-
Transfer Learning Study of Motion Transformer-based Trajectory Predictions
Authors:
Lars Ullrich,
Alex McMaster,
Knut Graichen
Abstract:
Trajectory planning in autonomous driving is highly dependent on predicting the emergent behavior of other road users. Learning-based methods are currently showing impressive results in simulation-based challenges, with transformer-based architectures technologically leading the way. Ultimately, however, predictions are needed in the real world. In addition to the shifts from simulation to the rea…
▽ More
Trajectory planning in autonomous driving is highly dependent on predicting the emergent behavior of other road users. Learning-based methods are currently showing impressive results in simulation-based challenges, with transformer-based architectures technologically leading the way. Ultimately, however, predictions are needed in the real world. In addition to the shifts from simulation to the real world, many vehicle- and country-specific shifts, i.e. differences in sensor systems, fusion and perception algorithms as well as traffic rules and laws, are on the agenda. Since models that can cover all system setups and design domains at once are not yet foreseeable, model adaptation plays a central role. Therefore, a simulation-based study on transfer learning techniques is conducted on basis of a transformer-based model. Furthermore, the study aims to provide insights into possible trade-offs between computational time and performance to support effective transfers into the real world.
△ Less
Submitted 9 July, 2024; v1 submitted 12 April, 2024;
originally announced April 2024.
-
VeSPA: The SuperWASP Variable Star Photometry Archive
Authors:
Adam McMaster,
Andrew J. Norton,
Hugh J. Dickinson,
Heidi B. Thiemann,
Ulrich C. Kolb
Abstract:
We present the first results from the SuperWASP Variable Stars (SVS) citizen science project. The photometry archive of the Wide Angle Search for Planets has previously been searched for periodic variations and the results of this search formed the basis of the SVS project on the Zooniverse. The SVS project asks volunteers to visually inspect light curve plots and categorize each one according to…
▽ More
We present the first results from the SuperWASP Variable Stars (SVS) citizen science project. The photometry archive of the Wide Angle Search for Planets has previously been searched for periodic variations and the results of this search formed the basis of the SVS project on the Zooniverse. The SVS project asks volunteers to visually inspect light curve plots and categorize each one according to a broad classification scheme. Results from the first two years of SVS have now been published online as the SuperWASP Variable Star Photometry Archive (VeSPA). The archive can be browsed online, downloaded in full, or queried, filtered, and sorted to export a refined set of results. An interactive light curve viewer also allows any light curve to be folded at a user-defined period. Analysis of citizen science results and development of VeSPA features are both ongoing. Updated results will be published every six months.
△ Less
Submitted 16 November, 2021;
originally announced November 2021.
-
SuperWASP Variable Stars: Classifying Light Curves Using Citizen Science
Authors:
Heidi B. Thiemann,
Andrew J. Norton,
Hugh J. Dickinson,
Adam McMaster,
Ulrich C. Kolb
Abstract:
We present the first analysis of results from the SuperWASP Variable Stars Zooniverse project, which is aiming to classify 1.6 million phase-folded light curves of candidate stellar variables observed by the SuperWASP all sky survey with periods detected in the SuperWASP periodicity catalogue. The resultant data set currently contains $>$1 million classifications corresponding to $>$500,000 object…
▽ More
We present the first analysis of results from the SuperWASP Variable Stars Zooniverse project, which is aiming to classify 1.6 million phase-folded light curves of candidate stellar variables observed by the SuperWASP all sky survey with periods detected in the SuperWASP periodicity catalogue. The resultant data set currently contains $>$1 million classifications corresponding to $>$500,000 object-period combinations, provided by citizen scientist volunteers. Volunteer-classified light curves have $\sim$89 per cent accuracy for detached and semi-detached eclipsing binaries, but only $\sim$9 per cent accuracy for rotationally modulated variables, based on known objects. We demonstrate that this Zooniverse project will be valuable for both population studies of individual variable types and the identification of stellar variables for follow up. We present preliminary findings on various unique and extreme variables in this analysis, including long period contact binaries and binaries near the short-period cutoff, and we identify 301 previously unknown binaries and pulsators. We are now in the process of develo** a web portal to enable other researchers to access the outputs of the SuperWASP Variable Stars project.
