BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Robotics Institute Carnegie Mellon University - ECPv6.15.12.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
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: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
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20191103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20200308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20201101T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20210314T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20211107T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20220313T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20221106T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20231105T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20240310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20241103T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20250309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20251102T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241206T140000
DTEND;TZID=America/New_York:20241206T150000
DTSTAMP:20260717T173551
CREATED:20241202T171900Z
LAST-MODIFIED:20241202T171900Z
UID:144633-1733493600-1733497200@www.ri.cmu.edu
SUMMARY:From Lab to Launch
DESCRIPTION:Bio: Nathan Michael is Shield AI’s Chief Technology Officer and a former Associate Research Professor in the Robotics Institute of Carnegie Mellon University (CMU). At CMU\, Nathan was the Director of the Resilient Intelligent Systems Lab\, a research lab dedicated to improving the performance and reliability of artificially intelligent and autonomous systems that operate in challenging\, real-world and GPS-denied environments. Nathan has authored over 150 publications on control\, perception\, and cognition for artificially intelligent single and multi-robot systems\, for which he has been nominee or recipient of nine best paper awards (ICRA\, RSS\, DARS\, CASE\, SSRR) as well as recipient of the Popular Mechanics Breakthrough Award and Robotics Society of Japan Best Paper Award. Over his decades-long career in research\, Nathan has led research programs supported by ARL\, AFRL\, DARPA\, DOE\, DTRA\, NASA\, NSF\, ONR\, and industry. \n  \nLocation: CIC (LL Level) Conference Room #1\nNote: Photos and video recordings are not permitted for this talk. Zoom is unavailable.
URL:https://www.ri.cmu.edu/event/from-lab-to-launch/
LOCATION:CIC\, CIC Buuilding Conference Room 1\, LL Level
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2024/12/Nathan-Michaels-Page.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241119T160000
DTEND;TZID=America/New_York:20241119T170000
DTSTAMP:20260717T173551
CREATED:20241108T194256Z
LAST-MODIFIED:20241108T194958Z
UID:144207-1732032000-1732035600@www.ri.cmu.edu
SUMMARY:A retrospective\, 40 Years of Field Robotics
DESCRIPTION:Abstract: Chuck has been building and deploying robots in the field for the past 40 years.  In this retrospective he will touch on the robots\, people and experiences that have been part of the journey.  From the early days in the 1980s with the Three Mile Island nuclear robots and the first outdoor autonomy robots Terregator and Navlab.  Through the 1990s with RedZone Robotics and nuclear robots plus the Dante volcanic explorer robot at CMU.  Transitioning back to CMU and the past 20 years of Field Robots.  The skills that are needed\, lessons learned\, achievements and great people through the decades.\n\n\nBio:\nChuck Whittaker\, Senior Field Robotics Specialist at Carnegie Mellon University.\n2004 -2024 Field Robotic Specialist CMU\, Field Robotics Center\, Robotics Institute\n2002 – 2004 Independent Robotic Contractor\n1988 – 2002 Technical Manager RedZone Robotics\n1985 – 1988 Technician CMU Civil Engineering and Robotics Institue\n1977 – 1985 Mining Engineer PBS Coals / Mincorp\n1974 – 1978 BS Civil Engineering University of Pittsburgh at Johnstown
URL:https://www.ri.cmu.edu/event/a-retrospective-40-years-of-field-robotics/
LOCATION:CIC\, CIC Buuilding Conference Room 1\, LL Level
CATEGORIES:Field Robotics Center Seminar,Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20241008T160000
DTEND;TZID=America/New_York:20241008T170000
DTSTAMP:20260717T173551
CREATED:20241007T173602Z
LAST-MODIFIED:20241007T173602Z
UID:143568-1728403200-1728406800@www.ri.cmu.edu
SUMMARY:Using Robotics\, Imaging and AI to Tackle Apple Fruit Production: Crop Harvest and Fire Blight Disease\, The Two Major Bottlenecks for U.S. Apple Producers
DESCRIPTION:Abstract \nTemperate tree fruit production is a significant agricultural sector in the United States\, encompassing a variety of fruits like apples\, pears\, cherries\, peaches and plums. The U.S. is the second-largest producer of apples in the world\, after China. Annual U.S. production is 10 – 11 billion pounds of apple. However\, apple production is complicated by two major bottlenecks: dependence on and diminishing availability of hand labor for apple harvesting and a devastating bacterial plant disease called fire blight\, that causes apple fruit losses and tree death. The former is the key issue driving many field robotics companies to develop semi-automated harvest platforms and test automated robotic harvesters. The latter is the one of the most difficult management problems for growers\, due to the difficult nature of the disease leading to the destruction of flowers\, fruit\, wood and whole trees. This seminar will present the key technological aspects of apple fruit industry revolution and how robotics shape the way apple trees are grown today and how it will play a key role in battling human labor issues that producers face. At last\, this talk will address the ideas for new ways how robotics and neural networks can help tackle one of the deadliest fire blight symptoms on apple which kill the whole trees: fire blight cankers. Carnegie Mallon stereo camera with strobe lighting and AI-recognition algorithms can help detect and map fire blight cankers on apple trees thus allowing their complete removal by pruning. Join us in discussing how field robotics can help agriculture in solving real agricultural problems.
URL:https://www.ri.cmu.edu/event/using-robotics-imaging-and-ai-to-tackle-apple-fruit-production-crop-harvest-and-fire-blight-disease-the-two-major-bottlenecks-for-u-s-apple-producers/
LOCATION:CIC\, CIC Buuilding Conference Room 1\, LL Level
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/png:https://www.ri.cmu.edu/app/uploads/2024/10/Srdjan-Acimovic-2.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20240725T150000
DTEND;TZID=America/New_York:20240725T160000
DTSTAMP:20260717T173551
CREATED:20240722T043156Z
LAST-MODIFIED:20240725T170429Z
UID:141898-1721919600-1721923200@www.ri.cmu.edu
SUMMARY:Robots Crossing Boundaries
DESCRIPTION:Abstract: \nOver the last 50 years\, autonomous robots have made the leap from being novel research contributions in university labs to becoming the fundamental technology upon which companies are built. While they traditionally have belonged to the engineering and computer science disciplines\, robots have now crossed into other areas of study and research – making impacts in oceanography\, geology\, archaeology\, biomechanics and biology. To exemplify these crossovers\, the speaker will discuss several interdisciplinary projects that his research team have contributed to: altruistic robotics\, underwater archeology\, autonomous shark tracking\, and education. These projects not only showcase several technical aspects of traditional robotics including motion planning\, machine learning\, state estimation\, systems integration\, and control theory\, but also highlight the impact of interdisciplinary research and education. \nBio: Christopher Clark is research scientist at Apple\, and has been a Professor at Harvey Mudd College since 2012. Before joining HMC’s faculty\, he served as a faculty member at the University of Waterloo and California Polytechnic State University\, San Luis Obispo. Clark is a Fulbright Scholar and for the 2011–2012 academic year\, he held the William R. Kenan\, Jr. Visiting Professorship for Distinguished Teaching at Princeton University.  In 2004\, he was a first hire at the startup company Kiva Systems (now Amazon Robotics)\, which changed warehouse management via multi-robot systems. He earned his undergraduate degree in engineering physics from Queen’s University\, Canada\, a master’s in mechanical engineering from the University of Toronto and a PhD in aeronautics and astronautics with a minor in computer science from Stanford University. Clark’s research areas include multi-robot systems\, underwater robot systems\, applied ML\, control theory\, intelligent vehicles\, state estimation and motion planning.
