Yiannis Aloimonos, Computer Vision Laboratory – University of Maryland, College Park
Hyper-dimensional Active Perception: a journey into the mind of a robotic bee
Abstract: Action and perception are often kept in separated spaces, which is a consequence of traditional vision being frame-based and only existing in the moment and motion being a continuous entity. This bridge is crossed by the dynamic vision sensor (DVS), a neuromorphic camera that can see the motion. We propose a method of encoding actions and perceptions together into a single space that is meaningful, semantically informed, and consistent by using hyper-dimensional binary vectors (HBVs). We show that the visual component can be bound with the system velocity to enable dynamic world perception, which creates an opportunity for real-time navigation and obstacle avoidance with active perception. Actions performed by an agent are directly bound to the perceptions experienced to form its own “memory.” Furthermore, because HBVs can encode entire histories of actions and perceptions—from atomic to arbitrary sequences—as constant-sized vectors, auto-associative memory was combined with deep learning paradigms for controls. Using this methodology we can implement, for manoquadrotors with all processing on board, a hierarchy of sensorimotor loops providing a set of competences (egomotion, moving object detection, obstacle avoidance, homing and landing) which can be interfaced with episodic, procedural and semantic memory, giving rising to a minimal cognitive system. The talk will conclude with a journey into the mind of a robotic bee using a variety of experimental results.
Biography: Yiannis Aloimonos is a Professor of Computational Vision and Intelligence at the Department of Computer Science, University of Maryland, College Park, and the Director of the Computer Vision Laboratory at the Institute for Advanced Computer Studies (UMIACS). He is also affiliated with the Institute for Systems Research, the Neural and Cognitive Science Program and the Maryland Robotics Center. He was born in Sparta, Greece and studied Mathematics in Athens and Computer Science at the University of Rochester, NY (PhD 1990). He is interested in Active Perception and the modeling of vision as an active, dynamic process for real-time robotic systems. For the past five years he has been working on bridging signals and symbols, specifically on the relationship of vision to control, and the relationship of action and language using Hyper-dimensional Computing.
Richard Bomphrey – Professor of Comparative Biomechanics – Royal Veterinary College
Bio-inspired aerodynamics, actuation, sensing and control
Abstract: How can the observation of animal aerodynamics inform bio-inspired aircraft? I will discuss how fundamental bioscience research can reach outside animal ecology and physiology to offer solutions to engineering challenges.
Flying animals must perceive and avoid obstacles, often in environments deprived of visual sensory cues. In my first example, I will show how collision-avoidance in nocturnal mosquitoes can be mediated by mechanosensory feedback, based on modulations of their own induced aerodynamic and acoustic fields as they enter ground- or wall-effect. Our computational fluid dynamics and aeroacoustic simulations are derived from detailed wing kinematics extracted from high-speed recordings of freely flying Culex quinquefasciatus mosquitoes. Based on Culex data, we have built palm-sized prototypes carrying a suite of bio-inspired sensor packages. I will discuss how aerodynamic and aeroacoustic cues that are associated with close proximity to the ground and wall planes could provide useful information to the flight controller (a mechanism we term ‘aerodynamic imaging’) and how mosquitoes perceive and interact with others using acoustic interactions.
In my second example, I will present our work based on measuring the changing shape of birds in flight. I will show how they minimise drag in a different way from conventional aircraft design on account of their smaller size, and how they remain unperturbed by strong gusts. Our detailed three-dimensional reconstructions of surface geometries show how wing elevation around the shoulder joint acts as an initially passive suspension system that rejects gusts, steadying the payload. The mechanism works most effectively when the aerodynamic centre of pressure is aligned with the mechanical centre of percussion, and therefore can be tuned either by changing wing shape or by the distribution of mass within the wing. Together, these topics give a small glimpse into the potential that continues to exist within the natural world for the improvement of small aircraft.
Biography: Richard’s research blends biology and engineering. He uses biomechanics as a tool to investigate evolutionary biology and how the physical environment determines the morphology and control systems of flying animals. He has worked on the sensory mechanisms of insects and birds, including flow-sensing, load-sensing, and optic flow. Richard’s work uses advanced equipment to investigate animal flight and understand their aerodynamic footprints by observing the motion of smoke or bubbles floating in the air. He has applied insights from biology to aerial robots inspired by birds and insects. Richard joined the Structure and Motion Laboratory at the Royal Veterinary College, University of London, in 2013 after reading Biological Sciences at Exeter, undertaking a DPhil (PhD) in Oxford, postdoctoral positions in Oxford and Bath, and an EPSRC Fellowship. He is currently Professor of Comparative Biomechanics at the Royal Veterinary College and Interim Vice Principal for Research.
Antonio Franchi – Professor in Aerial Robotics Control – Twente University
Is control a solved problem in robotics MAV research?
Abstract: In these days MAV and robotics research in general are dominated by data-driven approaches that are mainly focused on scientific problems related to perception and other AI-related problems. The design, modeling, and control problems, which have been the backbone of MAV research in the past decades seem to suffer a period of decline in terms of popularity. There are many possible explanations for this phenomenon. Perhaps there are no more relevant open control problems for UAVs; or maybe still unsolved problems are now perceived as solved because of some sort of ‘mass hallucination’; or perhaps the attention is now distracted by other problems which have been neglected in the past because too hard to solve and now seem/are finally within reach thanks to the new AI wave; or another explanation could be that the problems still open in control are so difficult that everybody pretend to ignore them, like the ‘Fox and the Grapes’. Or the motivation could be a different one of course. Or perhaps there is no decline in popularity at all and this research field is actually in very good shape but just not advertised enough to emerge.
