Onboard autonomous rock shape analysis for Mars rovers
The DOI matches its Crossref record; 17 of the 17 references it cites were found in a registry or an index.
Abstract
In the quest to maximize the scientific return of future robotic missions, it is imperative that our rovers be capable of determining the importance of the science they collect so that they may prioritize the acquisition and relay of that data. As an important step in this process, we present an automated technique to allow a rover to classify the shape and other geological characteristics of rocks from 2D photographic images and 3D stereographically produced data. Experiments were conducted in the Matlab environment using images returned by JPL's Mars Pathfinder mission. Our method begins by first segmenting the rocks from the background using a combination of image intensity and height data. Various metrics are then used to classify the region's sphericity, roundness, and other geometric properties. Preliminary experiments to determine the most useful metrics were conducted by characterizing the 2D rock shape while the 3D shape was later studied with metrics derived from these 2D techniques. Seven measures were developed and implemented. The performance of each measure was characterized by analyzing images from the Pathfinder mission and ranking the rocks according to the measured properties. Combined, the measures would provide a tool by which an automated rover could discover a greater amount of information about the data it collects, leading to a more productive mission.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Comparing images using the Hausdorff distance
Huttenlocher, D. P., Klanderman, G. A., & Rucklidge, W. J. (1993) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.232073 · Cited by 4,545
Direct least square fitting of ellipses
Fitzgibbon, A., Pilu, M., & Fisher, R. B. (1999) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/34.765658 · Cited by 2,752
Mathematical elements for computer graphics
Rogers, D. F., & Adams, J. A. (1976) · CERN Document Server (European Organization for Nuclear Research) · Cited by 1,072
Least-squares fitting of circles and ellipses
Gander, W., Golub, G. H., & Strebel, R. (1994) · BIT · doi:10.1007/bf01934268 · Cited by 860
Mathematical Elements for Computer Graphics
Mudur, S. P. (1986) · Advances in Computer Graphics I · doi:10.1007/978-3-642-46514-7_2 · Cited by 626 · cited 4×
“However, the fitting of b-splines to the data points via techniques derived in [20] smoothed the surfaces sufficiently that derivatives could then be calculated in an accurate and analytic method.”
BackgroundMethodInfluentialFitting conic sections to scattered data
Bookstein, F. L. (1979) · Computer Graphics and Image Processing · doi:10.1016/0146-664x(79)90082-0 · Cited by 528
Direct least squares fitting of ellipses
Fitzgibbon, A. W., Pilu, M., & Fisher, R. B. (1996) · Proceedings of 13th International Conference on Pattern Recognition · doi:10.1109/icpr.1996.546029 · Cited by 289 · cited 1×
“For a more detailed discussion of the ellipsespecific fitting problem, please see [7] or [8].”
BackgroundA note on the least squares fitting of ellipses
Rosin, P. L. (1993) · Pattern Recognition Letters · doi:10.1016/0167-8655(93)90062-i · Cited by 207 · cited 1×
“While numerous constraints have been proposed in the past [4] [10] [21], Fitzgibbon, Pilu, and Fisher set4ac b(2) = 1, thus forcing the discriminant to be negative and creating an ellipse specific minimization.”
BackgroundShape Representation and Recognition from Multiscale Curvature
Dudek, G., & Tsotsos, J. K. (1997) · Computer Vision and Image Understanding · doi:10.1006/cviu.1997.0533 · Cited by 110 · cited 1×
“However, a great deal of geological work has been performed classifying and categorizing the general appearance of microscopic particle grains with respect to various concrete properties [1], [6], [9], [13] [18], and the same basic concepts remain applicable even when scaled to the macroscopic realm of Martian rocks.”
BackgroundComparison of Combined Shape Descriptors for Irregular Objects.
