Samarth Manoj Brahmbhatt

76 Pages·2016·3.48 MB·English

Samarth Manoj Brahmbhatt
School Address Permanent Address
Institute for Robotics and Intelligent Machines 470 16th St NW, Apt. 2017
Georgia Institute of Technology Atlanta GA 30363
801 Atlantic Drive [phone]
Atlanta GA 30332
Webpage: www.cc.gatech.edu/~sbrahmbh E-mail: [email protected]
EDUCATION
Doctor of Philosophy, Robotics
School of Interactive Computing, Georgia Institute of Technology, GA expected May 2019
Master of Science in Engineering, Robotics
University of Pennsylvania, PA. May 2014
Bachelor of Technology, Electronics & Communication Engineering
Nirma University, Ahmedabad, India. May 2012
PUBLICATIONS
• “Occlusion-AwareObjectLocalization,SegmentationandPoseEstimation”-SamarthBrahmbhatt,
Heni Ben Amor and Henrik Christensen (British Machine Vision Conference 2015)
• “SingleImage3DObjectDetectionandPoseEstimationforGrasping”-MenglongZhu, KostaDerpa-
nis, Yinfei Yang, Samarth Brahmbhatt, Mabel Zhang, Cody Phillips and Kostas Daniilidis (IEEE
International Conference on Robotics and Automation 2014)
• “RoboCup 2013 Humanoid Kidsize League Winner” - Daniel D. Lee, Seung-Joon Yi, Stephen McGill,
Yida Zhang, Larry Vadakedathu, Samarth Brahmbhatt, Richa Agrawal and Vibhavari Dasagi
(RoboCup 2013: Robot World Cup XVII, Springer Berlin Heidelberg 2014)
• “Practical OpenCV” - Samarth Brahmbhatt (book published by Apress Media LLC)
RESEARCH EXPERIENCE
Dextro, Inc. New York City Summer 2015
Improving CNN-based object localization using local context
• ImplementedaConvolutionalNeuralNetworksystemforobjectlocalizationandsemanticsegmentation
• Used semantic segmentation as a local context signal to improve object detection
• Implemented CPU and GPU versions of various required layers in Caffe
GRASP Laboratory, University of Pennsylvania Spring 2014
Detecting Partially Occluded Objects in Images (Masters’ Thesis)
• Augmented the DPM object detection algorithm to detect up to 60% occluded objects
• UsedHOGfeaturesandgraph-cutstosegmentallpixelsinsidetheboundingboxtoobject/non-object
• Used Structural SVM to train HOG feature and graph edge weights
GRASP Laboratory, University of Pennsylvania Fall 2013
Active Deformable Part models inference implementation
WroteMEXimplementationoftheA-DPMobject-detectionalgorithminferencepart. Thisalgorithmtreats
part inference order in DPM as a scheduling problem and achieves up to 3x speedup over Cascade-DPM.
GRASP Laboratory, University of Pennsylvania Summer 2013
Detection and 6-DOF pose estimation of objects from a single 2D image
System to detect objects using their shape and estimate their 6-DOF pose by matching the outline with a
pre-computed 3D model. Works in heavily cluttered scenes. Contributed to:
S. Brahmbhatt
p. 2
• Dynamic programming based object outline matching for pose estimation
• Motion-field based algorithm for iteratively deciding the pose of the 3D model in space to match its
silhouette with outline of detected object
• Putting the silhouette extraction, detection and pose-estimation modules together into an efficient
pipeline executable on a Willow Garage PR2 robot
GRASP Laboratory, University of Pennsylvania Spring 2013
Robocup 2013 Humanoid Kid-size soccer international competition
Our team won the Kid-size competition after competing against international teams. Contributed to:
• Particlefilterbasedlocalizationsystemthatusedgoalpostsandfieldlinesaslandmarksandodometry
information from the walk engine
• Player self-localization orientation disambiguation based on goalkeeper ball estimate
Schneider India Innovation Challenge 2011 August 2011
Fuel saving at traffic signals
• Designed and prototyped a system that used accelerometers, magnetometers and wireless communica-
tion to automatically switch off engines of cars opposite red signals at traffic intersections
TEACHING EXPERIENCE
Teaching Assistant for
• Introduction to Robotics (MEAM 520): Responsible for weekly office hours and conducting a class
project on mobile robots.
• DesignofMechatronicsystems(MEAM510): Responsibleforweeklylabhoursandconductinganewly
added project in which students make an autonomous golfer robot.
ACADEMIC PROJECTS
Learning in Robotics (ESE 650) Spring 2013
• RGB-D point-cloud registration for 3D mapping (wiki)
• Planar Simultaneous Localization and Mapping using a particle filter (wiki)
• Image panoramas using 3-DOF orientation tracking by an Unscented Kalman Filter (wiki)
• Path planning in aerial photographs using imitation learning (wiki)
• Probabilistic color image segmentation using Gaussian Mixture Models (wiki)
Computer Vision and Computational Photography (CIS 581) Fall 2013
• Logo replacement using Shape Context feature matching (wiki)
• Panoramas by Corner appearance feature matching (wiki)
• Image Morphing by Thin Plate Splines (wiki)
Machine Perception (CIS 580) Spring 2013
• Image stitching using vanishing points and matching points (wiki)
• Logo warping using perspective transforms (wiki)
Intro to Parallel Programming (Udacity Online Course) Summer 2014
• Tone mapping using histogram equalization
• Poisson blending of masked images
COMPUTER SKILLS
• Programming Languages: C, C++, Matlab, Python
• Libraries and Tools: OpenCV, CUDA, Caffe, Vim, Git, LATEX
• Operating Systems: Microsoft Windows, Linux flavors, ROS
S. Brahmbhatt
p. 3
HONORS
• Best overall student in the Electronics and Communication department, Nirma University
• Second prize for final year project, Electronics and Communication department, Nirma University
• Dhirubhai Ambani Scholarship for all four years of undergraduate study
References available upon request.
Daniel Castro [email protected]
(404) - 9166421
Computer Vision Researcher
www.dcastro.me
Education
Georgia Institute of Technology, Atlanta, GA
​
Pursuing Doctorate of Philosophy in Computer Science August 2013 - Present
​
Bachelor of Science in Computer Science (GPA 3.35 - High Honors) August 2009 - May 2013
Research
Video Segmentation Project January 2013 - Present
​
Developed an online system to make our video segmentation algorithm more accessible to other researchers by building a platform on Google
Cloud Services (uses App Engine, Datastore and Compute Engine) to enable researchers to upload and segment their videos online.
Available at www.videosegmentation.com. This work was presented at a CVPR 2014 Panel.
​ ​