△ Less
Submitted 15 January, 2021;
originally announced January 2021.
-
Planet Hunters TESS II: Findings from the first two years of TESS
Authors:
Nora L. Eisner,
Oscar Barragán,
Chris Lintott,
Suzanne Aigrain,
Belinda Nicholson,
Tabetha S. Boyajian,
Steve B. Howell,
Cole Johnston,
Ben Lakeland,
Grant Miller,
Adam McMaster,
Hannu Parviainen,
Emily J. Safron,
Megan E. Schwamb,
Laura Trouille,
Sophia Vaughan,
Norbert Zicher,
Campbell Allen,
Sarah Allen,
Mark Bouslog,
Cliff Johnson,
Molly N. Simon,
Zach Wolfenbarger,
Elisabeth M. L. Baeten,
David M. Bundy
, et al. (1 additional authors not shown)
Abstract:
We present the results from the first two years of the Planet Hunters TESS citizen science project, which identifies planet candidates in the TESS data by engaging members of the general public. Over 22,000 citizen scientists from around the world visually inspected the first 26 Sectors of TESS data in order to help identify transit-like signals. We use a clustering algorithm to combine these clas…
▽ More
We present the results from the first two years of the Planet Hunters TESS citizen science project, which identifies planet candidates in the TESS data by engaging members of the general public. Over 22,000 citizen scientists from around the world visually inspected the first 26 Sectors of TESS data in order to help identify transit-like signals. We use a clustering algorithm to combine these classifications into a ranked list of events for each sector, the top 500 of which are then visually vetted by the science team. We assess the detection efficiency of this methodology by comparing our results to the list of TESS Objects of Interest (TOIs) and show that we recover 85 % of the TOIs with radii greater than 4 Earth radii and 51 % of those with radii between 3 and 4 Earth radii. Additionally, we present our 90 most promising planet candidates that had not previously been identified by other teams, 73 of which exhibit only a single transit event in the TESS light curve, and outline our efforts to follow these candidates up using ground-based observatories. Finally, we present noteworthy stellar systems that were identified through the Planet Hunters TESS project.
△ Less
Submitted 27 November, 2020;
originally announced November 2020.
-
Survey of Gravitationally-lensed Objects in HSC Imaging (SuGOHI). VI. Crowdsourced lens finding with Space Warps
Authors:
Alessandro Sonnenfeld,
Aprajita Verma,
Anupreeta More,
Elisabeth Baeten,
Christine Macmillan,
Kenneth C. Wong,
James H. H. Chan,
Anton T. Jaelani,
Chien-Hsiu Lee,
Masamune Oguri,
Cristian E. Rusu,
Marten Veldthuis,
Laura Trouille,
Philip J. Marshall,
Roger Hutchings,
Campbell Allen,
James O' Donnell,
Claude Cornen,
Christopher Davis,
Adam McMaster,
Chris Lintott,
Grant Miller
Abstract:
Strong lenses are extremely useful probes of the distribution of matter on galaxy and cluster scales at cosmological distances, but are rare and difficult to find. The number of currently known lenses is on the order of 1,000. We wish to use crowdsourcing to carry out a lens search targeting massive galaxies selected from over 442 square degrees of photometric data from the Hyper Suprime-Cam (HSC)…
▽ More
Strong lenses are extremely useful probes of the distribution of matter on galaxy and cluster scales at cosmological distances, but are rare and difficult to find. The number of currently known lenses is on the order of 1,000. We wish to use crowdsourcing to carry out a lens search targeting massive galaxies selected from over 442 square degrees of photometric data from the Hyper Suprime-Cam (HSC) survey. We selected a sample of $\sim300,000$ galaxies with photometric redshifts in the range $0.2 < z_{phot} < 1.2$ and photometrically inferred stellar masses $\log{M_*} > 11.2$. We