Human-Robot Collaboration and Companionship LabProject "Blossom" is a robot that is soft inside and out. It is built using traditional crafts like wool and wood. We wanted to bring back warm materials to home robotics, instead of more plastics, glass, and metal.
You can customize Blossom by knitting new exteriors and attaching different crafted parts to make each robot unique.
Blossom is currently used in research at Cornell University. We study new mechanical designs for social robots, including how to use soft materials and soft mechanisms in social robot design.
This robot is also used as a platform to study how Machine Learning can be used to make a robot react to YouTube content. One of the aims is to help children on the Autism spectrum to develop social skills. Blossom is built using TensorFlow, Google's open-source Machine Learning framework.
"Blossom" is a collaboration between Cornell University's Human-Robot Collaboration and Companionship Lab and Google Creative Technologies Singapore.
Blossom - A Handcrafted Social Robot - Soft Inside and OutHuman-Robot Collaboration and Companionship Lab2017-08-22 | Project "Blossom" is a robot that is soft inside and out. It is built using traditional crafts like wool and wood. We wanted to bring back warm materials to home robotics, instead of more plastics, glass, and metal.
You can customize Blossom by knitting new exteriors and attaching different crafted parts to make each robot unique.
Blossom is currently used in research at Cornell University. We study new mechanical designs for social robots, including how to use soft materials and soft mechanisms in social robot design.
This robot is also used as a platform to study how Machine Learning can be used to make a robot react to YouTube content. One of the aims is to help children on the Autism spectrum to develop social skills. Blossom is built using TensorFlow, Google's open-source Machine Learning framework.
"Blossom" is a collaboration between Cornell University's Human-Robot Collaboration and Companionship Lab and Google Creative Technologies Singapore.Mohamed Bouri - Supernumerary Robotic Manipulation for Laparoscopic Surgery. SRD@AIM-20Human-Robot Collaboration and Companionship Lab2020-07-29 | Supernumerary Robotic Manipulation for Laparoscopic surgery: Envisioned Scenarios and Results
Presented by Prof. Mohamed Bouri, École Polytechnique Fédérale de Lausanne (EPFL)
Invited Lecture at the 1st Workshop on Supernumerary Robotic Devices held as part of the 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020).
Abstract: Manipulation tasks are often bimanual, regardless that in several cases humans require to be assisted to successfully achieve their tasks, could be for assembly or positioning ; could be for precise applications or when manipulating heavy objects. To this end, more attention and coordination is required between the two persons, or more, involved in the same manipulation task. In our project we envision to give the control to only one operator, who will finally carry out the manipulation using his two biological hands, while being in control of two robotic arms- This supernumerary 4-handed manipulation has the main objective to increase the success of the operation by decreasing communication errors between the operator and his assistant, by increasing the manipulation area and by easing the execution of the tasks in giving the full control to only one operator. In [Abdi et al, 2016], we already proved that even in demanding tasks, three-handed manipulation is preferred to two-handed manipulation. This presentation will address the application of laparoscopic surgery [Abdi et al., 2015] [Hernandez et al., 2019]. This work is carried out in collaboration between the research group REHAssist and the lab LASA, both at Ecole Polytechnique Fédérale de Lausanne, Switzerland. It is a follow up of our investigation to the capabilities of surgeons to foot-control a laparoscopic instrument [Abdi et al., 2017]. A bipedal foot interface has been developed to achieved the control of two robotic arms, 5 degrees of freedom each. This foot-haptic interface [Hernandez et al., 2019] is capable to provide 5 DOFs foot force feedback to improve foot control perception. On the other side, the robots are implemented with assistance control strategies [Amanhoud et al., 2020]. The presented results will point out that force assistance improves human-robot interaction in four-handed manipulation, as well as reduces fatigue, improves ease of use and usefulness.Walterio Mayol-Cuevas - Handheld Robots: Bridging the Gap between Fully External and Wearable RobotsHuman-Robot Collaboration and Companionship Lab2020-07-29 | Handheld Robots: Bridging the Gap between Fully External and Wearable Robots
Presented by Prof. Walterio Mayol-Cuevas, University of Bristol
Invited Lecture at the 1st Workshop on Supernumerary Robotic Devices held as part of the 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020).