URL:https://www.ri.cmu.edu/event/robots-crossing-boundaries/
LOCATION:CIC\, CIC Buuilding Conference Room 1\, LL Level
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/png:https://www.ri.cmu.edu/app/uploads/2024/07/image.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20230404T130000
DTEND;TZID=America/New_York:20230404T140000
DTSTAMP:20260717T173551
CREATED:20230331T021352Z
LAST-MODIFIED:20230403T161505Z
UID:135443-1680613200-1680616800@www.ri.cmu.edu
SUMMARY:Autonomous mobility in Mars exploration: recent achievements and future prospects
DESCRIPTION:Abstract: \nThis talk will summarize key recent advances in autonomous surface and aerial mobility for Mars exploration\, then discuss potential future missions and technology needs for Mars and other planetary bodies. Among recent advances\, the Perseverance rover that is now operating on Mars includes new autonomous navigation capability that dramatically increases its traverse speed over previous rovers. Perseverance also carried the Ingenuity helicopter to Mars\, which is a technology demonstration of the first heavier-than-air aircraft ever to operate on another planet. The current mission objective for Perseverance involves driving a total distance of about 60 kilometers in about 10 Earth years. Rover mission concepts recently suggested for the Moon would drive about 1500 to 2000 km in under 4 years\, which requires significant advances in autonomy. Successors to the Ingenuity helicopter are now under development for use in a mission planned for later this decade to return Mars samples to Earth that Perseverance is collecting. Much larger helicopter concepts are being studied to enable carrying larger science instrument payloads for potential future Mars missions. Robotic surface and aerial vehicles\, as well as drilling systems for subsurface access\, potentially could play a role in NASA’s goals for a human mission to Mars roughly two decades from now. \nBio: \nDr. Larry Matthies is a Principal Technologist at the Jet Propulsion Laboratory (JPL) and serves as Technology Coordinator in the Mars Exploration Program Office. He joined JPL in 1989 after obtaining a PhD in computer science from Carnegie Mellon University and he supervised the JPL Computer Vision Group for 21 years prior to his current role. He and his group contributed new capabilities to every U.S. Mars surface mission since Mars Pathfinder in 1996\, including vision-based navigation capabilities for rovers\, precision landing\, and the Ingenuity helicopter. He has also been a principal investigator in many robotic air and ground vehicle programs funded by DARPA\, Army\, Navy\, and industrial sponsors\, with research emphasis on multi-sensor perception for autonomous navigation. He is a Fellow of the IEEE and a member of the editorial boards of the journals Field Robotics and Autonomous Robots.
URL:https://www.ri.cmu.edu/event/autonomous-mobility-in-mars-exploration-recent-achievements-and-future-prospects/
LOCATION:NSH 4305
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2023/03/Larry-Matthies.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221202T151500
DTEND;TZID=America/New_York:20221202T161500
DTSTAMP:20260717T173551
CREATED:20221130T184805Z
LAST-MODIFIED:20221130T184805Z
UID:134373-1669994100-1669997700@www.ri.cmu.edu
SUMMARY:Multi-Sensor Robot Navigation and Subterranean Exploration
DESCRIPTION:
URL:https://www.ri.cmu.edu/event/multi-sensor-robot-navigation-and-subterranean-exploration/
LOCATION:Newell-Simon Hall 4305
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2022/11/maurice_fallon-provided.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20221122T164500
DTEND;TZID=America/New_York:20221122T174500
DTSTAMP:20260717T173551
CREATED:20221115T222626Z
LAST-MODIFIED:20221115T222626Z
UID:134258-1669135500-1669139100@www.ri.cmu.edu
SUMMARY:Towards more effective remote execution of exploration operations using multimodal interfaces
DESCRIPTION:Abstract:\nRemote robots enable humans to explore and interact with environments while keeping them safe from existing harsh conditions (e.g.\, in search and rescue\, deep sea or planetary exploration scenarios). However\, the gap between the control station and the remote robot presents several challenges (e.g.\, situation awareness\, cognitive load\, perception\, latency) for effective teleoperation. Multimodal teleoperation interfaces have the benefits of offering feedback while reducing the burden of the operator’s visual channel. In this talk\, I will present the work of the MEROP team about multimodal interfaces (with emphasis on haptic devices) and how challenges for more effective remote teleoperation have been tackled. Results obtained and lessons learned from user studies performed in laboratory and field (e.g.\, in the AMADEE-20 Mars analog mission) will be presented. Finally\, directions for future work will be discussed.\nBio:\nJosé Luís Silva received a Ph.D. degree in Computer Science from the Portuguese MAP-i Consortium (University of Minho\, University of Aveiro and University of Porto)\, in 2012. From 2012 to 2013\, he performed a postdoc at the University of Toulouse (France) in collaboration with Airbus. From 2013 to 2016 he was Invited Assistant Professor at University of Madeira. He is\, since 2016\, Assistant Professor with the Department of Information Science and Technology\, Lisbon University Institute (ISCTE-IUL). He participated in several national and international research projects and in the AMADEE-20 Mars analog mission (organized by the Austrian Space Forum). His main research interests lie upon Software Engineering\, Human-Computer Interaction\, Human-Robot Interaction and in particular about multimodal interfaces for improved teleoperation of remote robots. He is a member of the Interactive Technologies Institute / LARSyS laboratory\, ISTAR-IUL research center and\, IFIP TC 13—Working Groups (13.2 and 13.10). His awards and honors include ISCTE-IUL Scientific Awards\, Best Iberian Ph.D. thesis from AISTI and PhD Award from Fraunhofer Portugal Challenge.
URL:https://www.ri.cmu.edu/event/towards-more-effective-remote-execution-of-exploration-operations-using-multimodal-interfaces/
LOCATION:1305 Newell Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2022/11/jose.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20200827T140000
DTEND;TZID=America/New_York:20200827T150000
DTSTAMP:20260717T173551
CREATED:20200815T160915Z
LAST-MODIFIED:20200828T142709Z
UID:123709-1598536800-1598540400@www.ri.cmu.edu
SUMMARY:Beyond ROS: Using a Data Connectivity Framework to build and run Autonomous Systems
DESCRIPTION:Virtual FRC Seminar:\nSeminar recording: https://cmu.zoom.us/rec/share/x84qF7_q8TlIcpHoyG_DRa58O6i8aaa8hCAW_fEPxEkBGjBVPyzW_lK0YW30RfJ3?startTime=1598551489000\nPasscode: qu6)ePH9 \n\nAbstract: \nNext-generation robotics will need more than the current ROS code in order to comply with the interoperability\, security and scalability requirements for commercial deployments. This session will provide a technical overview of ROS\, ROS2 and the Data Distribution Service™ (DDS) protocol for data connectivity in safety-critical cyber-physical systems. This session is intended for developers\, engineers and system architects of distributed autonomous systems. \nBill Drozd’s Team Explorer is a prime example of the systems where RTI Connext adds value: \n\n“Every day we expect more from the robots we make. For Team Explorer\, this meant increasing the size of our robot team from three to seven\, adding the capability for remote drone launches from ground vehicles\, all while remotely monitoring video streams from up to a dozen cameras (in less than 6 months). The lack of subterranean infrastructure means any software solution has to be tolerant to dropped packets\, high pings\, frequent disconnections and needs certain data to be resent automatically until reception is guaranteed. We chose RTI because we needed to go beyond what was possible with ROS-1 while still relying on a largely ROS-1 code base. The DDS messaging standard means that we could be certain that future versions of our software would integrate easily with any of the many existing commercial and government clients that rely on it\, and that the switch to ROS2 is always an option. Going far beyond simply providing this software we depend on for SubT\, the staff at RTI have at every opportunity worked hard to help us overcome our numerous and daunting technical challenges\, and we look forward to giving CMU the opportunity to learn from them.” – Bill Drozd\, Senior Research Programmer\, National Robotics Engineering Center\, Carnegie Mellon University \n\nThe focus will be on how the data-centric approach of DDS creates a scalable architecture that can address: \n\n\nReal-time performance\, determinism\, and production-level safety and security \n\n\nMulti-robot systems (swarms/platoons); inconsistent networks; and operation on constrained platforms \n\n\nModularity to manage development cost/risk\, complexity and enable scalability \n\n\nInteroperability\, ease of integration and common interfaces \n\n\n  \nSpeaker Bio: \nRoss Gilson is a Senior Field Applications Engineer with Real-Time Innovations (RTI)\, the leading DDS provider\, with over 1\,500 commercial projects in robotics\, autonomous systems\, connected healthcare\, aerospace and other complex distributed applications.  The company’s RTI Connext® DDS product enables intelligent architecture by sharing information in real time\, making large applications work together as one. \n 
URL:https://www.ri.cmu.edu/event/beyond-ros/
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2020/08/RTI_Logo_with-Tagline_Horizontal_RGB.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20191101T120000
DTEND;TZID=America/New_York:20191101T130000
DTSTAMP:20260717T173551
CREATED:20191025T025411Z
LAST-MODIFIED:20191025T025411Z
UID:118362-1572609600-1572613200@www.ri.cmu.edu
SUMMARY:Tartan AUV: A Dive into Carnegie Mellon’s RoboSub Team
DESCRIPTION:Abstract: \nFounded last year\, Tartan AUV is Carnegie Mellon’s undergraduate underwater robotics team which competes annually in the RoboSub competition. RoboSub teams must design\, build\, and test autonomous underwater vehicles that compete each August to complete tasks related to underwater navigation\, object detection and manipulation\, and acoustic beacon localization. In this talk we will provide an overview of the competition and our work so far\, then discuss our plans going forward. We will begin by laying out the tasks and challenges presented by the competition as well as some background information about the team. We will then discuss the design and implementation of our first vehicle\, which competed this past August\, before diving into the design of our next vehicle with a focus on our plans for reliability\, acoustic signal processing\, and underwater navigation and perception. The AUV operates in shallow\, turbid water with artifacts such as caustics from the surface and particulates in the water. Approaches such as GPS and LIDAR are unavailable underwater\, and robust underwater localization/perception sensors such as scanning sonars or doppler velocity logs (DVLs) are often prohibitively expensive. Our options include visual inertial odometry and landmark-based localization methods. We must overcome challenges such as low ground visibility in certain circumstances and supplement our localization with auxiliary sensor data. We will discuss possible solutions to these problems as well as the challenges faced. \n\n\n\nSpeaker Bios:\nTom Scherlis is a Junior at Carnegie Mellon majoring in ECE and Robotics. Tom co-founded Tartan AUV last year and developed the electrical system for the first vehicle. Now\, Tom leads the developmet of Tartan AUVs new software stack. Last summer\, Tom worked as an embedded software intern at Zipline International working on software for medical drone delivery. \nAdvaith Sethuraman is an undergraduate Junior in ECE at Carnegie Mellon University. Advaith joined Tartan AUV this year\, mainly focusing on solutions for Perception and Localization. Advaith has worked at Qualcomm Technologies as a Chipset Firmware Intern\, and more recently at Intel in the field of Cloud Mesh Streaming acceleration.
