Synthesis for Composable Robots: Guarantees and Feedback for Complex Behaviors

Hadas Kress-Gazit, Cornell University


Getting a robot to perform a complex task, for example completing the DARPA Robotics Challenge, typically requires a team of engineers who program the robot in a time consuming and error prone process and who validate the resulting robot behavior through testing in different environments. The vision of synthesis for robotics is to bypass the manual programming and testing cycle by enabling users to provide specifications – what the robot should do – and automatically generating, from the specification, robot control that provides guarantees for the robot’s behavior.

This talk will describe the work done in the verifiable robotics research group towards realizing the synthesis vision and will focus on synthesis for composable robots – modular robots and swarms. Such robotic systems require new abstractions and synthesis techniques that address the overall system behavior in addition to the individual control of each component, i.e. module or swarm member.

Explorations using Telepresence Robots in the Wild

Susan Fussell and Elijah Webber-Han


Mobile Robotic (Tele)Presence (MRP) systems are a promising technology for distance interaction because they provide both embodiment and mobility.  In principle, MRPs have the potential to support a wide array of informal activities, such as walking across campus, attending a movie or visiting a restaurant. However, realizing this potential has been challenging, due to a host of issues including internet connectivity, audio interference, limited mobility and limited line of sight. We will describe some ongoing work looking at the benefits and challenges of using MRPs in the wild.  The goal of this work is to develop a framework for understanding MRP use in informal social settings that captures key relationships among the physical requirements of the setting, the social norms of the setting, and the challenges posed for MRP pilots and people in the local environment.  This framework will then inform the design of novel user interfaces and crowdsourcing techniques to help MRP pilots anticipate and overcome challenges of specific informal social settings Joint Work: Sue Fussell, Elijah Weber-Han, Dept. of Communication & Dept. of Info. Science at Cornell University

What We Talk About When We Talk About Design

Panel, Cornell University


Panelists: Keith Evan Green, Kirstin Hagelskjaer Petersen, Guy Hoffman, Rob Shepherd and François Guimbretière. As panelists, we will interact with each other and the audience on the topic of what design means for robotics and what robotics means for design. Panelists would also like to discuss briefly the Q-exam in design.

Sailing in Space

Bo Fu, Cornell University


Solar sail is a type of spacecraft propelled by harvesting momentum from solar radiation. Compared with spacecraft propelled by traditional chemical rockets or the more advanced electric propulsion engines, the unique feature of solar sails is that they do not use fuel for propulsion. This allows for the possibility of return-type (round-trip) missions to other heavenly bodies, which would be difficult or near impossible with conventional propulsion methods. This also makes solar sails highly promising candidates for service as interplanetary cargo ships in future space missions. Solar sail research is quite broad and multi-disciplinary. In this talk, an overview of solar sail technology including the history, the fundamentals of photon-sail interaction, and the state of the art of solar sailing is presented. One specific area solar sail research – attitude dynamics and control – is discussed in detail. Attitude control of large sails poses a challenge because most methods developed for solar sail attitude control require the controller mass to scale with the sail’s surface area. This is addressed by a newly proposed tip displacement method (TDM), where by moving the wing tips, the geometry of sail film is exploited to generate the necessary control forces and torques. The TDM method is described as it applies to a square solar sail that consists of four triangular wings. The mathematical relationship between the displacement of the wing tip and the control torque generated is fully developed under quasi-static condition and assuming the wing takes on the shape of a right cylindrical shell. Results from further investigation by relaxing previous modeling assumptions are presented. Future research directions in aerospace engineering spanning field of autonomy, sensing, controls, and modeling are discussed.

Science Fiction / Double Feature: Design Q Exam and Nonverbal Behaviors

Guy Hoffman, Cornell University


In this informal meeting of the robotics seminar, we will do good on our promise to discuss the structure of the new(ish) Design Q exam, including presentations by faculty of the expectations, war stories from students who took the Design Q, and Q&A (no pun intended). The second part of this double feature seminar is going to be a presentation and discussion on one of the classics papers at the foundation of HRI, Paul Ekman’s 1969 article “The Repertoire of Nonverbal Behavior: Categories, Origins, Usage, and Coding”, which is at the basis of decades of research on body language, and a must-know for any researcher interested in HRI systems using gestures and facial expressions. For a non-light reading:

 Autonomy, Embodiment, and Anthropomorphism: the Ethics of Robotics

Ross Knepper, Cornell University


A robot is an artificially intelligent machine that can sense, think, and act in the world. It’s physical, embodied aspect sets a robot apart from other artificially intelligent systems, and it also profoundly affects the way that people interact with robots. Although a robot is an autonomous, engineered machine, its appearance and behavior can trigger anthropomorphic impulses in people who work with it. In many ways, robots occupy a niche that is somewhere between man and machine, which can lead people to form unhealthy emotional attitudes towards them. We can develop unidirectional emotional bonds with robots, and there are indications that robots occupy a distinct moral status from humans, leading us to treat them without the same dignity afforded to a human being. Are emotional relationships with robots inevitable? How will they influence human behavior, given that robots do not reciprocate as humans would? This talk will examine issues such as cruelty to robots, sex robots, and robots used for sales, guard or military duties. This talk was previously presented in spring 2017 as part of CS 4732: Social and Ethical Issues in AI.

How do people, and how should legged robots, avoid falling down?

Andy Ruina


What actuators does a person or legged robot have available to help prevent falls? Only ones that can more the relative horizontal position of the support point and the center of mass. What are these? Ankle torques, distortions of the upper body (bending at hips, swinging arms), stepping and pushing off. Of these, by far the biggest control authority is in stepping and pushing off. And these can be well understood, and well approximated, by a point mass model. Why? Because the same things that can’t help much, namely ankle torques and upper body distortions, can’t hurt much either. Thus, we believe we can design a robust balance controller using foot placement and pushoff and nothing else. And, reverse engineering, we think this explains most of what people do also, at least when recovering from large disturbances. A balanced broomstick, a Segway, a bicycle, a walking robot and a walking person all use the same basic idea.

Multi-Robot Mini Symposium

Kirstin H. Petersen, Cornell University


This Multi-Robot Mini Symposium will feature a series of brief talks by students and professors related to recent work on Multi-Robot/Swarm Robotics research. The goal is to identify and inspire new ideas among the multi-robot community at Cornell. We are looking for speakers – please notify Kirstin Petersen (khp37) if you would like to do a pitch!

Robotics Debate/Discussion

3/21/18 This week we will host a debate/discussion on some topics in robotics. There is still time to contribute discussion questions here:

Anything is fair game. The topics will be announced Wednesday morning. Good questions have an opportunity for deep discussion, support a variety of viewpoints, and engage the broad robotics community. You may sign your name or leave your question anonymous. If you put your name, you are volunteering to give a few sentence explanation of the question and its implications. -Ross

Maidbot: Designing and Building Rosie the Robot for the Hospitality Industry

Steve Supron, Maidbot


Steve Supron joined Maidbot as Manufacturing Lead during its incubation days over two years ago at REV Ithaca. Micah Green, a former Cornellian and the founder and CEO of Maidbot, hired Steve to help bring his dream of Rosie the Robot to the hotel industry. Steve will present the company’s story as well as the challenges and considerations of robotics in a hospitality setting. Steve will review some of the unique design decisions and technology and production choices the team has made along the way from early prototypes to testable pilot units and on to the production design.