Abstract: In this talk, we will discuss our past and recent work on the development of handheld robots. A Handheld robot is a person-oriented robot that shares properties of a handheld tool while being enhanced with autonomous motion as well as the ability to process task-relevant information and user signals. The application possibilities include helping inexperienced users to perform power tool-type tasks without much task knowledge and with limited training on the tool usage. These robots exploit the Moravec paradox by combining the strengths of human users such as innate obstacle avoidance and navigation skills, with precise motion and memory enhancement by robotic devices. In our recent work we have explored issues of the way to predict user intention from minimal user input such as gaze detection and motion, and issues of conflict of interests between user and robot. We have built prototypes of handheld robots and conducted various pilot studies on the effects of intention prediction, robot rebellion and tele-operation on simulated copy-block and maintenance tasks. We believe handheld robots bridge the gap that currently exists between fully independent robots for which full autonomy is the goal, and wearable or supernumerary robots where the robot is tightly coupled with the user. With handheld robots, we hope to tap into the millions of years that humans have used handheld tools but now with the enhancement possibilities that Robotics can offer. http://handheldrobotics.orgMonica Malvezzi and Domenico Pratichizzo - The Robotic Sixth Finger. SRD@AIM-20Human-Robot Collaboration and Companionship Lab2020-07-29 | The Robotic Sixth Finger
Presented by Prof. Monica Malvezzi and Prof. Domenico Prattichizzo, University of Siena and Istituto Italiano di Tecnologia (IIT)
Invited Lecture at the 1st Workshop on Supernumerary Robotic Devices held as part of the 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020).
Abstract: This talk will summarise the main phases of the research that we developed in the framework of human augmentation with supernumerary extra limbs. Starting from our previous experience on robotic hands and grasping, the device that we developed was aimed at integrating the human hand by means of a robotic extra finger. This topic opened a wide set of research questions, both theoretical/methodological and design/technological. The first tests and prototypes were realised with a modular fully actuated extra finger, in which each phalanx was controlled with one actuator, in a serial chain. This first solution allowed us to develop strategies for mapping human hand motion to an augmented bio-artificial hand. A fully actuated solution was not suitable for inexperienced users, due to the complexity and lack of robustness. For this reason we moved to underactuated compliant solution. The solution that we developed was actuated by only one motor, placed on the bracelet supporting the device. Finger closure motion is given by a tendon connecting all the phalanges. Finger phalanges are furthermore connected by deformable joints, with an equivalent compliance defined in the design phase to obtain the desired closure motion and realised by properly controlling manufacturing parameters and material mechanical properties. In parallel, user interfaces were developed for device control. Also for the user interfaces, wearability, robustness and easy of use were the criteria that guided the design. The most impactful application that we identified for this device was for subjects suffering of upper limb pathologies or diseases, as for instance the stroke. Some preliminary tests conducted with some stroke patients demonstrated that this type of device represents a very useful tool for recovering a part of grasping capabilities and for supporting the patient in several bimanual tasks that are very frequent in everyday life. A video showing the potential of the sixth finger for stroke patients is available at youtube.com/watch?v=zbybITNk0N8Masahiko Inami - Virtual Cyborgs: Freedom from Body Limitations. SRD@AIM-20Human-Robot Collaboration and Companionship Lab2020-07-28 | Virtual Cyborgs: Freedom from Body Limitations
Presented by Prof. Masahiko Inami, University of Tokyo
Invited Lecture at the 1st Workshop on Supernumerary Robotic Devices held as part of the 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020).
Abstract: The social revolutions have accompanied innovation of the view of the body. If we regard the information revolution as establishment of a virtual society against the real society, it is necessary to design a new view of body "JIZAI body Virtual Cyborg)", which can adapt freely to the change of social structure, and establish a new view of the body. In this talk, we discuss how we understand of basic knowledge about the body editing for construction of JIZAI body (Virtual Cyborg) based on VR, AR and Robotics. Superhuman Sports: Applying Human Augmentation to Physical Exercise. This talk will also present Superhuman Sports, a form of "Human-Computer Integration” to overcome somatic and spatial limitation of humanity by merging technology with the body. We hope to create a future of sports where everyone, strong or weak, young or old, non-disabled or disabled, can play and enjoy playing without being disadvantaged.Aldo Faisal - Playing Piano w/ 11 Fingers: Neurobehavioural Constraints of Human Robot AugmentationHuman-Robot Collaboration and Companionship Lab2020-07-28 | Playing the Piano with 11 Fingers: The Neurobehavioural Constraints of Human Robot Augmentation
Presented by Prof. Aldo Faisal, Imperial College London
Invited Lecture at the 1st Workshop on Supernumerary Robotic Devices held as part of the 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020).