Iivarinen, J., Peura, M., Särelä, J., & Visa, A. J. E. (1997) · Cited by 63
Research works printed but not linked: 8
- press, 1997, Numerical Recipes in C, 394
- turner, 1999, An algorithm for fitting an ellipsoid to data, Technical Report RR9803
- kwon, 1998, Hierarchical hausdorff distance matching algorithm using pyramidal structures, Proceedings of 4th Japan-Korea Joint Workshop on Computer Vision, 80
- krumbein, 1951, Statigruphy and Sedimentation
- blatt, 1972, Origin of Sedimentary Rocks Prentice-Hall Inc, 434
- 1995, Standard Test Method for Flat Particles Elongated Particles or Flat and Elongated Particles in Coarse Aggregate
- chetverikov, 1999, Detection of high curvature points in planar curves, Proc 23rd Workshop of the Austrian Pattern Recognition Group, 175
- frost, 1996, Digital analysis of aggregate particle shape, Center for Aggregates Research Fourth Annual Symposium Proceedings
Cited by
Oasis: Onboard autonomous science investigation system for opportunistic rover science
Castaño, R., Estlin, T. A., Anderson, R. C., Gaines, D. M., Castaño, A., Bornstein, B., Chouinard, C., & Judd, M. (2007) · Journal of Field Robotics · doi:10.1002/rob.20192 · Cited by 86
Self-citationPerformance Comparison of Rock Detection Algorithms for Autonomous Planetary Geology
Thompson, D. R., & Castaño, R. (2007) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero.2007.352699 · Cited by 64
“Stereo range data is a component in several rock detectors including those by Gor [8], Fox [6] and Thompson [12].”
MethodProbabilistic planning for robotic exploration
Simmons, R. G., & Smith, T. (2007) · Cited by 69
Multi-scale Features for Detection and Segmentation of Rocks in Mars Images
Dunlop, H., Thompson, D. R., & Wettergreen, D. (2007) · 2007 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2007.383257 · Cited by 55
“First, we calculate the residual error between the rock region’s contour and its corresponding best-fit ellipse [2].”
MethodRock modeling and matching for autonomous long‐range Mars rover localization
Li, R., Di, K., Howard, A. B., Matthies, L. H., Wang, J., & Agarwal, S. (2007) · Journal of Field Robotics · doi:10.1002/rob.20182 · Cited by 50
Automatic detection and classification of features of geologic interest
Thompson, D., Niekum, S., Smith, T., & Wettergreen, D. (2005) · 2005 IEEE Aerospace Conference · doi:10.1109/aero.2005.1559329 · Cited by 45
A Kernel-Based Multi-Featured Rock Modeling and Detection Framework for a Mars Rover
Xiao, X., Yao, M., Liu, H., Wang, J., Zhang, L., & Fu, Y. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3131206 · Cited by 46
“Another solution is to model the 3-D structure of the rock, involving stereo-based approaches [11], [12] and point-cloudbased methods [13]–[15].”
BackgroundAutomated Rock Detection and Shape Analysis from Mars Rover Imagery and 3D Point Cloud Data
Di, K., Yue, Z., Liu, Z., & Wang, S. (2013) · Journal of Earth Science · doi:10.1007/s12583-013-0316-3 · Cited by 42
Rover traverse science for increased mission science return
Castafio, R., Anderson, R. C., Estlin, T. A., DeCoste, D., Fisher, F., Gaines, D. M., Mazzoni, D., & Judd, M. (2004) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero.2003.1235546 · Cited by 41
Self-citationAutonomous rock detection on mars through region contrast
Xiao, X., Cui, H., Yao, M., & Tian, Y. (2017) · Advances in Space Research · doi:10.1016/j.asr.2017.04.028 · Cited by 35
Autonomous robotic rock breaking using a real‐time 3D visual perception system
Lampinen, S., Niu, L., Hulttinen, L., Niemi, J., & Mattila, J. (2021) · Journal of Field Robotics · doi:10.1002/rob.22022 · Cited by 39
“In Fox et al. (2002), 2D camera images were combined with range data to detect larger rocks autonomously.”
MethodR ockster : Onboard Rock Segmentation Through Edge Regrouping
Burl, M. C., Thompson, D. R., deGranville, C. K., & Bornstein, B. (2016) · Journal of Aerospace Information Systems · doi:10.2514/1.i010381 · Cited by 33
“2) Stereo height segmentation: fitting a ground plane based on stereo range data and transforming to a height map [34,35]; although this techniqueworks well for large rocks in close proximity to the rover, it does not work well further downrange and cannot detect flush rock surfaces such as outcrops.”
BackgroundAutomatic Identification of Large Fragments in a Pile of Broken Rock Using a Time-of-Flight Camera
McKinnon, C. D., & Marshall, J. A. (2014) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2014.2308011 · Cited by 27
Data Mining During Rover Traverse: From Images to Geologic Signatures
Thompson, D. R., Smith, T., & Wettergreen, D. S. (2005) · Cited by 25
Auto Rock Detection via Sparse-Based Background Modeling for Mars Rover
Xiao, X., Cui, H., Yao, M., Fu, Y., & Qi, W. (2018) · IEEE Congress on Evolutionary Computation (CEC) · doi:10.1109/cec.2018.8477665 · Cited by 23
“Another line of rock detection is 3D modelling, including stereo-based approaches presented in [9], [10], pure point cloud methods in [11] and combination approach with 2D images[12].”
