The MADMAX data set for visual‐inertial rover navigation on Mars

Meyer, Lukas; Smíšek, Michal; Villacampa, Alejandro Fontan; Maza, Laura Oliva; Medina, Daniel; Schuster, Martin J.; Steidle, Florian; Vayugundla, Mallikarjuna; Müller, Marcus G.; Rebele, Bernhard; Wedler, Armin; Triebel, Rudolph

doi:10.1002/rob.22016

Cited by 33 publications

(33 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The visual navigation dataset of the planetary rovers used in this research is part1 and part2 of the Katwijk beach planetary rover dataset [49] from the European Space Agency (ESA) [50][51][52][53], which contains 2250 frames of the image. The Katwijk dataset is a professional open dataset for the navigation vision of the planetary rover research, and many studies use the Katwijk dataset as the planetary environment [44,52,54].…”

Section: The Real-life Visual Navigation Dataset For the Planetary Roversmentioning

confidence: 99%

Rock Segmentation in the Navigation Vision of the Planetary Rovers

et al. 2021

View full text Add to dashboard Cite

Visual navigation is an essential part of planetary rover autonomy. Rock segmentation emerged as an important interdisciplinary topic among image processing, robotics, and mathematical modeling. Rock segmentation is a challenging topic for rover autonomy because of the high computational consumption, real-time requirement, and annotation difficulty. This research proposes a rock segmentation framework and a rock segmentation network (NI-U-Net++) to aid with the visual navigation of rovers. The framework consists of two stages: the pre-training process and the transfer-training process. The pre-training process applies the synthetic algorithm to generate the synthetic images; then, it uses the generated images to pre-train NI-U-Net++. The synthetic algorithm increases the size of the image dataset and provides pixel-level masks—both of which are challenges with machine learning tasks. The pre-training process accomplishes the state-of-the-art compared with the related studies, which achieved an accuracy, intersection over union (IoU), Dice score, and root mean squared error (RMSE) of 99.41%, 0.8991, 0.9459, and 0.0775, respectively. The transfer-training process fine-tunes the pre-trained NI-U-Net++ using the real-life images, which achieved an accuracy, IoU, Dice score, and RMSE of 99.58%, 0.7476, 0.8556, and 0.0557, respectively. Finally, the transfer-trained NI-U-Net++ is integrated into a planetary rover navigation vision and achieves a real-time performance of 32.57 frames per second (or the inference time is 0.0307 s per frame). The framework only manually annotates about 8% (183 images) of the 2250 images in the navigation vision, which is a labor-saving solution for rock segmentation tasks. The proposed rock segmentation framework and NI-U-Net++ improve the performance of the state-of-the-art models. The synthetic algorithm improves the process of creating valid data for the challenge of rock segmentation. All source codes, datasets, and trained models of this research are openly available in Cranfield Online Research Data (CORD).

show abstract

Section: The Real-life Visual Navigation Dataset For the Planetary Roversmentioning

confidence: 99%

Rock Segmentation in the Navigation Vision of the Planetary Rovers

et al. 2021

View full text Add to dashboard Cite

show abstract

“…As an additional source of planetary-analogue data, we consider the MADMAX dataset [38]. It provides data from a Mars-analogue scenario in the desert of Morocco that does not fit the conditions of Phobos precisely, but features long navigation trajectories and results from state-of-the-art algorithms as benchmarks.…”

Section: Testingmentioning

confidence: 99%

The MMX Rover on Phobos: The Preliminary Design of the DLR Autonomous Navigation Experiment

Vayugundla

Bodenmüller

Schuster

et al. 2021

2021 IEEE Aerospace Conference (50100)

Self Cite

View full text Add to dashboard Cite

This paper summarizes the challenges of deep space planetary robotic-exploration missions, using the example of the DLR Autonomous Navigation Experiment within the MMX Rover project, and presents a preliminary design of the proposed solution to safe navigation of the MMX Rover on Phobos. The MMX Rover, a joint contribution of the German Aerospace Center (DLR) and the Centre National d'Etudes Spatiales (CNES) is part of the Martian Moons eXploration (MMX) Mission by the Japan Aerospace Exploration Agency (JAXA), whose mission objective is to understand the origin of the two Martian moons. The mission involves scientific data collection and a sample return from Phobos, the bigger of the two Martian moons. The mission is scheduled to launch in the third quarter of 2024. The MMX Rover will fly as a payload and will be jettisoned onto Phobos from a low altitude of about 40 metres. The rover has multiple objectives: terrain assessment to mitigate the risk for the sample-return approach of the spacecraft, exploration of the surface of Phobos and act as a scientific and technology demonstration platform. In this context, the Robotics and Mechatronics Institute, DLR, is preparing to perform an Autonomous Navigation Experiment as a technology demonstration to showcase the advantages of robot autonomy for exploring distant celestial bodies. If successful, the MMX Rover will be the first man-made object to land on and explore Phobos.

show abstract

“…In the last decades, academic research has brought up various different solution strategies to tackle the visual odometry problem. However, when evaluating recent state-of-theart systems on datasets which resemble Mars-like environ-ments such as the one recorded in the Morocco-Acquired Dataset of Mars-Analogue eXploration (MADMAX) [1], we especially found that methods relying on descriptor-based feature matching lack robustness. While direct methods seem promising, they still require significantly more computational resources and are overall not as mature.…”

Section: Introductionmentioning

confidence: 97%

Towards Robust Monocular Visual Odometry for Flying Robots on Planetary Missions

Wudenka

Müller

Demmel

et al. 2021

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

In the future, extraterrestrial expeditions will not only be conducted by rovers but also by flying robots. The technical demonstration drone Ingenuity, that just landed on Mars, will mark the beginning of a new era of exploration unhindered by terrain traversability. Robust self-localization is crucial for that. Cameras that are lightweight, cheap and information-rich sensors are already used to estimate the egomotion of vehicles. However, methods proven to work in manmade environments cannot simply be deployed on other planets. The highly repetitive textures present in the wastelands of Mars pose a huge challenge to descriptor matching based approaches.In this paper, we present an advanced robust monocular odometry algorithm that uses efficient optical flow tracking to obtain feature correspondences between images and a refined keyframe selection criterion. In contrast to most other approaches, our framework can also handle rotation-only motions that are particularly challenging for monocular odometry systems. Furthermore, we present a novel approach to estimate the current risk of scale drift based on a principal component analysis of the relative translation information matrix. This way we obtain an implicit measure of uncertainty. We evaluate the validity of our approach on all sequences of a challenging realworld dataset captured in a Mars-like environment and show that it outperforms state-of-the-art approaches. The source code

show abstract

The MADMAX data set for visual‐inertial rover navigation on Mars

Cited by 33 publications

References 35 publications

Rock Segmentation in the Navigation Vision of the Planetary Rovers

Rock Segmentation in the Navigation Vision of the Planetary Rovers

The MMX Rover on Phobos: The Preliminary Design of the DLR Autonomous Navigation Experiment

Towards Robust Monocular Visual Odometry for Flying Robots on Planetary Missions

Contact Info

Product

Resources

About