Video Annotation Project August 2014 - Present
​
This project is an ongoing effort to develop pixel-level annotations for a large dataset of videos. We have approached this problem from
numerous angles, one of which was building a test set from public Flickr videos for users to annotate (available at www.videoannotation.com).
​ ​
We are currently building a larger video dataset for the research community that we hope will enhance future endeavors in object detection.

Egocentric Daily Activities Project January 2015 - Present
​
This endeavour focuses on the use of egocentric images for activity recognition. Our first approach involved the collection of just over 40,000
egocentric images, each with an activity label. These images were then analyzed through machine learning approaches, specifically a
convolutional neural network combined with a random decision forest. See publication below for more details.

Egocentric Vacation Project August 2014 - Present
​
The egocentric vacation project focused on detecting picturesque moments in long video streams by leveraging context for the data. We
processed over 20 hours of video recorded over a period of two weeks across the United States by a single participant and generated video
highlights for the participants vacation. One of the key contributions of this project was the introduction of a clever technique to detect well
aligned egocentric images without the use of traditional horizon detection techniques. See publication below for more details.
Industry Work
Google, Inc - Software Engineering Internships
​
Google Research - Atlanta, GA (part-time) August 2015 - Present
​ ​
Google Research - Mountain View, CA (full-time) May 2015 - August 2015
​​
Working with Dr. Grundmann and Dr. Raveendran under Google Research. The details of this work are to be released.

