRI). He received a B.S.E. in Computer Engineering from the University of Michigan in 1999, and a Ph.D. in Electrical Engineering and Computer Science from MIT in 2004. Dr. Tedrake is the Director of the MIT CSAIL Center for Robotics and was the leader of MIT's entry in the DARPA Robotics Challenge. He is a recipient of the NSF CAREER Award, the MIT Jerome Saltzer Award for undergraduate teaching, the DARPA Young Faculty Award in Mathematics, the 2012 Ruth and Joel Spira Teaching Award, and was named a Microsoft Research New Faculty Fellow. His research has been recognized with numerous conference best paper awards, including ICRA, Robotics: Science and Systems, Humanoids, Hybrid Systems: Computation and Control, as well as the inaugural best paper award from the IEEE RAS Technical Committee on Whole-Body Control.

Session 2 (Wednesday June 22)
Session Chair: Lydia Tapia
11:15-11:30 Cong Wei and Derek A. Paley. Distributed spacing control for multiple, buoyancy-controlled underwater robots.
11:30-11:45 Mitchell Jones, Max Heger and Jur van den Berg. Lane-Level Route Planning for Autonomous Vehicles.
11:45-12:00 Saurav Agarwal and Srinivas Akella. The Correlated Arc Orienteering Problem.
12:00-12:15 Zhaoming Xie, Xingye Da, Buck Babich, Animesh Garg and Michiel van de Panne. GLiDE: Generalizable Quadrupedal Locomotion in Diverse Environments with a Centroidal Model. [supplementary material]
12:15-12:30 Q&A for Session 2
Open Problems (Wednesday June 22, 14:00pm)

An important and longstanding component of the WAFR program is the open problem session, in which members of the WAFR community share open problems that they find impactful (important, interesting, useful, beautiful, etc.).

The WAFR 2022 organizing committee is soliciting open problem suggestions. We will pick three or four problems to become part of this year's open problem session.

If your open problem suggestion is selected for the session, then you will be expected to prepare a short three minute overview talk (chalk talk and/or presentation) motivating/describing the open problem. After your introduction WAFR participants will discuss the problem and its potential solutions.

Open problems discussed in years past have been solved by the community and papers describing the solution have even been submitted to (and presented at) subsequent WAFR conferences.

Please use this link to submit ideas for an open problem that you would like to discuss at WAFR!

Session 3 (Wednesday June 22)
Session Chair: Jaime Fisac
15:30-15:45 Hai Nguyen, Zhihan Yang, Andrea Baisero, Xiao Ma, Robert Platt and Christopher Amato. Hierarchical Reinforcement Learning under Mixed Observability. [supplementary material]
15:45-16:00 Yazied Hasan, John Baxter, Cesar A. Salcedo, Elena Delgado and Lydia Tapia. Reinforcement Learning of Coordinated Shepherds for Flock Navigation and Confinement. [supplementary material]
16:00-16:15 Wenhao Luo, Wen Sun and Ashish Kapoor. Sample-efficient Safe Learning for Online Nonlinear Control with Control Barrier Functions. [supplementary material]
16:15-16:30 Q&A for Session 3
Session 4 (Thursday June 23)
Session Chair: Dylan Shell
8:30-8:45 Jiaheng Hu, Howard Coffin, Julian Whitman, Matthew Travers and Howie Choset. Large-scale Heterogeneous Multi-Robot Coverage via Domain Decomposition and Generative Allocation.
8:45-9:00 Christoforos Mavrogiannis, Jonathan DeCastro and Siddhartha Srinivasa. Implicit Multiagent Coordination at Uncontrolled Intersections via Topological Braids.
9:00-9:15 Mitali Gandhe and Michael Otte. Decentralized Robot Swarm Clustering: Adding Resilience to Malicious Masquerade Attacks. [supplementary material]
9:15-9:30 Aviv Adler, Oscar Mickelin, Ragesh Ramachandran, Gaurav Sukhatme and Sertac Karaman. The Role of Heterogeneity in Autonomous Perimeter Defense Problems.
9:30-9:45 Q&A for Session 4
Keynote 2 (Thursday June 23, 9:45am)

The Missing Skill

Hadas Kress-Gazit

Abstract: To enable robots to perform complex tasks — interact with people, react to events, change their task — we must focus on specifications (what the robot should do), skills (what it can do), and composition (how the robot will do what it should do, given what it can do). There is extensive work on all three of these themes; in this talk I will focus on finding the “missing skill(s)” — given a task, if the robot’s set of skills is not enough to achieve it, how can we automatically identify and create additional skills that will enable the robot to achieve the task.

