Orateurs invités

Chair in Optimization and Biomechanics for Human Centred Robotics
Director of KIT BioRobotics Lab
Institute for Anthropomatics and Robotics (IAR)
Karlsruhe Institute of Technology (KIT), Germany
Canada Excellence Research Chair in Human Centred Robotics and Machine Intelligence, University of Waterloo, Canada

Endowing humanoid robots with motion intelligence: the roles of bio-inspiration, optimization and learning

In the past years, the field of humanoid robots has seen an amazing progress with many new prototypes being developed in industry as well as in academic labs worldwide. The vision of humanoids truly being able to provide support to humans in a wide range of dirty, dull and dangerous jobs and to becoming a companion or carer of humans gets closer to being fulfilled. However, generating and controlling motions for whole-body - i.e. bipedal – humanoid robots with their nonlinear dynamics with many degrees of freedom, underactuation and inherent instability is still very challenging, and in general humanoids do not yet achieve human motor skill levels.

In this talk, I will give an overview of our research on endowing humanoid robots with motion intelligence or embodied artificial intelligence that makes the robot aware of how it moves in and interacts with its dynamic environment and with humans. As they need to either physically interact with humans or replace human actions, they must be able to predict human behavior, recognize human intent, know how to safely interact with them and how to best support them. In addition, they should convey their motion intent to humans by moving in a human-like way and thus making it easier for humans to interact with the robots.

Our research aims to discover and describe general principles of human movement such as stability and efficiency or optimization principles underlying certain behaviors. We aim not only for qualitative understanding but also for quantitative knowledge about human movement that can be formulated in terms of mathematical models. These can serve as bioinspiration for humanoid motions, but also support understanding of human intent. Building on top of this understanding, we develop efficient computational methods combining model-based methods such as optimization with model-free approaches, that allow to control and improve motions for humanoids adapted to the specific situation.

I will show a range of different motions , which we generated for our Reem-C humoind robot at my lab in Waterloo and which all require an effective coordination of the whole body and control of stability. Motions covered include walking on different terrains, balancing, bimanual manipulation of objects, riding personal transporters and skateboards, natural body language during multimodal communication, and finally close-proximity physical-social human-robot interactions which are important in many applications ranging from dancing to healthcare.

Biography

Katja Mombaur joined the Karlsruhe Institute of Technology in Germany in 2023 as Full Professor, Chair for Optimization & Biomechanics for Human-Centred Robotics and Director of the BioRobotics Lab. In addition, she holds an affiliation with the University Waterloo in Canada where she has been Full Professor and Canada Excellence Research Chair (CERC) for Human-Centred Robotics & Machine Intelligence since 2020. Prior to moving to Canada, she has been a Full Professor at Heidelberg University where she directed the Optimization, Robotics & Biomechanics Chair, as well as the Heidelberg Center for Motion Research. Her international experience includes two years as a visiting researcher at LAAS-CNRS in Toulouse and one year at Seoul National University. She studied Aerospace Engineering at the University of Stuttgart and SupAéro in Toulouse and holds a PhD in Mathematics from Heidelberg University.
Katja Mombaur currently serves as the Vice President for Member Activities of the IEEE Robotics & Automation Society and co-chair of the TC on Humanoid Robots. She is a Senior Editor of the IEEE Transactions on Robotics and has actively contributed to the organization of many conferences and workshops, in various roles e.g. as Program Chair of the IEEE RAS Humanoids 2023 conference.
Katja’s research focuses on understanding human movement by a combined approach of model-based optimization and experiments and using this knowledge to improve motions of humanoid robots and the interactions of humans with exoskeletons, prostheses and external physical devices. Her goal is to endow humanoid and wearable robots with motion intelligence that allow them to operate safely in a complex human world. The development of efficient algorithms for motion generation, control and learning is a core component of her research.

Dr. Daniele Pucci

Istituto Italiano di Tecnologia (IIT), Italy
Head of Artificial and Mechanical Intelligence
Researcher Tenure Track - Principal Investigator

Towards a Shared Embodied Intelligence of Humanoid Robotics: the human-aware ergoCub

Shared embodied intelligence is a phenomenon observed across various animal species, enabling herds to accomplish deeds that exceed individual capacities. It consists in the ability to coordinate with other agents, known as shared intelligence, and body exploitation, shaped by the evolutionary process, known as embodied intelligence. Drawing inspiration from Nature’s collabo- rative dynamics, this talk introduces a methodology to co-design and control humanoid robots to achieve shared-embodied intelligence, specifically focusing on human-robot physical collaboration.

Our contribution is twofold, encompassing scientific and technological aspects. Scientifically, we present models for human-robot physical interaction depending on the robot hardware. Additionally, we propose robot control frameworks that incorporate the physical intelligence of human partners, fostering partner awareness and enabling efficient collaborations. These components are leveraged to define a shared embodied intelligence modular architecture, developed to co-design and control humanoid robots. Technologically, we introduce the humanoid robot ergoCub, a tangible implementation of the proposed archi- tecture. The ergoCub robot is characterised by body and behaviors optimised considering human ergonomics metrics. We analyze the process that optimised ergoCub body characteristics and we showcase ergoCub ability to collaborate with humans, manipulate objects, and walk robustly. Our approach explores how robots can adapt to the cognitive and physical attributes of human coun- terparts, with applications spanning from industrial to assistive robotics.

Biography

Daniele received the bachelor and master degrees in Control Engineering with highest honors from ”Sapienza”, University of Rome, in 2007 and 2009, respectively. In 2013, he earned the PhD title with a thesis prepared at INRIA Sophia Antipolis, France, with the supervision of Tarek Hamel, Salvatore Monaco, and Claude Samson. From 2013 to 2017, he has been a postdoc at the Istituto Italiano di Tecnologia (IIT) working within the EU project CoDyCo focusing on the balancing problem of the iCub humanoid robot. From August 2017 to August 2021, he has been the head of the Dynamic Interaction Control lab, a group of about 20 members focusing on the iCub locomotion walking problem. In this period, Daniele also laid the basis for the "Aerial Humanoid Robotics", a new branch of Robotics whose main aim is to achieve flying humanoid robots. Daniele has also been the scientific PI of the H2020 European Project AnDy, and now is: task leader of the H2020 European Project SoftManBot, coordinator of the joint laboratory between IIT and Honda JP, principal investigator (PI) in the Camozzi-IIT and Danieli Automation-IIT joint labs. Lastly, Daniele is the coordinator of the ergoCub project, a 5 million, three year joint project between INAIL and IIT.  Since September 2021, Daniele is the PI leading the  Artificial and Mechanical Intelligence  research line at IIT, a team composed of about forty members that combines AI and Mechanics to devise the next generation of the iCub humanoid robot.
Concerning awards and recognitions, in 2009 Daniele received the ”Excellence Path Award” from Sapienza for his master studies. In 2019, he was awarded as Innovator of the year Under 35 Europe  from the MIT Technology Review magazine for conceiving the iRonCub, the first jet-powered flying humanoid robot. Since 2020, Daniele is a visiting lecturer at University of Manchester. In July 2020, Daniele was selected to represent Italy at the Global Partnership on Artificial Intelligence (GPAI, conceived by Canada and France during their G7 presidencies) for the working group on COVID-19 response. In 2022, Daniele was selected as reviewer of EU projects. Also, in 2022, Daniele was the leader of the iCub team during the ANA Avatar XPRIZE, a four-year 10M$ international competition for making robotic avatars. Starting from 100 initial teams, the iCub team scored 2nd after semifinal, and 14th after finals because of a system failure during day-1 testing; the iCub team scored 1st in the specific category of walking humanoids.