Google Research - Atlanta, GA (part-time) August 2014 - December 2014
​
Google Research - Mountain View, CA (full-time) May 2014 - August 2014
​​
Worked with Dr. Matthias Grundmann and Dr. Vivek Kwatra on determining video saliency and developing the framework for video segmentation.

Google Images - Mountain View, CA (full-time) May 2012 - August 2012
​
Worked on the Google Images team with David Chen - we assessed how to visualize stereoscopic imagery on the web.

Google Platforms - Mountain View, CA (full-time) May 2011 - August 2011
​
Teaching Experience
Georgia Tech Online Masters Graduate Teaching Assistant January 2014 - Present
​
Redesigned the majority of the assignments for the online offering of Computational Photography with Dr. Irfan Essa.
Georgia Tech College of Computing Teaching Assistant January 2013 - May 2013
On campus teaching assistant for CS 6476 - Computational Journalism with Dr. Irfan Essa.
Georgia Tech College of Computing Study Abroad Teaching Assistant - Barcelona, Spain May 2013 - August 2013
​
Teaching Assistant for Computational Photography and Computational Journalism in addition to helping manage and coordinate the Georgia
Tech Study Abroad program in Barcelona Spain (~55 students traveling abroad).
Publications
Discovering Picturesque Highlights from Egocentric Vacation Videos - Winter Applications of Computer Vision March 2016
​​
Vinay Bettadapura, Daniel Castro, Irfan Essa - http://goo.gl/RvgW6v
​ ​ ​
Predicting Daily Activities from Egocentric Images using Deep Learning - International Symposium on Wearable Computing September 2015
​​
Daniel Castro, Steven Hickson, Vinay Bettadapura, Edison Thomaz, Gregory Abowd, Henrik Christensen, Irfan Essa - http://goo.gl/y6ZYDP
​ ​
Calibration Free Rolling Shutter Removal - International Conference on Computational Photography - Best Paper April 2012
​​ ​
Matthias Grundmann, Vivek Kwatra, Daniel Castro, Irfan Essa - http://research.google.com/pubs/pub37744.html
​ ​ ​
Awards
Tower Awards Recipient 2013
Awarded for Undergraduate Academic Excellence for underrepresented students. 2012
2011
Outstanding Undergraduate Research Award - College of Computing, Georgia Tech 2012
​
Best Paper Award - International Conference on Computational Photography 2012
​​
Activities / Skills
Panamanian Students @ Georgia Tech (PANAS) - Technology Chair August 2015 - Present
​ ​
Helped with any technology-related issues for the organization.
Member of the Georgia Tech Skydiving Club September 2009 - May 2010
13 jumps, finished AFF certification (*no longer certified due to inactivity).
Extensive Travelling Experience - 28 Countries
​
More than comfortable in an international community and with traveling locally / abroad as part of the job.
​ ​
Speak Spanish and English natively and have some experience in French.
​ ​​ ​ ​ ​
Computational Skills
Programming Languages - Python, C++, Javascript, Java.
​
Web Development Frameworks - Django, webapp2, Google Cloud Services (Compute Engine, App Engine, Cloud Datastore, Cloud Storage).
​
Deep Learning Frameworks - I work mainly with caffe, but have dabbled with tensorflow, torch, and theano.
​
Dongsik Chang
6302DenmeadeDrNE–Atlanta,Georgia30345
[phone] • [email protected] • users.ece.gatech.edu/~dchang33
RESEARCH INTERESTS
Robotics,mobilesensornetworks,flowfieldmodeling,environmentalsensingandmodeling,pathplanning,
optimalcontrol,andstochasticsystems
EDUCATION
GeorgiaInstituteofTechnology Atlanta,Georgia