Bio: Hadas Kress-Gazit is the Geoffrey S.M. Hedrick Sr. Professor at the Sibley School of Mechanical and Aerospace Engineering at Cornell University. She received her Ph.D. in Electrical and Systems Engineering from the University of Pennsylvania in 2008 and has been at Cornell since 2009. Her research focuses on formal methods for robotics and automation and more specifically on synthesis for robotics — automatically creating verifiable robot controllers for complex high-level tasks. Her group explores different types of robotic systems including modular robots, soft robots and swarms and synthesizes (pun intended) ideas from different communities such as robotics, formal methods, control, hybrid systems and computational linguistics. She is an IEEE Fellow and has received multiple awards for her research, teaching and advocacy for groups traditionally underrepresented in STEM. She lives in Ithaca with her partner and two kids.

Session 5 (Thursday June 23)
Session Chair: Stephen L. Smith
11:15-11:30 Aakriti Upadhyay, Boris Goldfarb, Weifu Wang and Chinwe Ekenna. A New Application of Discrete Morse Theory to Optimizing Safe Motion Planning Paths.
11:30-11:45 Rachel Luo, Shengjia Zhao, Jonathan Kuck, Boris Ivanovic, Silvio Savarese, Edward Schmerling and Marco Pavone. Sample-Efficient Safety Assurances using Conformal Prediction. [supplementary material]
11:45-12:00 Gustavo Cardona and Cristian-Ioan Vasile. Partial Satisfaction of Signal Temporal Logic Specifications for Coordination of Multi-Robot Systems.
12:00-12:15 Glen Chou, Necmiye Ozay and Dmitry Berenson. Safe Output Feedback Motion Planning from Images via Learned Perception Modules and Contraction Theory. [supplementary material]
12:15-12:30 Q&A for Session 5
Session 6 (Thursday June 23)
Session Chair: Dmitry Berenson
14:00-14:15 Deepak Edakkattil Gopinath, Andrew Thompson and Brenna Argall. Information Theoretic Intent Disambiguation via Contextual Nudges for Assistive Shared Control. [supplementary material]
14:15-14:30 Haimin Hu and Jaime Fisac. Active Uncertainty Learning for Human-Robot Interaction: An Implicit Dual Control Approach. [supplementary material]
14:30-14:45 Ewerton Vieira, Edgar Granados, Aravind Sivaramakrishnan, Marcio Gameiro, Konstantin Mischaikow and Kostas E. Bekris. Morse Graphs: Topological Tools for Analyzing the Global Dynamics of Robot Controllers. [supplementary material]
14:45-15:00 Q&A for Session 6
Thursday June 23, 15:30pm

Remembering Jean-Paul Laumond: Steve LaValle.

Session 7 (Friday June 24)
Session Chair: Ken Goldberg
8:30-8:45 Yulin Zhang and Dylan Shell. Nondeterminism subject to output commitment in combinatorial filters. [supplementary material]
8:45-9:00 Basak Sakcak, Vadim Weinstein and Steven M. LaValle. The Limits of Learning and Planning: Minimal Sufficient Information Transition Systems. [supplementary material]
9:00-9:15 Michelle Ho, Alec Farid and Anirudha Majumdar. Towards a Framework for Comparing the Complexity of Robotic Tasks.
9:30-9:45 Q&A for Session 7
Keynote 3 (Friday June 24, 9:45am)

Water, Earth, Fire, and Air: What Basic Elements Form the Foundation of Robotics?

Oliver Brock

Abstract: What once was the foundation of physics, today seems quaint. Over the centuries, most scientific disciplines have experienced disruptive changes to their foundation. These changes are, in fact, a characteristic of mature scientific disciplines. Looking at robotics: What are the chances that what we consider the foundation today, will remain forever? If the history of science is an indication, the chances are close to zero. One might argue that the foundation of robotics is changing already through the advent of new methods and technologies, most recently, for example, through deep learning or depth cameras. But these methods and technologies are applied within a conceptualization of robotics that remains unchanged. One might also argue that robotics is not starting from scratch but is building on centuries of advances in other disciplines. But the foundations of these disciplines are tailored to the problems addressed in these disciplines. We have imported them and structured robotics accordingly. If robotics should evolve as a scientific discipline, it might be worth examining the implicit assumptions we have imported along with the foundations of other disciplines — and the resulting consequences. In this talk, I will attempt to evaluate the current foundation of robotics and speculate about possible changes and what future advances they might enable.