URL:https://www.ri.cmu.edu/event/tartan-auv-a-dive-into-carnegie-mellons-robosub-team/
LOCATION:NSH 4305
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/png:https://www.ri.cmu.edu/app/uploads/2019/10/tom_advaith-1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20191016T113000
DTEND;TZID=America/New_York:20191016T133000
DTSTAMP:20260717T173551
CREATED:20191003T182933Z
LAST-MODIFIED:20191003T182933Z
UID:118001-1571225400-1571232600@www.ri.cmu.edu
SUMMARY:Multiple Drone Vision and Cinematography
DESCRIPTION:Abstract:\nThe aim of drone cinematography is to develop innovative intelligent single- and multiple-drone platforms for media production to cover outdoor events (e.g.\, sports) that are typically distributed over large expanses\, ranging\, for example\, from a stadium to an entire city.  The drone or drone team\, to be managed by the production director and his/her production crew\, will have: a) increased multiple drone decisional autonomy\, hence allowing event coverage in the time span of around one hour in an outdoor environment and b) improved multiple drone robustness and safety mechanisms (e.g.\, communication robustness/safety\, embedded flight regulation compliance\, enhanced crowd avoidance and emergency landing mechanisms)\, enabling it to carry out its mission against errors or crew inaction and to handle emergencies. Such robustness is particularly important\, as the drones will operate close to crowds and/or may face environmental hazards (e.g.\, wind). Therefore\, it must be contextually aware and adaptive\, towards maximizing shooting creativity and productivity\, while minimizing production costs.\nDrone vision plays an important role towards this end\, covering the following topics: a) drone visual mapping and localization\, b) drone visual analysis for target/obstacle/crowd/POI detection\, c) 2D/3D target tracking and d) privacy protection technologies in drones (face de-identification).\nThis lecture will offer an overview of current research efforts on all related topics\, ranging from visual semantic world mapping to multiple drone mission planning and control and to drone perception for autonomous target following\, tracking and AV shooting.\n \nSpeaker Bio:\nProf. Ioannis Pitas (IEEE fellow\, IEEE Distinguished Lecturer\, EURASIP fellow) received the Diploma and PhD degree in Electrical Engineering\, both from the Aristotle University of Thessaloniki\, Greece. Since 1994\, he has been a Professor at the Department of Informatics of the same University. He served as a Visiting Professor at several Universities.\nHis current interests are in the areas of autonomous systems\, machine learning\, computer vision\, 3D and biomedical imaging. He has published over 1090 papers\, contributed in 50 books in his areas of interest and edited or (co-)authored another 11 books. He has also been a member of the program committee of many scientific conferences and workshops. In the past he served as Associate Editor or co-Editor of 9 international journals and General or Technical Chair of 4 international conferences. He participated in 69 R&D projects\, primarily funded by the European Union and is/was principal investigator/researcher in 41 such projects. He has 29200+ citations to his work and h-index 80+ (Google Scholar).\nProf. Pitas leads the big European H2020 R&D project MULTIDRONE: https://multidrone.eu/. He is chair of the IEEE Autonomous Systems Initiative (ASI) https://ieeeasi.signalprocessingsociety.org/
URL:https://www.ri.cmu.edu/event/multiple-drone-vision-and-cinematography/
LOCATION:3305 Newell-Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2019/10/pitas.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190918T130000
DTEND;TZID=America/New_York:20190918T140000
DTSTAMP:20260717T173551
CREATED:20190912T014226Z
LAST-MODIFIED:20190915T234806Z
UID:117682-1568811600-1568815200@www.ri.cmu.edu
SUMMARY:Self-Supervised Learning on Mobile Robots Using Acoustics\, Vibration\, and Visual Models to Build Rich Semantic Terrain Maps
DESCRIPTION:Abstract:\nHumans and robots would benefit from having rich semantic maps of the terrain in which they operate.  Mobile robots equipped with sensors and perception software could build such maps as they navigate through a new environment.  This information could then be used by humans or robots for better localization and path planning\, as well as a variety of other tasks.  However\, it is hard to build good semantic maps without a great deal of human effort and robot time.  Others have addressed this problem\, but they don’t provide a high level of semantic richness\, and in some cases their approaches require extensive human data labeling and robot driving time. \nWe use a combination of better sensors and features\, both proprioceptive and exteroceptive\, and self-supervised learning to solve this problem.  We enhance proprioception by exploring the use of new sensing modalities such as sound and vibration\, and in turn we increase the number and variety of terrain types that can be estimated.  We build a supervised proprioceptive multiclass model that can predict up to seven terrain classes.  The proprioceptive predictions are then used as labels to train a self-supervised exteroceptive model from camera data.  The exteroceptive model uses up-to-date vision learning techniques.  This exteroceptive model can then estimate those same terrain types more reliably in new environments.  The exteroceptive semantic terrain predictions can then be spatially registered into a larger map of the surrounding environment.  3d point clouds from rolling/tilting ladar are used to register the proprioceptive and exteroceptive data\, as well as to register the resulting exteroceptive predictions into the larger map.  Our claim is that self-supervised learning makes the exteroception more reliable since it can be automatically retrained for new locations without human supervision.  We conducted experiments to support this claim by collecting data sets from different geographical environments and then comparing classification accuracies.  Our results show that our self-supervised learning approach is able to outperform supervised visual learning techniques. \nSpeaker Bio:\nJacqueline Libby is a PhD candidate in the Robotics Institute.  Her research interests are focused on developing real-world robotics systems.  At her time in the Robotics Institute\, she has worked with the best field roboticists to explore how complex robotic systems can interact with the world in complex outdoor environments.  She hopes in the future to apply what she has learned in the arenas of environmental sustainability or medicine.  Her thesis work (as described in the abstract above) is focused on sensor fusion techniques from a variety of sensing modalities as a way of enhancing robot perception.
