Towards Building an Automated Rock Classifier for Planetary Rover Missions
نویسندگان
چکیده
Introduction: Building upon our previous work on an automated mineral classifier [1], we are working to extend its functionality towards classification of rocks. The current work focuses on building an automated classifier using Raman spectral data to identify key minerals contained in igneous rocks. The ability to identify minerals in igneous rock samples provides inportant information on their composition. Raman spectroscopy provides unique chemical signatures for minerals and other geological samples. It is a non-destructive technique that does not require sample preparation. Thus it makes an excellent choice to use in conjunction with image data since particular areas of an image (e.g., specific crystals, grains, layers or veins), may correspond to specific Raman signals. Using Artificial Neural Networks to develop a Rock Classifier As in our previous study [1], we are using artificial neural networks (ANNs) because of their ability to learn by example. More specifically, we used a Multilayer Perceptron (MLP), which is an ANN that performs its example based learning using backpropagation, a gradient descent algorithm, which seeks the lowest point of error. Methods: Using the Raman spectra from rock samples and ANNs, we built a classifier to determine if key minerals were present in rocks. As with the mineral classifier in our previous work [1], this classifier had four stages: The input stage, the preprocessing stage, the classification algorithm stage, and the output stage.
منابع مشابه
Vision-based Terrain Classification and Classifier Fusion for Planetary Exploration Rovers
Autonomous rover operation plays a key role in planetary exploration missions. Rover systems require more and more autonomous capabilities to improve efficiency and robustness. Rover mobility is one of the critical components that can directly affect mission success. Knowledge of the physical properties of the terrain surrounding a planetary exploration rover can be used to allow a rover system...
متن کاملSupporting Increased Autonomy for a Mars Rover
This paper presents an architecture and set of technology for supporting increased autonomy for a planetary rover. The MER rovers have outperformed all expectations by lasting over 1300 sols (or Martian days), which is an order of magnitude longer than their original mission goal. The longevity of these vehicles will have significant effects on future rover missions. Common objectives for futur...
متن کاملAutonomous Science Technologies for a Mars Rover
This paper presents an architecture and a set of technology for performing autonomous science and commanding for a planetary rover. The MER rovers have outperformed all expectations by lasting over 1100 sols (or Martian days), which is an order of magnitude longer than their original mission goal. The longevity of these vehicles will have significant effects on future mission goals, such as obj...
متن کاملSelf-Supervised Terrain Classification for Planetary Surface
Exploration Rovers 2 3 4 Christopher A. Brooks, Karl Iagnemma 5 Department of Mechanical Engineering 6 Massachusetts Institute of Technology 7 Cambridge, MA 02139 8 [email protected], [email protected] 9 10 Abstract 11 12 For future planetary exploration missions, improvements in autonomous rover 13 mobility have the potential to increase scientific data return by providing safe 14 access to geol...
متن کاملRock Modeling and Matching for Autonomous Mars Rover Localization
In Mars rover missions, a high degree of accuracy in localization of the rover and mapping of the surrounding terrain is of fundamental importance for safe rover navigation and for achievement of scientific and engineering goals. In the Mars Exploration Rover (MER) 2003 mission, ground image-based incremental bundle adjustment (BA) technology has been performed on Earth to correct rover positio...
متن کامل