Ph.D.(candidate)inElectricalandComputerEngineering Aug. 2016(expected)
GeorgiaInstituteofTechnology Atlanta,Georgia
M.S.inElectricalandComputerEngineering Dec. 2010
HanyangUniversity Seoul,Korea
B.S.inElectricalandComputerEngineering Feb. 2007
HONORS, AWARDS, & CERTIFICATES
• Best Paper Award Finalist at the 12th International Symposium on Distributed 2014
AutonomousRoboticSystems(DARS2014)
• Certificate in Embedded Systems Engineering from University of California at 2005
Irvine,Irvine,California
• KoreanGovernmentScholarship,supportingavisitingstudentattheUniversity 2005
ofCaliforniaatIrvine
PUBLICATIONS & PRESENTATIONS
JournalPapers................................................................................................................
[1] DongsikChang,WencenWu,CatherineR.Edwards,andFuminZhang,“MotionTomography: Map-
pingFlowFieldsUsingAutonomousUnderwaterVehicles,”InternationalJournalofRoboticsResearch,
2015(underreview).
[2] DongsikChang,CatherineR.Edwards,andFuminZhang,“Real-TimeGuidanceofUnderwaterGliders
AssistedbyPredictiveOceanModels,”JournalofAtmosphericandOceanicTechnology,32(3): 562-578,2015.
BookChapters................................................................................................................
[1] DongsikChang,WencenWu,andFuminZhang,“GliderCT:AnalysisandExperimentalValidation,”in
DistributedAutonomousRoboticSystems,ser. SpringerTractsinAdvancedRobotics. SpringerJapan,Volume
112,285-298,2016.
[2] Dongsik Chang, Xiaolin Liang, Wencen Wu, Catherine R. Edwards, and Fumin Zhang, “Real-Time
ModelingofOceanCurrentsforNavigatingUnderwaterGliderSensingNetworks,”inCooperativeRobots
andSensorNetworks,ser. StudiesinComputationalIntelligence. SpringerBerlinHeidelberg,Volume507,
61-75,2014.
ConferencePapers..........................................................................................................
[1] DongsikChangandFuminZhang,“DistributedOptimizationforMotionTomography,”submittedto
the55thIEEEConferenceonDecisionandControl(CDC2016),2016(underreview).
[2] Dongsik Chang and Fumin Zhang, “Resolving Temporal Variations in Data-Driven Flow Models
ConstructedbyMotionTomography,”submittedto10thIFACSymposiumonNonlinearControlSystems
(NOLCOS2016),2016(underreview).
[3] DongsikChangandFuminZhang,“DistributedMotionTomographyforTime-VaryingFlowFields,”
submittedtoOCEANS’16MTS/IEEEShanghai,2016(accepted).
[4] DongsikChang,WencenWu,andFuminZhang,“GliderCT:AnalysisandExperimentalValidation,”
inProceedingsofthe12thInternationalSymposiumonDistributedAutonomousRoboticSystems(DARS2014),
196-209,2014. BestPaperAwardFinalist.
[5] WencenWu,DongsikChang,andFuminZhang,“GliderCT:ReconstructingFlowFieldsFromPredicted
MotionofUnderwaterGliders,”inProceedingsoftheEighthACMInternationalConferenceonUnderwater
Networks&Systems(WUWNet2013),Article47,2013.
[6] WencenWu,DongsikChang,andFuminZhang,“ABio-InspiredRobust3DPlumeTrackingStrategy
UsingMobileSensorNetworks,”inProceedingsofthe52ndIEEEConferenceonDecisionandControl(CDC
2013),4571-4578,2013.
[7] DongsikChang,WencenWu,DonaldR.Webster,MarcJ.Weissburg,andFuminZhang,“ABio-Inspired
PlumeTrackingAlgorithmforMobileSensingSwarmsinTurbulentFlow,”inProceedingsof2013IEEE
InternationalConferenceonRoboticsandAutomation(ICRA2013),921-926,2013.
[8] XiaolinLiang,WencenWu,DongsikChang,andFuminZhang,“Real-TimeModelingofTidalCurrent
forNavigatingUnderwaterGliderSensingNetworks,” ProcediaComputerScience,10:1121-1126,2012.
