BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Robotics Institute Carnegie Mellon University - ECPv6.15.12.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Robotics Institute Carnegie Mellon University
X-ORIGINAL-URL:https://www.ri.cmu.edu
X-WR-CALDESC:Events for Robotics Institute Carnegie Mellon University
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20050403T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20051030T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20060402T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20061029T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20070311T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20071104T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20080309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20081102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20090308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20091101T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20100314T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20101107T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20110313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20111106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20120311T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20121104T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20130310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20131103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20140309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20141102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20150308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20151101T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20160313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20161106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20170312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20171105T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20180311T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20181104T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171011T120000
DTEND;TZID=America/New_York:20171011T130000
DTSTAMP:20260719T154951
CREATED:20171010T154204Z
LAST-MODIFIED:20171010T154516Z
UID:101171-1507723200-1507726800@www.ri.cmu.edu
SUMMARY:The Critical Radius in Sampling-Based Motion Planning
DESCRIPTION:Abstract \n———— \nMotion planning is a fundamental problem in robotics: allowing autonomous robots to efficiently navigate in environments cluttered with obstacles. Sampling-based algorithms\, which were first described two decades ago\, have made a great impact on the field of robotics by providing simple but highly-effective tools for motion planning. These techniques aim to capture the structure of the problem by randomly sampling robot configurations and connecting nearby samples\, to form discrete graphs which approximate the robot’s range of movements. \nIn this talk I will describe a new result concerning the theoretical asymptotic guarantees of sampling-based planners\, which significantly improves upon the celebrated work of Karaman and Frazzoli (2011). Particularly\, we prove that the number of neighbors considered for connection to each sample can be reduced from O(log n) (where n is the number of samples) to only O(1)\, without sacrificing asymptotic completeness or optimality. Continuum percolation theory plays an important role in our proofs. \nThe talk is based on the following paper:  \nKiril Solovey and Michal Kleinbort\, “The Critical Radius in Sampling-Based Motion Planning”\, arXiv:1709.06290\, 2017. \nBiography \n————— \nKiril Solovey is a Ph.D. student at the School of Computer Science\, Tel-Aviv University\, Israel\, working under the guidance of Dan Halperin. His research interests include sampling-based techniques for robot motion planning\, multi-robot systems\, and computational geometry. Kiril is a Clore Scholar and a recipient of the Best Student Paper award at Robotics: Science and Systems 2015. \nContact information: kirilsol@post.tau.ac.il / kirilsol.github.io
URL:https://www.ri.cmu.edu/event/critical-radius-sampling-based-motion-planning/
LOCATION:Newell Simon Hall 1507
CATEGORIES:CFR Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2017/10/cfr_logo.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20110405T163000
DTEND;TZID=America/New_York:20110405T000000
DTSTAMP:20260719T154951
CREATED:20190617T003731Z
LAST-MODIFIED:20190617T003731Z
UID:114663-1302021000-1301961600@www.ri.cmu.edu
SUMMARY:Robust Grasping Under Uncertainty in Pose and/or Shape
DESCRIPTION:Event Location:  NSH 3305Bio: Kaijen Hsiao received her B.S.E. degree in Mechanical Engineering in 2002\nfrom Princeton University and her Ph.D. degree in Computer Science in 2009\nfrom MIT\, where she worked with Tomas Lozano-Perez and Leslie Kaelbling. She\nis now a research scientist at Willow Garage\, where her research interests\nlie in grasping and manipulation. \nAbstract: Robotic manipulation of objects is much more difficult in unstructured\nenvironments (such as peoples’ homes) than in structured ones (such as\nfactories) because of the presence of uncertainty. Uncertainty comes in many\nforms in manipulation tasks: uncertainty in object identities\, poses\, or\nshapes\, or in robot poses or shapes\, for instance. In this talk\, I will\ndiscuss my thesis work on grasping objects of known shape robustly under\nsignificant uncertainty in object pose. To reason explicitly about\nuncertainty while grasping\, we model the problem as a partially observable\nMarkov decision process (POMDP). We derive a closed-loop strategy that\nmaintains a belief state (a probability distribution over world states)\, and\nselect actions with a receding horizon using forward search through the\nbelief space. Our actions are world-relative trajectories (WRT): fixed\ntrajectories expressed relative to the most-likely state of the world. We\nlocalize the object\, ensure its reachability\, and robustly grasp it at a\nspecified position by using information-gathering\, reorientation\, and\ngoal-seeking WRT actions. This framework is used to grasp objects (including\na power drill and a Brita pitcher) despite significant pose uncertainty\,\nusing a 7-DOF Barrett Arm and attached 4-DOF Barrett Hand equipped with\nforce and contact sensors. I will also present more recent work on\nreactively adjusting grasps in a model-free way using tactile sensors during\ngrasp execution\, as well as work on selecting grasps under uncertainty in\nobject shape\, in which we probabilistically combine results from multiple\ngrasp planners/evaluators on multiple potential object representations to\nselect grasps that are most likely to work given all the possible object\nhypotheses. Both are demonstrated using the PR2 robot from Willow Garage.