URL:https://www.ri.cmu.edu/event/self-supervised-learning-on-mobile-robots-using-acoustics-vibration-and-visual-models-to-build-rich-semantic-terrain-maps-2/
LOCATION:3305 Newell-Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2019/09/libby_jacqueline_cropped.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190910T160000
DTEND;TZID=America/New_York:20190910T170000
DTSTAMP:20260717T173551
CREATED:20190826T150604Z
LAST-MODIFIED:20190915T234659Z
UID:117434-1568131200-1568134800@www.ri.cmu.edu
SUMMARY:AI in Space - From Earth Orbit to Mars and Beyond!
DESCRIPTION:Abstract: \nArtificial Intelligence is playing an increasing role in our everyday lives and the business marketplace. This trend extends to the space sector\, where AI has already shown considerable success and has the potential to revolutionize almost every aspect of space exploration. \nWe first highlight a number of success stories of the tremendous impact of Artificial Intelligence in Space: over a dozen years of operations of the Autonomous Sciencecraft on EO-1\, the Earth Observing Sensorweb tracking volcanoes\, flooding and wildfires and automated targeting onboard the MER and MSL rovers. \nNext we describe how AI-based scheduling is being deployed to NASA’s next rover to Mars\, the M2020 rover. \nFinally we discuss why AI is essential to the search for life beyond Earth\, highlighting the key role of AI in Europa Submersible and Interstellar mission concepts. \nBios: \nDr. Steve Chien is a Senior Research Scientist at the Jet Propulsion Laboratory\, California Institute of Technology where he leads efforts in autonomous systems for space exploration. Dr. Chien has received numerous awards for his research in space autonomous systems including: NASA Medals in 1997\, 2000\, 2007\, and 2015; he is a four time honoree in the NASA Software of the Year competition (1999\, 1999\, 2005\, 2011); and in 2011 he was awarded the inaugural AIAA Intelligent Systems Award. He has led the deployment of ground and flight autonomy software to numerous missions including the Autonomous Sciencecraft/Earth Observing One\, WATCH/Mars Exploration Rovers\, Earth Observing Sensorwebs\, IPEX\, ESA’s Rosetta\, ECOSTRESS and OCO-3 missions and is currently contributing to onboard and ground scheduling for the M2020 rover mission. \n\n\nMs. Jagriti Agrawal is a Member of Technical Staff  in the Artificial Intelligence Group at the Jet Propulsion Laboratory\, California Institute of Technology where she works on automated scheduling for the upcoming M2020 Mars Rover mission.
URL:https://www.ri.cmu.edu/event/ai-in-space-from-earth-orbit-to-mars-and-beyond/
LOCATION:3305 Newell-Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/png:https://www.ri.cmu.edu/app/uploads/2019/08/chien_agrawal.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190521T120000
DTEND;TZID=America/New_York:20190521T130000
DTSTAMP:20260717T173551
CREATED:20190515T193352Z
LAST-MODIFIED:20190515T193352Z
UID:112973-1558440000-1558443600@www.ri.cmu.edu
SUMMARY:From Farm to Takeoff: Ground and Aerial Robots for Biological Systems Analysis
DESCRIPTION:Abstract: \nBiological and agricultural environments are dynamic\, unstructured\, and uncertain\, posing challenges for environmental data collection at the necessary spatial and temporal scales to enable meaningful systems analysis. Small unmanned systems\, however\, can overcome some of these challenges by enabling autonomous or human-assisted image-based and in situ environmental data collection. This talk will present a suite of new assistive technologies that leverage robotics and computer vision to broaden sensing and sense-making across different types of biological and agricultural environments. Demonstrative case studies in each system will be presented\, including high-throughput\, autonomous\, mobile phenotyping of row crops and focused human-robot interaction studies for aerial telemanipulation. The material covered in this talk will illustrate how the strategic\, user-focused design of robotics and automated systems to accomplish unique environmental data collection can enable better understanding and decision-making for dynamic biological systems. \n  \nSpeaker Bio: \nDr. Sierra Young is an Assistant Professor of Biological and Agricultural Engineering at North Carolina State University\, focusing on the use of robotics and automation for sensing and sense-making in agricultural and natural systems\, and human-robot interaction for small unmanned systems. Before arriving at North Carolina State\, she worked as a Visiting Scholar in the Agricultural and Biosystems Engineering Department at Iowa State University. Dr. Young received her Ph.D. in Civil Engineering as a Department of Defense NDSEG Fellow with a focus on the human-robot interaction for physical object manipulation by small unmanned aerial systems from the University of Illinois at Urbana-Champaign in 2018.
URL:https://www.ri.cmu.edu/event/from-farm-to-takeoff-ground-and-aerial-robots-for-biological-systems-analysis/
LOCATION:1305 Newell Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/png:https://www.ri.cmu.edu/app/uploads/2019/05/young.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190517T120000
DTEND;TZID=America/New_York:20190517T130000
DTSTAMP:20260717T173551
CREATED:20190514T171621Z
LAST-MODIFIED:20190514T171621Z
UID:112951-1558094400-1558098000@www.ri.cmu.edu
SUMMARY:Event Cameras: Image Reconstruction\, Convolutions and Color
DESCRIPTION:Abstract: \nEvent cameras are novel\, bio-inspired visual sensors\, whose pixels output asynchronous and independent timestamped spikes at local intensity changes\, called ‘events’. Event cameras offer advantages over conventional frame-based cameras in terms of latency\, high dynamic range (HDR) and temporal resolution. Event cameras do not output conventional image frames\, thus\, image reconstruction from events enables visualisation and frame-based processing. Convolution is a fundamental tool for computer vision\, however\, its application to event cameras is unclear. We show how to compute convolution event-by-event. Until recently\, event cameras have been limited to grayscale intensity. Now\, a new color event camera is available and we unveil the color information contained in events by reconstructing color images\, and release the Color Event Camera Dataset. \n  \nSpeaker Bio: \nCedric completed his Master of Engineering (Mechanical) at the University of Melbourne in 2016. In 2015 he worked as a research assistant in the Fluid Dynamics lab at Melbourne before completing an exchange semester at ETH Zurich. In 2017 Cedric commenced his PhD in Robotic Vision under the supervision of Rob Mahony at the ANU. His PhD topic is the development of novel optical flow algorithms capable of running in real-time for high-speed robotics applications. Cedric is currently pursuing this research using event cameras\, which are bio-inspired vision sensors with microsecond temporal resolution.
URL:https://www.ri.cmu.edu/event/event-cameras-image-reconstruction-convolutions-and-color/
LOCATION:Newell-Simon Hall 4305
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2019/05/scheerlinck.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190405T143000
DTEND;TZID=America/New_York:20190405T153000
DTSTAMP:20260717T173551
CREATED:20190329T030431Z
LAST-MODIFIED:20190329T030431Z
UID:112426-1554474600-1554478200@www.ri.cmu.edu
SUMMARY:Automatic Real-time Anomaly Detection for Autonomous Aerial Vehicles
DESCRIPTION:Abstract: \nThe recent incidents with Boeing 737 Max 8 aircraft have raised concerns about the safety and reliability of autopilots and autonomous operations. There is a growing need for methods to monitor the status of aircraft and report any faults and anomalies to the human pilot or to the autopilot to deal with the emergency situation. We present an online approach using the Recursive Least Squares method to detect anomalies in the behavior of an aircraft. The method models the relationship between correlated input-output pairs and uses the model to detect the anomalies. The result is an easy-to-deploy anomaly detection method that does not assume a specific aircraft model\, does not require any training\, works in real-time and can detect many types of faults and anomalies in a wide range of autonomous aircraft\, including fixed-wing planes\, multirotors and VTOLs. Extensive experimentation in flight tests shows that the method can detect the faults with a high rate and low latency. \n  \nSpeaker Bio: \n\nAzarakhsh is a Ph.D. student in the AIR Lab at the Robotics Institute\, advised by Dr. Sebastian Scherer. His research focuses on developing safe and robust controllers for various autonomous aerial vehicles in real-world applications to accelerate the integration of drones in our everyday lives. Before Carnegie Mellon\, he had received a B.Sc. in Electrical Engineering\, majoring in Electronics Engineering and a B.Sc. in Computer Engineering\, majoring in Software Engineering\, both from Shahid Beheshti University.