In this keynote I will give my historical and technical point of view on the question whether there are still significant open scientific problems in UAV research on the design and control side and, in case the answer is yes, I will try to critically rank such open problems in terms of difficulty and provide possible technical directions to explore the paths toward their solutions.
Biography: Antonio Franchi (https://homepages.laas.fr/afranchi/robotics/) is a Full Professor in Aerial Robotics Control at the University of Twente, The Netherlands, in the Robotics and Mechatronics department, and an associated researcher at LAAS-CNRS (RIS team), Toulouse, France. He is a IEEE Senior Member.
From 2014 to 2019 he was a Permanent Researcher at CNRS and the leader of the aerial robotics activities at LAAS-CNRS. From 2010 to 2013 he was a Research Scientist and then a Senior Research Scientist at the Max Planck Institute for Biological Cybernetics in Germany, and the scientific leader of the group “Autonomous Robotics and Human Machine Systems”. He received the Laurea (M.Sc.) degree (summa cum laude) in Electronic Engineering and the Ph.D. degree in System Engineering (Feb. 2010) from Sapienza University of Rome, Italy. In 2009 he was a visiting student at the University of California at Santa Barbara.
His main research interests lie in the robotics area, with a special regard to control and estimation problems and applications ranging across motion and physical interaction control, decentralized control/estimation/coordination, haptics, and hardware/software architectures. His main areas of expertise are aerial robotics and multiple-robot systems.
He published more than 150 papers in international journals, books, and conferences and gave more than 90 invited talks in international venues since 2010. In 2010 he was awarded with the ‘’IEEE RAS ICYA Best Paper Award’’ for one of his works on Multi-robot Exploration. In 2018 he was a recipient of the 2018 IEEE RAS Most Active Technical Committee Award.
He was Associate Editor of the IEEE Transactions on Robotics from 2016 until 2021 and he is Senior Editor for IEEE ICRA since 2020. He has been associate editor of the IEEE Robotics & Autom. Mag. (2013 to 2016), IEEE ICRA (2014 to 2019), IEEE/RSJ IROS (2014 to 2017) and the IEEE Aerospace and Electric Systems Magazine (2015).
He is the project coordinator of JCJC ANR MuRoPhen, co-coordinator of the FlyCrane project, and a was a contributor to PRO-ACT H2020, all focused on multi-robot coordination/manipulation. He is the local coordinator of the EU H2020 Aerial-CORE project, he has been local coordinator in the EU H2020 AEROARMS project, the creator of the ANR PRC ‘The Flying Co-worker’ project, and he contributed to the EU FP7 ARCAS project, all focused on aerial robotic manipulation.
He is the co-founder and co-chair of the IEEE RAS Technical Committee on Multiple Robot Systems, http://multirobotsystems.org/ (>450 members).
He co-funded and was the program co-chair of the IEEE-sponsored International Symposium on Multi-robot and Multi-agent Systems (MRS 2017, 2019, and 2021).
He co-organized the IEEE-RAS-sponsored 2019 and 2016 Summer Schools on Multiple Robot Systems at NUS, Singapore and CTU, Prague, and more than 15 workshops on Multi-robot Systems, Aerial Robots and Teleoperation at IEEE ICRA, IEEE/RSJ IROS, and RSS, among others.
Since 2010, he has been the mentor of 11 graduated PhD student and he is currently mentoring 8 PhD candidates. One the PhD theses he mentored won the Best Robotics Thesis French National Award in 2019.
Laurent PERCHAIS, CEO – Rémi POTET & Corentin GIRAUD, Drone Engineers at DRONISOS
Storytelling with drone swarms
Abstract: Dronisos, with its unrivalled technology developed completely in-house, has turned the sky into the world’s biggest 3-dimensional screen using thousands of drones as animated pixels.
Major Events, Global Brands and world-renowned Theme Parks have all adopted this new medium of communication and entertainment, making Dronisos a worldwide leader in drone entertainment with over 15,000 performances to date.
Dronisos has invented a new medium for storytellers to bring their stories to life on the world’s biggest stage – the sky.
However, for this keynote, we want to tell you OUR story. The story of a 5-year journey a team of engineers passionate about drones took on, inventing a new medium of expression while tackling technical and scientific challenges of autonomous drone swarms along the way.
Biography:
Laurent PERCHAIS is the CEO of Dronisos. He is a graduate of the Ecole Polytechnique (Paris, France) and HEC Business School (Paris, France). Following his first role at Canal+ (French TV player), he held various positions at Orange (French Telco) both as an Engineer and as a Business manager, where he ended as Head of Strategy for Orange Content (TV, Music, Gaming….) before becoming Dronisos CEO in 2017.
Remi POTET is a former IMAV participant and now a Dronisos expert in drone engineering. He holds an MSc from ISAE SUPAERO(Toulouse, France) and has been part of Dronisos’ journey from almost the beginning (2018).
Corentin GIRAUD is a drone industry veteran. After a successful experience at Azur Drones, he joined Dronisos in 2020 to lead the development of Dronisos in the Defense & Safety sector. He graduated from ISAE-SUPAERO (Toulouse, France).