crowdsourced lens finding on this sample of galaxies on the Zooniverse platform, as part of the Space Warps project. The sample was complemented by a large set of simulated lenses and visually selected non-lenses, for training purposes. Nearly 6,000 citizen volunteers participated in the experiment. In parallel, we used YattaLens, an automated lens finding algorithm, to look for lenses in the same sample of galaxies. Based on a statistical analysis of classification data from the volunteers, we selected a sample of the most promising $\sim1,500$ candidates which we then visually inspected: half of them turned out to be possible (grade C) lenses or better. Including lenses found by YattaLens or serendipitously noticed in the discussion section of the Space Warps website, we were able to find 14 definite lenses, 129 probable lenses and 581 possible lenses. YattaLens found half the number of lenses discovered via crowdsourcing. Crowdsourcing is able to produce samples of lens candidates with high completeness and purity, compared to currently available automated algorithms. A hybrid approach, in which the visual inspection of samples of lens candidates pre-selected by discovery algorithms and/or coupled to machine learning is crowdsourced, will be a viable option for lens finding in the 2020s.
△ Less
Submitted 4 July, 2021; v1 submitted 1 April, 2020;
originally announced April 2020.
-
Planet Four: Probing Springtime Winds on Mars by Map** the Southern Polar CO$_2$ Jet Deposits
Authors:
K. -Michael Aye,
Megan E. Schwamb,
Ganna Portyankina,
Candice J. Hansen,
Adam McMaster,
Grant R. M. Miller,
Brian Carstensen,
Christopher Snyder,
Michael Parrish,
Stuart Lynn,
Chuhong Mai,
David Miller,
Robert J. Simpson,
Arfon M. Smith
Abstract:
The springtime sublimation process of Mars' southern seasonal polar CO$_2$ ice cap features dark fan-shaped deposits appearing on the top of the thawing ice sheet. The fan material likely originates from the surface below the ice sheet, brought up via CO$_2$ jets breaking through the seasonal ice cap. Once the dust and dirt is released into the atmosphere, the material may be blown by the surface…
▽ More
The springtime sublimation process of Mars' southern seasonal polar CO$_2$ ice cap features dark fan-shaped deposits appearing on the top of the thawing ice sheet. The fan material likely originates from the surface below the ice sheet, brought up via CO$_2$ jets breaking through the seasonal ice cap. Once the dust and dirt is released into the atmosphere, the material may be blown by the surface winds into the dark streaks visible from orbit. The location, size and direction of these fans record a number of parameters important to quantifying seasonal winds and sublimation activity, the most important agent of geological change extant on Mars. We present results of a systematic map** of these south polar seasonal fans with the Planet Four online citizen science project. Planet Four enlists the general public to map the shapes, directions, and sizes of the seasonal fans visible in orbital images. Over 80,000 volunteers have contributed to the Planet Four project, reviewing 221 images, from Mars Reconnaissance Orbiter's HiRISE (High Resolution Imaging Science Experiment) camera, taken in southern spring during Mars Years 29 and 30. We provide an overview of Planet Four and detail the processes of combining multiple volunteer assessments together to generate a high fidelity catalog of $\sim$ 400,000 south polar seasonal fans. We present the results from analyzing the wind directions at several locations monitored by HiRISE over two Mars years, providing new insights into polar surface winds.
△ Less
Submitted 24 October, 2018; v1 submitted 27 March, 2018;
originally announced March 2018.