URL:https://www.ri.cmu.edu/event/automatic-real-time-anomaly-detection-for-autonomous-aerial-vehicles/
LOCATION:3305 Newell-Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2019/03/keipour.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190403T110000
DTEND;TZID=America/New_York:20190403T120000
DTSTAMP:20260717T173551
CREATED:20190330T212600Z
LAST-MODIFIED:20190330T212748Z
UID:112466-1554289200-1554292800@www.ri.cmu.edu
SUMMARY:Improving Multirotor Trajectory Tracking Performance using Learned Dynamics Models
DESCRIPTION:Abstract: \n\nMultirotors and other aerial vehicles have recently seen a surge in popularity\, partly due to a rise in industrial applications such as inspection\, surveillance\, exploration\, package delivery\, cinematography\, and others. Crucial to multirotors’ successes in these applications\, and enabling their suitability for other applications\, is the ability to accurately track trajectories at high speed and high acceleration. In this talk\, I will show how trajectories can be precisely tracked with a multirotor even in the presence of external disturbances\, such as wind and varying payload\, and modeling errors arising from for example\, poor system identification and calibration\, rotor degradation\, and other unmodeled dynamics. We are able to achieve improved tracking performance by inverting an acceleration error dynamics model learned using incremental regression techniques. Dynamically inverting this model results in vehicle control inputs that are more precise\, improving performance without requiring stiffer feedback gains. Simulation and hardware results will be presented that highlight the benefits of the proposed approach. \n\n\n\n\n  \n\n\n\nBio: \n\nAlex Spitzer is a Ph.D. student in the Resilient Intelligent Systems Lab (RISLab) at the Robotics Institute\, advised by Professor Nathan Michael. His research focuses on combining machine learning techniques with optimization and control theory to improve robot performance. Prior to CMU\, Alex received B.S. degrees in Computer Science and Electrical and Computer Engineering from Cornell University.
URL:https://www.ri.cmu.edu/event/improving-multirotor-trajectory-tracking-performance-using-learned-dynamics-models/
LOCATION:3305 Newell-Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2019/03/spitzer_alexander_2016.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20190327T140000
DTEND;TZID=America/New_York:20190327T150000
DTSTAMP:20260717T173551
CREATED:20190325T173423Z
LAST-MODIFIED:20190325T173926Z
UID:112347-1553695200-1553698800@www.ri.cmu.edu
SUMMARY:Toward intuitive human controlled MAVs: motion primitives based teleoperation
DESCRIPTION:Abstract: \nHumans excel at composing high-level plans that achieve a complex\, multimodal objective; however\, achieving proficiency in teleoperating multi-rotor aerial vehicles (MAVs) in unstructured environments with stability and safety requires significant skill and training. In this talk\, we present human-in-the-loop control of a MAV via teleoperation using motion primitives that addresses these concerns. We show that we can increase naive user proficiency by utilizing known vehicle models. We remove the requirement of maintaining stability and dynamic feasibility from the operator by generating snap-continuous motion primitives at user specified transition points. We further provide safety by incorporating reactive collision avoidance via input space search with locally generated depth maps using onboard depth cameras. Motion primitives based teleoperation is readily extendable to shared control\, e.g. through sampling-based adaptation based on local directional intent prediction over motion primitive libraries. \n  \nSpeaker Bio: \nXuning Yang is a Ph.D. student in the Resilient Intelligent Systems Lab at the Robotics Institute\, advised by Prof. Nathan Michael. Her research focuses on enabling efficient human-in-the-loop control for mobile robots\, including online modelling and prediction of operator intent in order to enable agile safe navigation in unstructured environments. Prior to CMU\, She received her B.A.Sc. in Engineering Science from University of Toronto\, majoring in Aerospace Engineering.
URL:https://www.ri.cmu.edu/event/toward-intuitive-human-controlled-mavs-motion-primitives-based-teleoperation/
LOCATION:GHC 6501
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2019/03/yang_xuning.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181130T140000
DTEND;TZID=America/New_York:20181130T150000
DTSTAMP:20260717T173551
CREATED:20181115T205315Z
LAST-MODIFIED:20181115T205618Z
UID:109939-1543586400-1543590000@www.ri.cmu.edu
SUMMARY:Visual SLAM with Semantic Scene understanding
DESCRIPTION:Abstract: Simultaneous localization and mapping (SLAM) has been widely used in autonomous robots and virtual reality. It estimates the sensor motion and maps the environment at the same time. The classic sparse feature point map of visual SLAM is limited for many advanced tasks including robot navigation and interactions\, which usually require a high-level understanding of 3D object and planes. Most approaches solve SLAM and scene understanding sequentially. \n\nIn this work\, we propose a tightly coupled monocular object and plane SLAM to build a more accurate\, meaningful and dense map. More importantly\, it demonstrates that scene understanding and SLAM can improve each other in one system. To do that\, we first propose an efficient 3D object detection without shape priors and graphical layout inference without box room assumptions. Then we propose a bundle adjustment system to jointly optimize camera poses with objects and planes. They can provide additional geometric\, long-term scale\, and semantic constraints to improve SLAM estimation. Dynamic object movements are also modeled explicitly to achieve 4D mapping and improve the estimation. Experiments on some public TUM\, KITTI and collected datasets show that the proposed algorithm can get the state-of-the-art monocular camera localization accuracy and also improve the 3D object detection.  \n\nSpeaker Bio: Shichao Yang is a Ph.D. student in the Mechanical Engineering at Carnegie Mellon University\, advised by Prof. Sebastian Scherer in the Robotics Institute. He received a B.S in Mechanical Engineering from Shanghai Jiao Tong University in 2013. His research focuses on simultaneous localization and mapping (SLAM) and semantic scene understanding.
URL:https://www.ri.cmu.edu/event/visual-slam-with-semantic-scene-understanding/
LOCATION:3305 Newell-Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/11/shichaoy.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20181128T103000
DTEND;TZID=America/New_York:20181128T113000
DTSTAMP:20260717T173551
CREATED:20181122T064349Z
LAST-MODIFIED:20181122T065054Z
UID:110159-1543401000-1543404600@www.ri.cmu.edu
SUMMARY:Autonomous drone cinematographer: Using artistic principles to create smooth\, safe\, occlusion-free trajectories for aerial filming
DESCRIPTION:Abstract: Autonomous aerial cinematography has the potential to enable automatic capture of aesthetically pleasing videos without requiring human intervention\, empowering individuals with the capability of high-end film studios. Current approaches either only handle off-line trajectory generation\, or offer strategies that reason over short time horizons and simplistic representations for obstacles\, which result in jerky movement and low real-life applicability. In this work we develop a method for aerial filming that is able to trade off shot smoothness\, occlusion\, and cinematography guidelines in a principled manner\, even under noisy actor predictions. We present a novel algorithm for real-time covariant gradient descent that we use to efficiently find the desired trajectories by optimizing a set of cost functions. Experimental results show that our approach creates attractive shots\, avoiding obstacles and occlusion 65 times over 1.25 hours of flight time\, re-planning at 5 Hz with a 10 s time horizon. We robustly film human actors\, cars and bicycles performing different motion among obstacles\, using various shot types.  \nSpeaker Bio: Rogerio Bonatti is a PhD student in the Air Lab at the Robotics Institute advised by Prof. Sebastian Scherer. He is broadly interested in combining motion planning techniques together with learning methods for robust robot decision making in real-life conditions. His current research is focused on autonomous aerial cinematography. Prior to CMU\, Rogerio received his B.S. in Mechatronics Engineering from University of Sao Paulo\, together with a yearlong study-abroad program at Cornell University. 