[9] DongsikChang,FuminZhang,andMichaelWest,“DiagnosisandPrognosisofScrubberFaultsfor
UnderwaterRebreathersBasedonStochasticEventModels,”inProceedingsof2011IEEEInternational
ConferenceonPrognosticsandHealthManagement(PHM2011),2011.
Presentations.................................................................................................................
[1] Dongsik Chang, Catherine R. Edwards, and Fumin Zhang, “Flow Mapping Based on the Motion-
IntegrationErrorsofAutonomousUnderwaterVehicles,”2016OceanSciencesMeeting(OSM2016),New
Orleans,LA,Feb. 21-26,2016.
[2] DongsikChangandFuminZhang,“MotionTomographyPerformedbyMobileSensorNetworks,”2015
DecisionandControlLaboratoryStudentSymposium,GeorgiaTech,Atlanta,GA,Apr. 24,2015.
[3] Dongsik Chang, Catherine R. Edwards, and Fumin Zhang, “Real-Time Guidance of Autonomous
UnderwaterVehicleNavigationUsingPredictiveOceanModels,”2013SIAMConferenceonControland
ItsApplications(CT13),SanDiego,CA,July8-10,2013.
[4] DongsikChang,KlementynaSzwaykowska,CatherineR.Edwards,andFuminZhang,“TheGlider
CoordinatedControlSystemforStudyofPersistentWintertimePhytoplanktonBloomsNearLongBay,
SC,”2012OceanSciencesMeeting(OSM2012),SaltLakeCity,UT,Feb. 20-24,2012.
ACADEMIC EXPERIENCE
GeorgiaInstituteofTechnology Atlanta,Georgia
GraduateResearchAssistantattheGeorgiaTechSystemsResearch(GTSR)Lab 2009–Present
• Ph.D.research: Todevelopefficientalgorithmsforreal-timehigh-resolutionenvi- 2011–Present
ronmentalflowfieldmodelingandestimationusingmobilesensornetworks
• Study of environmental flow field mapping based on trajectory information of 2013–Present
underwatermobilesensornetworks
- Developedamethodreferredtoasmotiontomographyforcreatingamapofa
flowfieldbasedontrajectoryinformationofmobilesensingagents
- Developednon-parametricandparametricflowfieldestimationalgorithms
- Validatedmotiontomographythroughsimulationsandexperiments
- Developingadistributedflowfieldestimationalgorithm
• FieldexperimentforanoilspillsurveyintheGulfofMexico(EcosystemImpactsof June–July2015
Oil&GasInputstotheGulf[ECOGIG])usingunderwatergliders,incollaboration
withtheSkidawayInstituteofOceanography
- Contributedtoglideroperationanddatacollection
• Study of acoustic signal detection by autonomous underwater vehicles within Sept.–Oct. 2014
an array of acoustic transmitters at Gray’s Reef National Marine Sanctuary, in
collaborationwiththeSkidawayInstituteofOceanography
- Contributedtoglideroperationanddatacollection
• Fielddemonstrationofoceanobservingtechnologies(Gliderpalooza),incollabo- Sept. 2013
rationwiththeSkidawayInstituteofOceanography
- Led the GT-SkIO glider team for glider operation to support an ensemble of
oceanmodelsbyprovidingoceansamplescollectedfromgliders
• Bio-inspiredplumetracking 2013
- Modeledstochasticeventsofplumedetectioninspiredbybluecrabs
- Contributedtodevelopingtwo-/three-dimensionalplumetrackingalgorithms
basedontheproposedstochasticplumedetectioneventmodel
• Oceancurrentmodelingfornavigatingunderwatergliders 2013
- Developedaparametricmodelforoceancurrentsusingbasisfunctions
- ValidatedtheproposedmodelusingHF-radardataandADCIRCoceanmodel
output,anddemonstratedthemodelthroughguidingsimulatedgliders
• StudyofphytoplanktonbloomformationoffLongBay,SouthCalifornia,incollab- 2011–2013
orationwiththeSkidawayInstituteofOceanography(SupportedbyNSF,PIs: Drs.
JimNelson,CatherineR.Edwards,andHarveySeim)
- LedtheGT-SkIOgliderteamintwofieldexperimentsin2012and2013