Bio: Oliver Brock is the Alexander-von-Humboldt Professor of Robotics in the School of Electrical Engineering and Computer Science at the Technische Universität Berlin, a German "University of Excellence". He received his Ph.D. from Stanford University in 2000 and held postdoctoral positions at Rice University and Stanford University. He was an Assistant and Associate Professor in the Department of Computer Science at the University of Massachusetts Amherst before moving back to Berlin in 2009. The research of Brock's lab, the Robotics and Biology Laboratory, focuses on robot intelligence, mobile manipulation, interactive perception, grasping, manipulation, soft material robotics, interactive machine learning, deep learning, motion generation, and the application of algorithms and concepts from robotics to computational problems in structural molecular biology. Oliver Brock directs the Research Center of Excellence "Science of Intelligence". He is an IEEE Fellow and was president of the Robotics: Science and Systems Foundation from 2012 until 2019.

Q&A on NSF Robotics Programs (Friday June 24, 11:15am)

This session will include presentations on NSF funding opportunities of interest to the robotics community. The cross-Directorate robotics core program Foundational Research in Robotics (FRR) was founded in 2020 to eliminate artificial distinctions between the engineering and computer science elements of robotics, and to support a broad base of community-driven innovation in robotics. NSF program officers will present the goals of the FRR program and provide an overview of other robotics-relevant NSF programs.

It was recently announced that the National Robotics Initiative (NRI) — which has been NSF's flagship robotics program for over a decade — will be sunset. NSF program officers will also discuss the implications of the NRI sunset on support for robotics at NSF and offer guidance for prospective researchers. NSF remains committed to supporting and growing a thriving robotics research community. The FRR program will now provide a single home for foundational research in robotics across NSF. FRR welcomes proposals on a broad spectrum of foundational research in robotics, including the topics of collaborative robotics and integration in robotics that were previously supported by NRI.

Following the presentations, NSF program officers will take questions from the audience. More information about the robotics programs at NSF can be found at the Robotics@NSF website.

Lab Tours (Friday June 24, 11:15am)

Join in a tour of the Maryland Robotics Center labs including live demos of indoor and outdoor navigation with ground robots, legged robots, and aerial robots.

The lab tours will happen concurrently with the Q&A session with NSF program managers. Participants can choose to attend either of the sessions.

Session 8 (Friday June 24)
Session Chair: Kostas Bekris
14:00-14:15 Yahav Avigal, Jeffrey Ichnowski, Max Cao and Ken Goldberg. GOMP-ST: Grasp Optimized Motion Planning for Suction Transport.
14:15-14:30 Khen Elimelech, Lydia E. Kavraki and Moshe Y. Vardi. Automatic cross-domain task plan transfer by caching abstract skills. [supplementary material]
14:30-14:45 Alexandre Amice, Hongkai Dai, Peter Werner, Annan Zhang and Russ Tedrake. Finding and Optimizing Certified, Collision-Free Regions in Configuration Space for Robot Manipulators. [supplementary material]
14:45-15:00 Q&A for Session 8
Session 9 (Friday June 24)
Session Chair: Dan Halperin
15:30-15:45 Tristan Schäfer, Jan Bessai, Constantin Chaumet, Jakob Rehof and Christian Riest. Design Space Exploration for Sampling-Based Motion Planning Programs with Combinatory Logic Synthesis. [supplementary material]
15:45-16:00 Mohamed Khalid M Jaffar and Michael Otte. PiP-X: Funnel-based Online Feedback Motion Planning/Replanning in Dynamic Environments. [supplementary material]
16:00-16:15 Marcus Hoerger, Hanna Kurniawati, Dirk Kroese and Nan Ye. Adaptive Discretization using Voronoi Trees for Continuous-Action POMDPs. [supplementary material]
16:15-16:30 Zhongqiang Ren, Sivakumar Rathinam and Howie Choset. A Lower Bounding Framework for Motion Planning amid Dynamic Obstacles in 2D. [supplementary material]
16:30-16:45 Q&A for Session 9