URL:https://www.ri.cmu.edu/event/autonomous-drone-cinematographer-using-artistic-principles-to-create-smooth-safe-occlusion-free-trajectories-for-aerial-filming/
LOCATION:Gates Hillman Center 4405
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2016/12/bonatti_rogerio_2016.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180802T140000
DTEND;TZID=America/New_York:20180802T150000
DTSTAMP:20260717T173551
CREATED:20180730T160940Z
LAST-MODIFIED:20180730T161425Z
UID:106884-1533218400-1533222000@www.ri.cmu.edu
SUMMARY:Toward Invariant Visual Inertial State Estimation using Information Sparsification
DESCRIPTION:Abstract\nIn this work\, we address two current challenges in real-time visual-inertial odometry (VIO) systems – efficiency and accuracy. To this end\, we present a novel approach to tightly couple visual and inertial measurements in a fixed-lag VIO framework using information sparsification. To bound computational complexity\, fixed-lag smoothers perform marginalization of variables but consequently deteriorate accuracy and especially efficiency. Current state-of-the-art approaches work around this by selectively discarding measurements and marginalizing additional variables. However\, such strategies are sub-optimal from an information-theoretic perspective. In contrast\, our approach formulates an optimization based on Kullback-Leibler divergence to preserve most of the information. To validate our approach\, we conduct extensive real-time drone tests and perform comparisons to current state-of-the-art fixed-lag VIO methods in the EuRoC visual-inertial dataset. The experimental results show that the proposed method achieves competitive and superior accuracy in almost all trials. \n\nIn achieving a more efficient and accurate state estimator\, the second part of the work presents the on-going progress in formulating an optimization-based VIO system using Matrix Lie Groups. Inspired by the recently developed Invariant-EKF framework\, the proposed framework presents better convergence and addresses the consistency problem commonly seen in EKF-based and fixed-lag frameworks. In particular\, we provide detailed derivations of a novel IMU preintegration framework using the group affine properties. Simulation results show our proposed formulation allows the nonlinear optimizer to converge with significantly fewer iterations\, as compared to the state-of-the-art IMU preintegration scheme.\n\n\nSpeaker Bio\nShih-Chieh (Jerry) Hsiung is an M.S. student in the Robot Perception Lab at the Robotics Institute advised by Prof. Michael Kaess. He is broadly interested in the intersection of information theory\, probabilistic theory and nonlinear optimization. His current research is focused on improving optimization methods in visual-inertial state estimation. Prior to CMU\, Jerry received his B.S. in Computer Science from Harvey Mudd College.
URL:https://www.ri.cmu.edu/event/toward-invariant-visual-inertial-state-estimation-using-information-sparsification/
LOCATION:1305 Newell Simon Hall
CATEGORIES:Field Robotics Center Seminar,MSR Thesis Presentation,Seminar,Student Talks
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/07/hsuing_jerry_2016.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180727T150000
DTEND;TZID=America/New_York:20180727T160000
DTSTAMP:20260717T173551
CREATED:20180723T231223Z
LAST-MODIFIED:20180724T130042Z
UID:106687-1532703600-1532707200@www.ri.cmu.edu
SUMMARY:Wire Detection\, Reconstruction\, and Avoidance for Unmanned Aerial Vehicles
DESCRIPTION:Abstract\nThin objects\, such as wires and power lines are one of the most challenging obstacles to detect and avoid for UAVs\, and are a cause of numerous accidents each year. This thesis makes contributions in three areas of this domain: wire segmentation\, reconstruction\, and avoidance. \nPixelwise wire detection can be framed as a binary semantic segmentation task. Due to the lack of a labeled dataset for this task\, we generate one by rendering photorealistic wires from synthetic models and compositing them into frames from publicly available flight videos. We show that dilated convolutional networks trained on this synthetic dataset in conjunction with a few real images perform with higher accuracy and speed on real-world data on a portable GPU as compared to multiple baselines. \nGiven the pixel-wise segmentations\, we develop a method for 3D wire reconstruction. We propose a model-based multi-view algorithm\, which employs a minimal parameterization of wires as catenary curves. We use a bundle adjustment-style framework to recover the model parameters using non-linear least squares optimization. In addition\, we propose a model-free voxel grid method to reconstruct wires via a pose graph of disparity images\, and briefly discuss the pros and cons of each method. \nTo close the sensing-planning loop for wire avoidance\, we demonstrate a reactive\, trajectory library-based planner coupled with our model-free reconstruction method in experiments with a real UAV. Finally\, we propose a novel framework for long-range\, deliberate avoidance. In current state-of-the-art sampling-based planners\, collision checking is often the bottleneck and worst case asymptotic guarantees are the main focus. We propose a method to learn adaptive sampling distributions\, such that the expected search effort is minimized for finite-time problems. Our non-stationary sampling heuristic leverages both the instantaneous search tree and the workspace environment\, and shows significant runtime improvements in simulated experiments. \nSpeaker Bio\nRatnesh Madaan is an M.S. student in the Robotics Institute at Carnegie Mellon University\, advised by Dr. Sebastian Scherer. He is broadly interested in perception and planning algorithms for UAVs. His current research interests lie in thin obstacle detection and mapping\, and at the intersection of machine learning and traditional motion planning algorithms. Prior to CMU\, Ratnesh received a B.Tech. in Mechanical Engineering from Indian Institute of Technology\, Roorkee in 2015.
URL:https://www.ri.cmu.edu/event/wire-detection-reconstruction-and-avoidance-for-unmanned-aerial-vehicles/
LOCATION:1305 Newell Simon Hall
CATEGORIES:Field Robotics Center Seminar,MSR Thesis Presentation,Seminar,Student Talks
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/07/madaan.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180727T133000
DTEND;TZID=America/New_York:20180727T143000
DTSTAMP:20260717T173551
CREATED:20180723T230547Z
LAST-MODIFIED:20180724T125947Z
UID:106682-1532698200-1532701800@www.ri.cmu.edu
SUMMARY:Autonomous 3D Reconstruction in Underwater Unstructured Scenes
DESCRIPTION:Abstract\nReconstruction of marine structures such as pilings underneath piers presents a plethora of interesting challenges. It is one of those tasks better suited to a robot due to harsh underwater environments. Underwater reconstruction typically involves human operators remotely controlling the robot to predetermined way-points based on some prior knowledge of the location and model of the object of interest. However\, it is impractical and dangerous to manually control the robot to perform the reconstruction task in an unstructured scene where prior knowledge of the locations and shapes of objects of interest is not available \n\nBased on the state-of-the-art mapping and planning methods\, this thesis presents an approach that enables the robot to perform reconstruction task in an underter unstructured scene autonomously without any prior knowledge except the bouding box within which the robot should operate. In particular\, the key challenge lies in working with a world representation that is compatible with both mapping and planning algorithms. We address this challenge with the proposed virtual occupancy grid map or VOG-Map. VOG-Map is represented by a collection of local occupancy grid maps whose respective poses are optimized to account for the drift or accumulated noise. \nBased on the VOG-Map\, a path planning algorithm is able to plan safer way-points for collision-free paths as well as more informative way-points for accurate reconstruction than is that based on a global occupancy grid map. By adding additional constraints on the way-points and modifying how their respective information gains are computed\, we show how our mapping algorithm could work well with the way-points returned by the planner based on VOG-Map. \nThe quality of both VOG-Map and scene reconstruction depends on the mapping algorithm. We employ smoothing-based pose graph simultaneous localization and mapping (SLAM) algorithm which is capable of correcting for drift upon loop closures when the algorithm determines that the robot has come back to a previously visited area. However\, in scene that lacks geometric structures\, the process of determining loop closures via methods such as iterative closest point (ICP) is prone to error. We incorporate the approach of determining degeneracy in the scene into our ICP method and add a loop closure constraint to the SLAM optimization problem\, constraining only well-conditioned directions based on principal component analysis. \nWe demonstrate the use of VOG-Map for implementing an underwater system in which the robot actively plans paths to generate accurate 3D scene reconstructions. We evaluate our system qualitatively and quantitatively on simulated as well as real-world experiments. \n\nSpeaker Bio\n \nBing-Jui Ho is a masters student in the Robot Perception Lab at The Robotics Institute\, he is advised by Prof. Michael Kaess. His research is focused on 3D reconstruction in underwater unstructured scenes. Bing received his B.S. in Computer Science from the University of Illinois at Urbana-Champaign. His work has been published at IROS.\n\n  \n 
URL:https://www.ri.cmu.edu/event/autonomous-3d-reconstruction-in-underwater-unstructured-scenes/
LOCATION:GHC 4405
CATEGORIES:Field Robotics Center Seminar,MSR Thesis Presentation,Seminar,Student Talks