-
Floating Forests: Quantitative Validation of Citizen Science Data Generated From Consensus Classifications
Authors:
Isaac S. Rosenthal,
Jarrett E. K. Byrnes,
Kyle C. Cavanaugh,
Tom W. Bell,
Briana Harder,
Alison J. Haupt,
Andrew T. W. Rassweiler,
Alejandro Pérez-Matus,
Jorge Assis,
Ali Swanson,
Amy Boyer,
Adam McMaster,
Laura Trouille
Abstract:
Large-scale research endeavors can be hindered by logistical constraints limiting the amount of available data. For example, global ecological questions require a global dataset, and traditional sampling protocols are often too inefficient for a small research team to collect an adequate amount of data. Citizen science offers an alternative by crowdsourcing data collection. Despite growing popular…
▽ More
Large-scale research endeavors can be hindered by logistical constraints limiting the amount of available data. For example, global ecological questions require a global dataset, and traditional sampling protocols are often too inefficient for a small research team to collect an adequate amount of data. Citizen science offers an alternative by crowdsourcing data collection. Despite growing popularity, the community has been slow to embrace it largely due to concerns about quality of data collected by citizen scientists. Using the citizen science project Floating Forests (http://floatingforests.org), we show that consensus classifications made by citizen scientists produce data that is of comparable quality to expert generated classifications. Floating Forests is a web-based project in which citizen scientists view satellite photographs of coastlines and trace the borders of kelp patches. Since launch in 2014, over 7,000 citizen scientists have classified over 750,000 images of kelp forests largely in California and Tasmania. Images are classified by 15 users. We generated consensus classifications by overlaying all citizen classifications and assessed accuracy by comparing to expert classifications. Matthews correlation coefficient (MCC) was calculated for each threshold (1-15), and the threshold with the highest MCC was considered optimal. We showed that optimal user threshold was 4.2 with an MCC of 0.400 (0.023 SE) for Landsats 5 and 7, and a MCC of 0.639 (0.246 SE) for Landsat 8. These results suggest that citizen science data derived from consensus classifications are of comparable accuracy to expert classifications. Citizen science projects should implement methods such as consensus classification in conjunction with a quantitative comparison to expert generated classifications to avoid concerns about data quality.
△ Less
Submitted 25 January, 2018;
originally announced January 2018.
-
The K2-138 System: A Near-Resonant Chain of Five Sub-Neptune Planets Discovered by Citizen Scientists
Authors:
Jessie L. Christiansen,
Ian J. M. Crossfield,
Geert Barentsen,
Chris J. Lintott,
Thomas Barclay,
Brooke D. Simmons,
Erik Petigura,
Joshua E. Schlieder,
Courtney D. Dressing,
Andrew Vanderburg,
David R. Ciardi,
Campbell Allen,
Adam McMaster,
Grant Miller,
Martin Veldthuis,
Sarah Allen,
Zach Wolfenbarger,
Brian Cox,
Julia Zemiro,
Andrew W. Howard,
John Livingston,
Evan Sinukoff,
Timothy Catron,
Andrew Grey,
Joshua J. E. Kusch
, et al. (3 additional authors not shown)
Abstract:
K2-138 is a moderately bright (V = 12.2, K = 10.3) main sequence K-star observed in Campaign 12 of the NASA K2 mission. It hosts five small (1.6-3.3R_Earth) transiting planets in a compact architecture. The periods of the five planets are 2.35 d, 3.56 d, 5.40 d, 8.26 d, and 12.76 d, forming an unbroken chain of near 3:2 resonances. Although we do not detect the predicted 2-5 minute transit timing…
▽ More
K2-138 is a moderately bright (V = 12.2, K = 10.3) main sequence K-star observed in Campaign 12 of the NASA K2 mission. It hosts five small (1.6-3.3R_Earth) transiting planets in a compact architecture. The periods of the five planets are 2.35 d, 3.56 d, 5.40 d, 8.26 d, and 12.76 d, forming an unbroken chain of near 3:2 resonances. Although we do not detect the predicted 2-5 minute transit timing variations with the K2 timing precision, they may be observable by higher cadence observations with, for example, Spitzer or CHEOPS. The planets are amenable to mass measurement by precision radial velocity measurements, and therefore K2-138 could represent a new benchmark systems for comparing radial velocity and TTV masses. K2-138 is the first exoplanet discovery by citizen scientists participating in the Exoplanet Explorers project on the Zooniverse platform.