Presented by Prof. Hiroyasu Iwata, Waseda University
Invited Lecture at the 1st Workshop on Supernumerary Robotic Devices held as part of the 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020).
Abstract: This talk covers the development of a voluntarily manipulative wearable robot arm called "3rd Arm", which allows humans to comprehensively perform multiple tasks. Especially, I would like to address the design and evaluation of an intuitive interface to operate "3rd Arm" focusing on face vector and a new concept of SR device called "Detachable body" allowing us to perform a concurrent tasks at two different places simultaneously. We've focused on head movement for manipulating an extra arm and propose a new interface based on a“face vector.”It's a point instruction-type interface, unlike conventional path control-type interfaces, which reduces the cognitive load of the conventional head movement interface. We gave a cocrete body as an eyeglass-shape interface device mounting the gyro sensor and distance sensor for getting coordinate. Also, a laser pointer is attached to the interface to give visual biofeedback to the users, since Because a face vector itself is invisible. The result of experiments under two dual-task conditions shows that the cognitive load of the proposed interface is low enough, even when a high cognitive load task is included in dual-task conditions.It demonstrates that the point instruction interface proposed in this study can shorten working time for both direct and indirect reaching tasks compared to the path control interface in all single task conditions. Detachable body is a new concept of a wearable robot arm as an extended body. It can be detached from user's natural body and attached not only to the other person's body but also anywhere even in the environment. Humans can become to perform dual-presence tasks that allow concurrent tasks executed in two distant places using the detachable body. Also, we designed an information presenting interface to handle both the natural body and the detached body at two distant places simultaneously. The interface consists of a vision presenting system that superimposed two environment images with binocular disparity, and a proprioception presenting system that makes somatosensory feedback of detached arm's position. The usability of the proposed interface was evaluated by the work efficiency and subjective evaluation in a dual-presence task. The result suggested the effects of disparity in the vision presenting system and feedback information.AUR Fluency ExperimentHuman-Robot Collaboration and Companionship Lab2018-11-12 | Illustration of the Fluency experiment described in
An autonomous robot using an embodied cognition architecture working together with a human on a mock painting task.Goosebumps - Texture-Changing Soft Robotic SkinHuman-Robot Collaboration and Companionship Lab2018-05-01 | We developed a soft robotic skin that can change its texture to express its internal state, such as emotions. The prototype skin can change its skin through a combination of goosebumps and spikes.
This is in contrast to most socially expressive robots, which use either gestures or facial expressions to communicate. While these are common human expression channels, humans and many animals also show emotions on their skin, like ruffled feathers and goosebumps.
This research project aims to give social robots a similar capability.
Kip1, is an Empathy Object in the form of a peripheral robotic conversation companion. The robot’s function is to promote non-aggressive conversation between people. It monitors the conversation around it and reacts emotionally to aggressive speech.
Kip1, is an Empathy Object in the form of a peripheral robotic conversation companion. The robot’s function is to promote non-aggressive conversation between people. It monitors the conversation around it and reacts emotionally to aggressive speech.
Travis is fully controlled by an Android phone through Google's ADK, and can therefore be expanded by custom mobile apps.
In this demo, it functions as an intelligent speaker dock, which can select songs from a library based on a rhythmic input, analyze the song's beat and genre and produce beat-matched and genre-specific movements.
Credits:
PIs: Guy Hoffman / IDC Herzliya - http://guyhoffman.com Gil Weinberg / Georgia Tech - http://gtcmt.gatech.edu/
Robot Design: Guy Hoffman Rob Aimi / Alium Labs
Music Intelligence: Gil Weinberg Jimmy ONeill Mason Bretan
Motion Design & Engineering Guy Hoffman Orr Gottlieb Assaf Mashiah
Engineering Design Support: Ian CampbellVyo - Social Robot for the Smart HomeHuman-Robot Collaboration and Companionship Lab2017-08-17 | Vyo - a social robot designed to manage your smart home.
A collaboration between the IDC Media Innovation Lab (milab), the Cornell Human-Robot Collaboration & Companionship Lab (HRC^2), and SK Telecom
Credits: Guy Hoffman, Oren Zuckerman, Sung Park Michal Luria, Benny Megidish Rob Aimi Maayan Polak, Leor Alon, Paul Osman Videography: Dorin Haveri Music: Nicolai Heidlas - Take The Chance