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/07/ho.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180726T103000
DTEND;TZID=America/New_York:20180726T113000
DTSTAMP:20260717T173551
CREATED:20180723T230921Z
LAST-MODIFIED:20180724T125848Z
UID:106685-1532601000-1532604600@www.ri.cmu.edu
SUMMARY:Learning Reactive Flight Control Policies: from LIDAR measurements to Actions
DESCRIPTION:Abstract\nThe end goal of a reactive flight control pipeline is to output control commands based on local sensor inputs. Classical state estimation and control algorithms break down this problem by first estimating the robot’s velocity and then computing a roll and pitch command based on that velocity. However\, this approach is not robust in geometrically degenerate environments which do not provide enough information to accurately estimate vehicle velocity. Recent work has shown that learned end-to-end policies can unify obstacle detection and planning systems for vision-based systems. This work applies similar methods to learn an end-to-end control policy for a lidar equipped flying robot which replaces both the state estimator and controller while leaving long term planning to traditional planning algorithms. Specifically\, this work demonstrates the feasibility of training such a policy using imitation learning and RNNs to map directly from lidar range measurements to robot accelerations without an explicit state estimate. The policy is fully trained on simulated data using procedurally generated environments\, achieving an average of over 1.7km mean distance between collisions. Additionally\, various real-world flight tests through tunnel and tunnel-like environments demonstrate that a policy learned in simulation can successfully control a real quadcopter. \n  \nSpeaker Bio\nSam Zeng is an M.S. student in the Robotics Institute at Carnegie Mellon University advised by Dr. Sebastian Scherer. His current research is focused on applying machine learning to reactive control and state estimation for UAVs. Sam received his undergraduate degree from CMU in Mechanical Engineering with an additional major in Robotics in 2016. \n 
URL:https://www.ri.cmu.edu/event/learning-reactive-flight-control-policies-from-lidar-measurements-to-actions/
LOCATION:1305 Newell Simon Hall
CATEGORIES:Field Robotics Center Seminar,MSR Thesis Presentation,Seminar,Student Talks
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/07/zeng-1.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180702T110000
DTEND;TZID=America/New_York:20180702T120000
DTSTAMP:20260717T173551
CREATED:20180627T183734Z
LAST-MODIFIED:20180627T184217Z
UID:106486-1530529200-1530532800@www.ri.cmu.edu
SUMMARY:Learning Deep Multimodal Features for Reliable and Comprehensive Scene Understanding
DESCRIPTION:Abstract\nRobust scene understanding is a critical and essential task for autonomous navigation. This problem is heavily influenced by changing environmental conditions that take place throughout the day and across seasons. In order to learn models that are impervious to these factors\, mechanisms that intelligently fuse features from complementary modalities and spectra have to be integrated. \n\nIn the first part of the talk\, I will first briefly give an overview of our new architecture for semantic segmentation that incorporates our proposed multiscale residual units\, an efficient atrous spatial pyramid pooling module and a deep decoder with skip refinement stages. I will then describe our framework for probabilistically fusing features from multiple modality-specific streams adaptively based on the scene condition. This approach achieves state-of-the performance for multimodal segmentation on several benchmarks including Cityscapes\, Synthia\, SUN-RGBD\, ScanNet and Freiburg Forest. \nIn the second part of the talk\, I will describe our architecture for joint semantic motion segmentation where\, in addition to the object category\, our model also predicts the motion status of each pixel using consecutive monocular images. Our network fuses semantic features with learned motion features while suppressing any induced ego-flow to yield pixel-level semantic motion labels. This work is currently the state-of-the-art for semantic motion segmentation on the Cityscapes and KITTI datasets. \nSpeaker Bio\nAbhinav Valada is a Ph.D. candidate in the Autonomous Intelligent Systems lab at the University of Freiburg in Germany\, where he is advised by Prof. Wolfram Burgard. His research is focused on weakly-supervised/self-supervised deep learning algorithms for multimodal robot perception and state-estimation. His methods are currently the state-of-the-art on several scene understanding and pose estimation benchmarks. Abhinav received his M.S. in Robotics from Carnegie Mellon University. He was previously a research engineer at the Field Robotics Center and the National Robotics Engineering Center. His work has been published at major robotics conferences and journals including ICRA\, IROS\, RSS\, FSR\, ISRR\, ISER and IJRR.
URL:https://www.ri.cmu.edu/event/https-www-ri-cmu-edu-wp-admin/
LOCATION:1305 Newell Simon Hall
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/06/Abhinav_valada.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180511T113000
DTEND;TZID=America/New_York:20180511T123000
DTSTAMP:20260717T173551
CREATED:20180503T205748Z
LAST-MODIFIED:20180503T205748Z
UID:105887-1526038200-1526041800@www.ri.cmu.edu
SUMMARY:Composable Benchmarks for Safe Motion Planning on Roads
DESCRIPTION:Abstract\nNumerical experiments for motion planning of road vehicles require numerous components: vehicle dynamics\, a road network\, static obstacles\, dynamic obstacles and their movement over time\, goal regions\, a cost function\, etc. Providing a description of the numerical experiment precise enough to reproduce it might require several pages of information. Thus\, only key aspects are typically described in scientific publications\, making it impossible to reproduce results—yet\, reproducibility is an important asset of good science. Composable benchmarks for motion planning on roads (CommonRoad) are presented so that numerical experiments are fully defined by a unique ID. Each benchmark is composed by a vehicle model\, a cost function\, and a scenario (including goals and constraints). The scenarios are partly recorded from real traffic and partly hand-crafted to create dangerous situations. \nThe second part of the talk presents techniques to ensure safe planning in CommonRoad benchmarks. Since each traffic situation is unique\, we propose to verify the safety of autonomous vehicles online\, i.e.\, during their operation\, to account for any possible traffic situation. We use reachability analysis to bound possible behaviors of other traffic participants and plan fail-safe trajectories that ensure that reachable sets are avoided when the fail-safe maneuver is activated. This makes it possible to safeguard non-verifiable techniques\, such as machine learning. \nSpeaker Bio\nMatthias Althoff received the diploma in Mechatronics and Information Technology from the department of mechanical engineering at the Technische Universität München\, Germany\, in 2005. He received his PhD degree (summa cum laude) in electrical engineering from the same university under the supervision of Univ.-Prof. Dr.-Ing./Univ. Tokio Martin Buss in 2010. From 2010 – 2012 he was a postdoctoral researcher at Carnegie Mellon University\, USA\, with a joint appointment in electrical engineering and the Robotics Institute. He joined the computer science department at Ilmenau University of Technology\, Germany\, in 2012 as assistant professor for automation systems. Since 2013 Matthias Althoff is assistant professor in computer science at the Technische Universität München. \nHis research interests include the design and analysis of cyber-physical systems\, formal verification of continuous and hybrid systems\, reachability analysis\, planning algorithms\, robust and fault-tolerant control. Main applications of his research are automated vehicles\, robotics\, power systems\, and analog and mixed-signal circuits.
URL:https://www.ri.cmu.edu/event/composable-benchmarks-for-safe-motion-planning-on-roads/
LOCATION:Newell-Simon Hall 1305
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2018/05/Althoff.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20180124T120000
DTEND;TZID=America/New_York:20180124T130000
DTSTAMP:20260717T173551
CREATED:20180118T180624Z
LAST-MODIFIED:20180118T180624Z
UID:103864-1516795200-1516798800@www.ri.cmu.edu
SUMMARY:From Robust Real-time SLAM to Safe Collision Avoidance
DESCRIPTION:Abstract\nState estimation plays a critical role in a robotic system. The problem is to know where the robot is and how it is oriented. This is very often a building block in the navigation system\, which modules in charge of higher level tasks are relied on. Challenges are to carry out state estimation in 6-DOF\, in real-time at high frequencies\, with high precision\, robust to aggressive motion and environmental changes. The talk will start with state estimation leveraging range\, vision\, and inertial sensing. Then\, it will discuss more recent work regarding autonomous navigation of lightweight UAVs in cluttered environments\, avoiding obstacles at high speeds. The talk will finish with the latest results and take aways. \nSpeaker Bio\nJi Zhang is postdoctoral fellow at the Robotics Institute of CMU. He received his PhD degree in Feb. 2017. His PhD research focused on ego-motion estimation and mapping. His methods are ranked #1 and #2 on the odometry leaderboard of the internationally well-known KITTI Vision Benchmark\, and won the Microsoft Indoor Localization Competition in 2016 and 2017. His recent work inclined toward collision avoidance of aerial vehicles. Ji Zhang is founder and Chief Scientist of Kaarta\, a CMU spin-off commercializing 3D lidar mapping and 3D modeling technologies as the outcome of his research work.