△ Less
Submitted 19 January, 2018; v1 submitted 11 January, 2018;
originally announced January 2018.
-
Planet Four: Terrains - Discovery of Araneiforms Outside of the South Polar Layered Deposits
Authors:
Megan E. Schwamb,
Klaus-Michael Aye,
Ganna Portyankina,
Candice J. Hansen,
Campbell Allen,
Sarah Allen,
Fred J. Calef III,
Simone Duca,
Adam McMaster,
Grant R. M. Miller
Abstract:
We present the results of a systematic map** of seasonally sculpted terrains on the South Polar region of Mars with the Planet Four: Terrains (P4T) online citizen science project. P4T enlists members of the general public to visually identify features in the publicly released Mars Reconnaissance Orbiter CTX images. In particular, P4T volunteers are asked to identify: 1) araneiforms (including fe…
▽ More
We present the results of a systematic map** of seasonally sculpted terrains on the South Polar region of Mars with the Planet Four: Terrains (P4T) online citizen science project. P4T enlists members of the general public to visually identify features in the publicly released Mars Reconnaissance Orbiter CTX images. In particular, P4T volunteers are asked to identify: 1) araneiforms (including features with a central pit and radiating channels known as 'spiders'); 2) erosional depressions, troughs, mesas, ridges, and quasi-circular pits characteristic of the South Polar Residual Cap (SPRC) which we collectively refer to as 'Swiss cheese terrain', and 3) craters. In this work we present the distributions of our high confidence classic spider araneiforms and Swiss cheese terrain identifications. We find no locations within our high confidence spider sample that also have confident Swiss cheese terrain identifications. Previously spiders were reported as being confined to the South Polar Layered Deposits (SPLD). Our work has provided the first identification of spiders at locations outside of the SPLD, confirmed with high resolution HiRISE imaging. We find araneiforms on the Amazonian and Hesperian polar units and the Early Noachian highland units, with 75% of the identified araneiform locations in our high confidence sample residing on the SPLD. With our current coverage, we cannot confirm whether these are the only geologic units conducive to araneiform formation on the Martian South Polar region. Our results are consistent with the current CO2 jet formation scenario with the process exploiting weaknesses in the surface below the seasonal CO2 ice sheet to carve araneiform channels into the regolith over many seasons. These new regions serve as additional probes of the conditions required for channel creation in the CO2 jet process. (Abridged)
△ Less
Submitted 25 August, 2017;
originally announced August 2017.
-
A transient search using combined human and machine classifications
Authors:
Darryl E. Wright,
Chris J. Lintott,
Stephen J. Smartt,
Ken W. Smith,
Lucy Fortson,
Laura Trouille,
Campbell R. Allen,
Melanie Beck,
Mark C. Bouslog,
Amy Boyer,
K. C. Chambers,
Heather Flewelling,
Will Granger,
Eugene A. Magnier,
Adam McMaster,
Grant R. M. Miller,
James E. O'Donnell,
Helen Spiers,
John L. Tonry,
Marten Veldthuis,
Richard J. Wainscoat,
Chris Waters,
Mark Willman,
Zach Wolfenbarger,
Dave R. Young
Abstract:
Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combini…
▽ More
Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combining classifications from volunteers participating in a citizen science project with those from a convolutional neural network. The unique aspect of this work is the deployment, in combination, of both human and machine classifications for near real-time discovery in an astronomical project. We show that the combination of the two methods outperforms either one used individually. This result has important implications for the future development of transient searches, especially in the era of LSST and other large-throughput surveys.
△ Less
Submitted 17 July, 2017;
originally announced July 2017.
