VAST

Visual And Spectral Terrain classification in unstructured multi-class environments
(VAST)

Near Infrared spectroscopy (NIRS) is a powerful technique which allow for non-contact material determination of multitude of multiple materials using only reflected light. In this research we apply NIRS to build a dataset of high-spectral resolution reflectance measurements, and high-resolution RGB images captured from in front of a rover's wheels towards the goal of predicting wheel slippage and classifying the current terrain.