URL:https://www.ri.cmu.edu/event/robust-grasping-under-uncertainty-in-pose-and-or-shape/
CATEGORIES:CFR Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20100908T160000
DTEND;TZID=America/New_York:20100908T000000
DTSTAMP:20260719T154951
CREATED:20190617T003520Z
LAST-MODIFIED:20190617T003520Z
UID:114183-1283961600-1283904000@www.ri.cmu.edu
SUMMARY:Connection Vector Fields and Optimized Coordinates for Swimming Systems at Low and High Reynolds Numbers
DESCRIPTION:Event Location:  NSH 1507 \nAbstract: Several efforts have recently been made to relate the displacement of swimming three-link systems over strokes to geometric quantities of the strokes. While this approach has been successful for finding net rotations\, noncommutativity concerns have prevented it from working for net translations. Our recent results on other locomoting systems have shown that the degree of this noncommutativity is dependent on the coordinates used to describe the problem\, and that it can be greatly mitigated by an optimal choice of coordinates. Here\, we extend the benefits of this optimal-coordinate approach to the analysis of swimming at the extremes of low and high Reynolds numbers.
URL:https://www.ri.cmu.edu/event/connection-vector-fields-and-optimized-coordinates-for-swimming-systems-at-low-and-high-reynolds-numbers/
CATEGORIES:CFR Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20091020T163000
DTEND;TZID=America/New_York:20091020T000000
DTSTAMP:20260719T154951
CREATED:20190617T003517Z
LAST-MODIFIED:20190617T003517Z
UID:114175-1256056200-1255996800@www.ri.cmu.edu
SUMMARY:Synthesis of Controllers to Create\, Maintain\, and Reconfigure Robot Formations with Communication Constraints
DESCRIPTION:Event Location:  NSH 1507 \nAbstract: Using multiple robots in place of a single complex robot to accomplish\na task has many benefits\, including simplified system repair\, less\ndown time\, and lower cost. Combining groups of these multi-robot\nsystems allows addressing multiple subtasks in parallel\, reducing the\ntime it takes to address many problems\, such as search and rescue and\nautomated warehouse systems. \nI will address the synthesis of controllers for groups of multi-robot\nsystems that enable them to automatically create desired labeled\nformations and maintain those formations while traversing an\nenvironment with obstacles. The robots have constraints on\ncommunication\, both within and across groups. In a group\, individuals\nare capable of close coordination via\nhigh bandwidth communication\, since they are within a specified\ndistance of the other robots. Coordination across groups must be\nlimited because communication links can be sporadic or more expensive. \nI will describe a method for developing feedback controllers for\nreconfiguring groups of robots that is entirely automatic\, and\nprovably correct by construction. This work provides a framework with\nwhich navigation of multiple groups in environments with obstacles is\npossible\, and enables scaling to many groups of robots.
URL:https://www.ri.cmu.edu/event/synthesis-of-controllers-to-create-maintain-and-reconfigure-robot-formations-with-communication-constraints/
CATEGORIES:CFR Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20091019T153000
DTEND;TZID=America/New_York:20091019T000000
DTSTAMP:20260719T154951
CREATED:20190617T003518Z
LAST-MODIFIED:20190617T003518Z
UID:114177-1255966200-1255910400@www.ri.cmu.edu
SUMMARY:Towards Small Robot Aided Victim Manipulation and a Few Other Things.