BackgroundLife in the Atacama: Science autonomy for improving data quality
Smith, T., Thompson, D. R., Wettergreen, D. S., Cabrol, N. A., Warren‐Rhodes, K., & Weinstein, S. J. (2007) · Journal of Geophysical Research Atmospheres · doi:10.1029/2006jg000315 · Cited by 20
“[35] Several past systems have used stereo images to find rocks based on their height above the ground plane [Gor et al., 2000; Fox et al., 2002; Pedersen, 2002].”
BackgroundAutonomous onboard traverse science system
Castaño, R., Judd, M., Estlin, T. A., Anderson, R. C., Scharenbroich, L. J., Song, L., Gaines, D. M., Fisher, F., Mazzoni, D., & Castaño, A. (2005) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero.2004.1367601 · Cited by 17
“We begin by fitting an ellipse to the boundary points of the identified rock in the image [7].”
MethodSelf-citationMachine learning challenges in Mars rover traverse science
Castaño, R., Judd, M., Anderson, R. C., & Estlin, T. A. (2003) · Cited by 16
Self-citationKuang, B., Gu, C., Rana, Z. A., Zhao, Y., Sun, S., & Nnabuife, S. G. (2022) · Sensors · doi:10.3390/s22218393 · Cited by 15
“The planetary rover is an essential platform for planetary exploration, widely involved in various scientific activities, including geological exploration [6], planetary history investigation [7], extraterrestrial water resource exploration [8], unknown environment perception [9], life exploration [10], and Mars sampl…”
Li, Y., Xiao, Z., Ma, C., Zeng, L., Zhang, W., Peng, M., & Li, A. (2023) · Journal of Geophysical Research Planets · doi:10.1029/2022je007656 · Cited by 13
Ma, C., Li, Y., Lv, J., Xiao, Z., Zhang, W., & Mo, L. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3371684 · Cited by 22
“D scenes using stereo-vision or multiview techniques and detecting rocks through point cloud processing [8], [24], [25], [26], [27], [28].”
MethodProbabilistic surface classification for rover instrument targeting
Foil, G., Thompson, D. R., Abbey, W. J., & Wettergreen, D. S. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696439 · Cited by 12
“Advances include rock detection and classification of rock characteristics [2], [6], and detection and tracking of dust devils [7].”
BackgroundVisibility Classification of Pellets in Piles for Sizing without Overlapped Particle Error
Andersson, T., Thurley, M. J., & Marklund, O. (2007) · International Conference on Digital Image Computing: Techniques and Applications (DICTA) · doi:10.1109/dicta.2007.4426839 · Cited by 11
“2D shape features have been used to detect broad-leaved weeds in cereal crops [16], to allow a rover to classify the shape and other geologic characteristics of rocks [9], to investigate the suitability of an imaging system to measure shape of particles [4] and for detection and classification of rocks [18].”
BackgroundMulti-scale rock detection on Mars
Li, G., Geng, Y., & Xiao, X. (2018) · Science China Information Sciences · doi:10.1007/s11432-017-9277-x · Cited by 10
Robotic astrobiology – prospects for enhancing scientific productivity of mars rover missions
Ellery, A. A. (2017) · International Journal of Astrobiology · doi:10.1017/s1473550417000180 · Cited by 9
“Rock shape may be extracted using ellipsoid fitting or B-splines (Fox et al. 2002).”
MethodA GRADIENT-REGION CONSTRAINED LEVEL SET METHOD FOR AUTONOMOUS ROCK DETECTION FROM MARS ROVER IMAGE
Yang, J., & Kang, Z. (2019) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlii-2-w13-1479-2019 · Cited by 11
“Fox et al. (2002) segmented the rocks from the background based on intensity and height and classified the shape and other geologic characteristics of rocks from two-dimensional photographic images and three-dimensional stereographically produced data.”
BackgroundVISION BASED OBSTACLE DETECTION USING ROVER STEREO IMAGES
Wang, Y., Peng, M., Di, K., Wan, W., Liu, Z., Yue, Z., Xing, Y., Mao, X. H., & Teng, B. (2019) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlii-2-w13-1471-2019 · Cited by 7
“Fox et al. (2002) first fitted a plane using 3D range data to extract the rocks and then projected them onto 2D images to segment the images; the contours were then fitted into ellipse shapes and passed through several criteria to obtain final rock extraction results.”
