AI4MARS: A Dataset for Terrain-Aware Autonomous Driving on Mars

Swan, R. Michael; Atha, Deegan; Leopold, Henry A.; Gildner, Matthew; Oij, Stephanie; Chiu, Cindy C.W.; Ono, Masahiro

doi:10.1109/cvprw53098.2021.00226

Cited by 64 publications

(40 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work leverages the first large-scale dataset of Mars images labeled for the purposes of terrain classification and traversability assessment -AI4Mars [34]. This dataset consists of images from Curiosity rover's navigation camera (NAVCAM) and color mast camera (Mastcam).…”

Section: Problem Descriptionmentioning

confidence: 99%

“…In this approach, citizen scientists voluntarily identified four classes -soil, bedrock, sand, and big rock -in MSL images. These submissions were then aggregated based on several pixel-level agreement heuristics described in [34]. The "gold standard" test set includes 322 images where each pixel's label was unanimously determined by three experts.…”

Section: Problem Descriptionmentioning

confidence: 99%

“…This generalized representation learning is achieved by Figure 1. Sample MSL images and labels from the AI4Mars dataset [34].…”

Section: Network Architecturementioning

confidence: 99%

“…2) Supervised Finetuning: For supervised finetuning, a segmentation head consisting of an atrous convolution block and A pixel-wise cross entropy loss function which neglects unlabeled pixels is used in this work. Weighting the loss function by relative pixel frequencies was found in [34] to offer little in the way of performance but incur additional computational cost, and was therefore not investigated. A batch size of 8 and a learning rate of 0.057 0.002 × √ 8 is used for finetuning.…”

Section: Network Architecturementioning

confidence: 99%

See 3 more Smart Citations

Mars Terrain Segmentation with Less Labels

Goh¹,

Chen²,

Wilson³

2022

Preprint

View full text Add to dashboard Cite

Planetary rover systems need to perform terrain segmentation to identify drivable areas as well as identify specific types of soil for sample collection. The latest Martian terrain segmentation methods rely on supervised learning which is very data hungry and difficult to train where only a small number of labeled samples are available. Moreover, the semantic classes are defined differently for different applications (e.g., rover traversal vs. geological) and as a result the network has to be trained from scratch each time, which is an inefficient use of resources. This research proposes a semi-supervised learning framework for Mars terrain segmentation where a deep segmentation network trained in an unsupervised manner on unlabeled images is transferred to the task of terrain segmentation trained on few labeled images. The network incorporates a backbone module which is trained using a contrastive loss function and an output atrous convolution module which is trained using a pixel-wise cross-entropy loss function. Evaluation results using the metric of segmentation accuracy show that the proposed method with contrastive pretraining outperforms plain supervised learning by 2%-10%. Moreover, the proposed model is able to achieve a segmentation accuracy of 91.1% using only 161 training images (1% of the original dataset) compared to 81.9% with plain supervised learning.

show abstract

Section: Problem Descriptionmentioning

confidence: 99%

Section: Problem Descriptionmentioning

confidence: 99%

“…This generalized representation learning is achieved by Figure 1. Sample MSL images and labels from the AI4Mars dataset [34].…”

Section: Network Architecturementioning

confidence: 99%

Section: Network Architecturementioning

confidence: 99%

See 2 more Smart Citations

Mars Terrain Segmentation with Less Labels

Goh¹,

Chen²,

Wilson³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…SPOC analysis of images captured by the Navcam of the Curiosity rover to identify areas of slippage on Mars terrain and images captured by the HiRISE camera to determine the traversability of potential landing sites for the Mars 2020 Rover mission. SPOC has been successful for these purposes and will become more successful after refinement using the AI4MARS dataset [35]. AI4Mars is a large dataset totaling 35K high-resolution images taken on the surface of Mars from the Opportunity, Spirit, and Curiosity rovers, with approximately ten people labeling each image in the dataset to ensure that each image is high quality [35].…”

Section: Ii3 Spoc and Ai4marsmentioning

confidence: 99%

Machine Learning for Mars Exploration

Momennasab¹

2021

Preprint

View full text Add to dashboard Cite

Risk to human astronauts and interplanetary distance causing slow and limited communication drives scientists to pursue an autonomous approach to exploring distant planets, such as Mars. A portion of exploration of Mars has been conducted through the autonomous collection and analysis of Martian data by spacecraft such as the Mars rovers and the Mars Express Orbiter. The autonomy used on these Mars exploration spacecraft and on Earth to analyze data collected by these vehicles mainly consist of machine learning, a field of artificial intelligence where algorithms collect data and self-improve with the data. Additional applications of machine learning techniques for Mars exploration have potential to resolve communication limitations and human risks of interplanetary exploration. In addition, analyzing Mars data with machine learning has the potential to provide a greater understanding of Mars in numerous domains such as its climate, atmosphere, and potential future habitation. To explore further utilizations of machine learning techniques for Mars exploration, this paper will first summarize the general features and phenomena of Mars to provide a general overview of the planet, elaborate upon uncertainties of Mars that would be beneficial to explore and understand, summarize every current or previous usage of machine learning techniques in the exploration of Mars, explore implementations of machine learning that will be utilized in future Mars exploration missions, and explore machine learning techniques used in Earthly domains to provide solutions to the previously described uncertainties of Mars.

show abstract