- Developedalocaloceanpredictionalgorithmforglidernavigationbasedon
flowestimatesfromglidersandtheadvancedcirculation(ADCIRC)model
- DevelopedtheGlider-EnvironmentNetworkedInformationSystem(GENIoS),
aMATLAB-basedsoftwarepackageforautonomousunderwaterglidercontrol
- Builtareal-timeinterfacetoaccessdatafromoceanmodelssuchasADCIRC,
HYCOM,andSABGOM,andfromHF-radaroceanobservations
- Analyzed ocean currents using the ADCIRC predictive ocean model, glider
on-boardflowestimates,andWERAHF-radardata
- Contributedtogliderdeployment/operationanddatacollection
• Diagnosisandprognosisofscrubberfaultsforunderwaterrebreathers 2009–2010
- Modeledeventsofcarbondioxide(CO )absorptionfailureofCO scrubbers
2 2
forunderwaterrebreathersusingstochasticprocesses
- DevelopedalgorithmsfordiagnosisandprognosisofCO scrubberfaultsbased
2
onstochasticeventmodels.
HanyangUniversity Seoul,Korea
• Seniorproject 2006
- DesignedandimplementedBluetooth-basedreal-timewirelessnetworkingMP3
players
• Summerstudentinternattheembeddedcontrolsystemslab June–Aug. 2006
- Objective: ToimplementaMATLAB-basedcontrolleralgorithmonXC164
UniversityofCaliforniaatIrvine Irvine,California
• Courseprojectsinanembeddedsystemsengineeringcertificateprogram Feb.–July2005
- ImplementedaprocessoremulatoronFPGAusinganAlteraUP-2designlab
package
- ImplementedanembeddedschedulingsystemusingAtmega128
PROFESSIONAL EXPERIENCE
MontereyBayAquariumResearchInstitute MossLanding,California
SummerIntern June–Aug. 2012
• Mentors: Drs. JamesBellinghamandSergeyFrolov
• Project description: To design and implement a flow prediction algorithm for
autonomousunderwatervehiclenavigationbasedonanempiricalmodelderived
fromhistoricHFradardata
TEACHING EXPERIENCE
• Graduate mentor for the vertically integrated projects (VIP) program, Georgia Tech’s undergraduate
researchexperienceprogram,inFall2015andSpring2016
• GraduatementorforMeadowcreekHighSchoolstudentsforasummerresearchexperienceprogramin
Summer2015
• TeachingassistantforStochasticSystems(ECE6558)inFall2014
ACADEMIC SERVICE
• Reviewer for the Robotics and Automation Letters (RA-L), the Journal of Atmospheric and Oceanic
Technology(JTECH),theCyber-PhysicalSystems(CPS),theConferenceonDecisionandControl(CDC),
the American Control Conference (ACC), the International Conference on Robotics and Automation
(ICRA),andtheInternationalConferenceonIntelligentRobotsandSystems(IROS)
• Reviewerforthepresident’sundergraduateresearchawards(PURA)programatGeorgiaTechin2015
• Researchpresentationatthebriefingandexhibition“RobotsforGood”heldinWashington,D.C.in2014
inconjunctionwiththeCongressionalRoboticsCaucus
• Demonstrationofunderwaterrobotsatthe2012SavannahOceanExchangeannualevent
• StudentvolunteerfortheConferenceonDecisionandControl2010(CDC2010),Atlanta,Georgia
EXTRACURRICULAR ACTIVITIES
• GkMMSummerSchool2010on“Cooperativemonitoringinrobotandsensornetworks,”VirginiaTech,
Blacksburg,Virginia
• Cyber-PhysicalSystemsSummerSchool2010,GeorgiaTech,Atlanta,Georgia
OTHER EXPERIENCE
RepublicofKoreaAirForce Osan-si,Gyeonggi-do,Korea
MilitarySoldier,ElectricalandCommunicationsDepartment May2000–Nov. 2002
• Ledacompanyof150soldiers;dischargedatarankofAirmanFirstClass
SKILLS
• Languages: English,Korean
• OperatingSystems: Windows,Linux,MacOS
• ProgrammingLanguages: C,C++,MATLAB,VHDL,Verilog,Lisp
• ComputerTools: ArmDeveloperSuite,QuartusII,MAXPlus2,WinAVR,VisualStudio,PSpice
• RoboticPlatforms: Slocumglider,KheperaIII
Vivian Chu
[email protected]
www.cc.gatech.edu/~vchu7