MethodAutonomous Rock Instance Segmentation for Extra-Terrestrial Robotic Missions
Durner, M., Boerdijk, W., Fanger, Y., Sakagami, R., Risch, D. L., Triebel, R., & Wedler, A. (2023) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero55745.2023.10115717 · Cited by 8
“Several rock detectors [2], [27], [5] use stereo range data as input modality.”
MethodRobust Rock Detection and Clustering with Surface Analysis for Robotic Rock Breaking Systems
Lampinen, S., & Mattila, J. (2021) · IEEE/ASME International Conference on Advanced Intelligent Mechatronics · doi:10.1109/aim46487.2021.9517695 · Cited by 5
“proposed fusion of 2D imagery and range data for autonomous rock detection for a planetary rover [7].”
BackgroundAutomatic Extraction of 3-D Information of Lunar Surface Rocks Using Topography Data
Liu, X., & Fa, W. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3334151 · Cited by 5
“Other automatic methods based on pattern recognition [19], [20] and stereo image matching techniques [21], [22] are also developed to detect rocks on Mars using rover optical images.”
MethodTechniques for Onboard Prioritization of Science Data for Transmission
Castaño, R., Anderson, R. C., Estlin, T. A., DeCoste, D., Fisher, F., Gaines, D. M., Mazzoni, D., & Judd, M. (2003) · Interplanetary Network Progress Report · Cited by 4
Self-citationLBNet: A Lightweight Bilateral Network for Semantic Segmentation of Martian Rock
Wei, P., Sun, Z., & He, T. (2024) · IEEE Access · doi:10.1109/access.2024.3511084 · Cited by 3
Autonomous Geologist for Planetary Exploration
Sharif, H. (2014) · Carleton University's Institutional Repository (MacOdrum Library, Carleton University) · doi:10.22215/etd/2014-10141 · Cited by 2
Object Modeling and Matching from Multi-view Ground Images for Automated Mars Rover Localization
Li, R., Di, K., Agarwal, S., Wang, J., & Matthies, L. H. (2006) · IEEE Aerospace Conference (AERO) · doi:10.1109/aero.2006.1655782 · Cited by 1
“This method has been tested for use in autonomous rock shape analysis by the Mars rover, the aim of which is to enable the rover to direct its activities to regions of higher geologic significance and to prioritize the downlink of data, thus maximizing the scientific return of the mission [14, 15].”
MethodExpectation-maximization framework for rock textures segmentation
Lotfy, H. M. S., Elmaghraby, A. S., Kantardzic, M. M., & Hadizadeh, J. (2005) · Machine learning for signal processing ... · doi:10.1109/mlsp.2004.1423026 · Cited by 1
Autonomous sample selection and acquisition for planetary exploration
Pugh, S., & Barnes, D. P. (2007) · EPSC
Andersson, T. (2007) · KTH Publication Database DiVA (KTH Royal Institute of Technology)
Solís, M., Prudencio, D., Pérez, M., Prado, Á., Menéndez, O., & Arévalo-Ramirez, T. (2024) · IEEE International Conference on Automatica · doi:10.1109/ica-acca62622.2024.10766792
Metadata
- Authors
- Fox, J., Castano, R., & Anderson, R. C.
- Affiliations
- California Institute of Technology (US)
- ORCID
- Anderson, R.C.
- Venue
- Proceedings, IEEE Aerospace Conference
- Publisher
- IEEE
- Published
- 2003-06-25
- Volume
- 5
- Pages
- 5-2052
- ISSN
- 1095-323X (ISSN-L)
- OpenAlex
- W1868666719
- Cited by
- 38
- References
- 17
- Citations by year
- 2024: 3 · 2023: 3 · 2022: 1 · 2021: 3 · 2019: 2 · 2018: 2 · 2017: 2 · 2016: 1 · 2014: 2 · 2013: 2
- FWCI
- 10.64
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE Aerospace Conference (AERO) (publishedVersion)
- Funders
- Jet Propulsion Laboratory
- Topics
- Physical Sciences › Earth and Planetary Sciences › Space and Planetary Science › Planetary Science and Exploration
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image and Object Detection Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Keywords
- Mars rover, autonomous rover, rock classification, sphericity, roundness, image segmentation, Mars Pathfinder
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.