URL:https://www.ri.cmu.edu/event/robust-real-time-slam-safe-collision-avoidance/
LOCATION:Newell Simon Hall 1507
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2016/12/zhang_ji_2014_3.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171205T103000
DTEND;TZID=America/New_York:20171205T113000
DTSTAMP:20260717T173551
CREATED:20171204T185713Z
LAST-MODIFIED:20171204T185713Z
UID:102903-1512469800-1512473400@www.ri.cmu.edu
SUMMARY:Belief Space Planning for Reducing Terrain Relative Localization Uncertainty in Noisy Elevation Maps
DESCRIPTION:Abstract \nAccurate global localization is essential for planetary rovers to reach science goals and mitigate mission risk. Planetary robots cannot currently use GPS or infrastructure for navigating\, and hence rely on terrain for determining global position. Terrain relative navigation (TRN) compares planetary rover-perspective images and 3D models to existing satellite orbital imagery and digital elevation models (DEMs) for absolute positioning. However\, TRN is limited by the quality of orbital data and the presence and uniqueness of terrain features. This talk presents a method that reduces localization uncertainty from terrain relative navigation while planning global paths and is robust to noise and errors in DEMs. \nSpeaker Bio\nEugene Fang is a Ph.D. student in the Robotics Institute at Carnegie Mellon University\, advised by Prof. William “Red” Whittaker. He previously received his M.S. in Robotics at CMU and B.S. in Electrical Engineering and Computer Sciences at UC Berkeley. His interests are in route planning for planetary exploration rovers.
URL:https://www.ri.cmu.edu/event/belief-space-planning-for-reducing-terrain-relative-localization-uncertainty-in-noisy-elevation-maps/
LOCATION:GHC 4405
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2016/12/fang_eugene_2016.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171128T120000
DTEND;TZID=America/New_York:20171128T130000
DTSTAMP:20260717T173551
CREATED:20171127T190051Z
LAST-MODIFIED:20171127T190051Z
UID:102812-1511870400-1511874000@www.ri.cmu.edu
SUMMARY:Planning Algorithms for Multi-Robot Active Perception
DESCRIPTION:Abstract\nA fundamental task of robotic systems is to use on-board sensors and perception algorithms to understand high-level semantic properties of an environment. The performance of perception algorithms can be greatly improved by planning the motion of the robots to obtain high-value observations. \nIn this talk I will present a suite of planning algorithms we have been developing for these tasks. These methods aim to address the challenges of decentralised coordination\, long planning horizons\, unreliable communication\, predicting plans of other agents\, and exploiting characteristics of perception models. The proposed algorithms are motivated by several key ideas: Monte Carlo tree search\, self-organising maps\, branch and bound\, optimal stopping\, sweep planes\, variational methods\, orienteering problems\, set cover\, and Bayesian inference. \nSpeaker Bio\nGraeme Best is a PhD candidate at the Australian Centre for Field Robotics (ACFR) at The University of Sydney under the supervision of A/Prof. Robert Fitch. His current research interests include planning algorithms for multi-robot teams performing coordinated perception tasks\, with a particular emphasis on decentralised algorithms and probabilistic reasoning. Previously\, he worked on projects involving machine learning for legged robots\, marine robotics operations\, and human-robot interaction. He received the B.E. (Electrical) and B.Sc. (Computer Science) from Monash University in 2014
URL:https://www.ri.cmu.edu/event/planning-algorithms-for-multi-robot-active-perception/
LOCATION:Newell Simon Hall 1507
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2017/11/best.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20171121T153000
DTEND;TZID=America/New_York:20171121T163000
DTSTAMP:20260717T173551
CREATED:20171116T024719Z
LAST-MODIFIED:20171116T024719Z
UID:102675-1511278200-1511281800@www.ri.cmu.edu
SUMMARY:Dense Planar-Inertial SLAM for Large Indoor 3D Reconstruction
DESCRIPTION:Abstract\nReconstructing the dense 3D models of indoor environments in real-time is key to many robotics applications\, such as navigation\, inspection\, and augmented reality. It is also a challenging problem due to the accumulation of drift\, large amount of data\, limited computation\, and occasional lack of visual features. We develop an RGB-D simultaneous localization and mapping (SLAM) system that takes the commonly observed indoor planar structures (e.g.: walls and floors) as landmarks\, and outputs dense 3D point cloud models with refined maps of the landmark planes. Incorporating planes in the SLAM system reduces the accumulated drift\, and allows a more efficient global optimization. Then\, an inertial sensor is added into the system to further correct the drift and increase the robustness. With additional structural constraints between planes (e.g.: orthogonality and parallelism) and loop closure functions integrated with the dense planar-inertial SLAM system\, our final solution can reconstruct accurate dense 3D models of large indoor environments in real-time on CPU only. We demonstrate the state-of-the-art performance of our solution by comparing it with other existing methods on public datasets and evaluating it using a ground truth model from a survey lidar. \nSpeaker Bio\nMing Hsiao is a Ph.D. student in the Robotics Institute at Carnegie Mellon University\, advised by Prof. Michael Kaess. He previously received his M.S. in Electrical Engineering at National Taiwan University. His interests are in SLAM\, 3D perception\, and optimization.
URL:https://www.ri.cmu.edu/event/dense-planar-inertial-slam-large-indoor-3d-reconstruction/
LOCATION:Newell Simon Hall 1507
CATEGORIES:Field Robotics Center Seminar,Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2016/12/hsiao_ming_2014_sm.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20170922T123000
DTEND;TZID=America/New_York:20170922T133000
DTSTAMP:20260717T173551
CREATED:20170921T151441Z
LAST-MODIFIED:20170926T191232Z
UID:27829-1506083400-1506087000@www.ri.cmu.edu
SUMMARY:Fusion of Cameras and Sparse Ranging Measurements in Multi‐agent SLAM
DESCRIPTION:Abstract \nCameras are widely used for localization and navigation in GNSS‐denied environments. By exploiting\nVSLAM (Visual Simultaneous Localization and Mapping) techniques\, vehicles equipped with cameras are capable of estimating their own trajectories and simultaneously building a map of the surrounding environment. In many applications\, multiple cooperative robotic agents (robotic swarms) are used in order to improve the robustness to malfunctions\, exploration or mapping efficiency\, and localization accuracy. \nThis talk shows that by using sparse ranging measurements between a pair of dynamic rovers and onboard cameras\, the relative pose between the two agents can be estimated. If monocular cameras are used\, the global scale factors can be jointly obtained. The algorithm can also be applied in a single‐rover/single‐base‐station scenario\, but the polar angle of the rover with respect to the base station remains ambiguous. In addition\, it is shown that the error accumulation problem in VSLAM is mitigated by including ranging measurements\, without requiring the rovers to revisit mapped locations for loop closures. The localization accuracy improvement is analyzed using Cramér–Rao lower bounds. \n  \nBio \nChen Zhu is currently pursuing his Ph.D. at Technische Universität München (Technical University of Munich) in Munich\, Germany as a full‐time researcher at the Institute for Communications and Navigation. He received his B.Sc. in Automation Engineering from Tsinghua University\, in Beijing\, China in 2009\, and his M.Sc. in Communications Engineering in 2011 from Technische Universität München. In 2012\, he has won the IEEE Region 8 (Europe\, Middle East and Africa) student paper contest (2nd place) with the paper “High accuracy multi‐link synchronization in LTE: Applications in localization”. \nSince 2012\, Chen Zhu works on the DLR (German Aerospace Center) project “Valles Marineris Explorer” (VaMEx). His research interests include visual navigation\, robotic swarm navigation\, and multiple sensor fusion in autonomous vehicle navigation. \n 
URL:https://www.ri.cmu.edu/event/fusion-of-cameras-and-sparse-ranging-measurements-in-multi%e2%80%90agent-slam/
LOCATION:NSH 1507
CATEGORIES:Field Robotics Center Seminar
ATTACH;FMTTYPE=image/jpeg:https://www.ri.cmu.edu/app/uploads/2017/09/zhu.jpg
END:VEVENT
END:VCALENDAR