Education: Georgia Institute of Technology August 2013 - Present
Ph.D. in Robotics

University of Pennsylvania May 2013
M.S. in Robotics

University of California, Berkeley May 2009
B.S. in Electrical and Computer Engineering (EECS)

Work Experience: Socially Intelligent Machines Lab, RIM @ GT, Georgia Tech, Atlanta, GA
Graduate Research Assistant, August 2013 - Present
- Developed system to learn and predict affordances in objects using primitive action manipulation and
multimodal (haptic + vision) data: published (2), under review (1)
- Developed real-time multimodal (vision + audio) system to detect robot engagement: published (1)
- Architected new ROS software structure, maintenance, and integrating ROS and MEKA robot in simulation

Google[x], Mountain View, CA
User Experience Researcher Intern, May 2015 – August 2015
- Worked on Project Wing (UAV delivery project) where I conducted several user studies
- Prototyped delivery mechanisms and collected and analyze data for package delivery specifications

Honda Research Institute, Mountain View, CA
Software Research Intern, June 2013 - August 2013
- Developed human-car interfaces for eyes-free search queries and conducted user studies on system safety

Haptics Lab, GRASP Lab, University of Pennsylvania, Philadelphia, PA
Research Assistant, January 2012 - May 2013
- Co-lead team for DARPA-funded BOLT-E robotic language grounding program
- Collaborated in integrating Syntouch BioTacs, a new touch sensor, onto Willow Garage’s PR2
- Wrote real-time controller for PR2 in ROS to perform exploratory procedures on objects
- Published award-winning paper at ICRA 2013 based on the DARPA funded research with UC Berkeley

International Business Machine – Almaden Research Center, San Jose, CA
Software Engineer, August 2009 - August 2011
- Published paper in CIKM 2011 on refined pattern discovery tool for SystemT
- Integrated pattern discovery standalone tool for unstructured text into IBM’s Infosphere BigInsights
- Published paper in PVLDB 2010 on provenance research
- Developed queries for information integration on large (TB) datasets; redesigned SystemT UI & backend

Professor Dennis K. Lieu, UC Berkeley, Berkeley, CA
Undergraduate Research Assistant, August 2008 - May 2009
- Lead undergrad research team working on the integration of flywheels for a triple hybrid drive train
- Calculated efficiency of drive train integrated with different designs of flywheels and battery types

Residential Computing, RSSP-IT, UC Berkeley
Lead Desktop Administrator, August 2008 - June 2009, Desktop Administrator, August 2007 - May 2008
- Managed four-member Desktop Administration team with sole discretion over hiring and annual evaluations
- Managed six residential computing centers serving ~6,000 residents that offer 150 computers with Mac and
Windows run off Ethernet-base network hosted by UNIX-based server

Journal V. Chu, I. McMahon, L. Riano, C. G. McDonald, Q. He, J. M. Perez-Tejada, M. Arrigo, T. Darrell, K. J.
Articles: Kuchenbecker, Robotic Learning of Haptic Adjectives Through Physical Interaction. Robotics and
Autonomous Systems, 2015

Conference V. Chu, B. Akgun, A.L. Thomaz, Learning Haptic Affordances from Demonstration and Human-Guided
Proceedings: Exploration. HAPTICS, 2016 (In Press)

V.Chu, T. Fitzgerald, A.L. Thomaz, Learning Object Affordances by Leverage the Combination of Human-
Guidance and Self-Exploration. HRI, 2016 (In Press)

V. Chu, K. Bullard, A.L. Thomaz, Multimodal Real-Time Contingency Detection for HRI. IROS, 2014

V. Chu, I. McMahon, L. Riano, C. G. McDonald, Q. He, J. M. Perez-Tejada, M. Arrigo, N. Fitter, J. C. Nappo,
T. Darrell, K. J. Kuchenbecker. Using Robotic Exploratory Procedures to Learn the Meaning of Haptic
Adjectives. ICRA, 2013 (Best Cognitive Robotics Paper Award)

Y. Li, V. Chu, S. Blohm, H. Zhu, H. Ho. Facilitating Pattern Discovery for Relation Extraction with Semantic-