DESCRIPTION:Event Location:  NSH 1507Bio: Mark Yim is an Associate Professor at the University of Pennsylvania.\nPrior to this\, he was Principal Scientist at the Palo Alto Research\nCenter (formerly Xerox PARC) where he established a group developing\nmodular self-reconfigurable robots.  His group has demonstrated\nmodular robots that can form different shapes\, jump\, ride tricycles\,\nclimb stairs\, poles and fences\, manipulate objects and reassemble\nthemselves after being kicked into pieces.  His other research\ninterests include biologically inspired mechanims\, flying robots and\nmeso-scale MEMs devices. He has over 40 patents issued (perhaps most\nprominent are ones related to the Sony PS2 and Microsoft Xbox joypad\nvibration control which resulted in US$100\,000\,000 in litigation and\nsettlements). \nAbstract: Robotic manipulation of human victims is a problem that has been\navoided by the robotics community. This talk will present some of the\nissues in using small robots aiding the transport of incapacitated\nvictims. Focusing on the manipulation of victims in preparation for\ntransport\, the primary design specifications come from the required\nforces that a small robot would need to be capable of applying.  In\naddition\, I’ll talk about a variety of other things that are happening\nin the modlab part of the GRASP lab at the University of Pennsylvania.
URL:https://www.ri.cmu.edu/event/towards-small-robot-aided-victim-manipulation-and-a-few-other-things/
CATEGORIES:CFR Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20090929T163000
DTEND;TZID=America/New_York:20090929T000000
DTSTAMP:20260719T154951
CREATED:20190617T003519Z
LAST-MODIFIED:20190617T003519Z
UID:114179-1254241800-1254182400@www.ri.cmu.edu
SUMMARY:Structure in Flying Vertebrates: Movement in Unpredictable Environments
DESCRIPTION:Event Location:  NSH 1507 \nAbstract: Prior experimental data has indicated that flying animals operate at\nsafety factors much greater than those generally used in manufactured\nflying vehicles.  However\, comparative data on the range of structural\nstrengths represented by living flyers\, and their correlations with\nspecific modes of movement\, are considerably less common in the\nliterature.  Flying animals may provide insights for robotic design\,\nespecially with regards to micro air vehicles\, but also for any systems\ndesigned to move in unpredictable environments with strict weight\nrequirements.  Birds\, bats\, and pterosaurs all demonstrate a wide array of\nadaptations related to maintaining structural integrity in unpredictable\nenvironments.  In this talk\, I will present the results of my recent work\non the skeletal rigidity in the wings and hind limbs of a wide range of\nflying species.  I place special emphasis on pterosaurs\, which have often\nbeen overlooked in biomechanical analyses because they have no living\nrepresentatives.  Despite being an enigmatic fossil group\, pterosaurs\nprovide numerous insights regarding weight distribution\, passive load\naccommodation\, and multi-purpose limb systems.  Pterosaurs are especially\ninformative for understanding the size limits of flapping flyers (as in\nthe construction of ornithopters)\, as they included the largest known\nflying animals.  I will supply evidence that pterosaur forelimbs acted as\na multi-purpose module\, supplying both aerodynamic force in flight and\nacting as a primary launching module\, by supplying leaping force\, when\nentering flight from the ground.  I will further demonstrate that failure\nloads in the forelimb of living birds encompass a nearly seven-fold range\,\nwhile bats are considerably more constrained\, and utilize less rigid wing\nskeletons.  Finally\, adaptations to outboard stall reduction and gust\nalleviation will be considered.
URL:https://www.ri.cmu.edu/event/structure-in-flying-vertebrates-movement-in-unpredictable-environments/
CATEGORIES:CFR Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20060206T170000
DTEND;TZID=America/New_York:20060206T000000
DTSTAMP:20260719T154951
CREATED:20190617T003519Z
LAST-MODIFIED:20190617T003519Z
UID:114181-1139245200-1139184000@www.ri.cmu.edu
SUMMARY:Algebraic Topological Tools for Sensor Networks
DESCRIPTION:Event Location:  NSH 1507 \nAbstract: As sensor engineering and manufacturing evolve\nto produce smaller devices\, we will face the problem of dealing\nwith large collections of local networked devices. What types\nof global problems can be solved by a swarm of local sensors?\nTopologists solved a similar problem nearly a century ago\, and in\nso doing invented some very sophisticated algebraic tools. This\ntalk will demonstrate the surprising effectiveness of algebraic\ntopology as a toolbox for working with sensor networks having\nneither localization capabilities nor probabilistic assumptions.
URL:https://www.ri.cmu.edu/event/algebraic-topological-tools-for-sensor-networks/
CATEGORIES:CFR Seminar
END:VEVENT
END:VCALENDAR