Unsupervised classification of slip events for planetary exploration rovers. (October 2017)

• Record Type: Journal Article
• Authors: Bouguelia, Mohamed-Rafik; Gonzalez, Ramon; Iagnemma, Karl; Byttner, Stefan
• Is Part Of: Journal of terramechanics. Volume 73(2017:Oct.)
• Page Start: 95
• Page End: 106
• Publication Date: 2017-10
• Subjects: Unsupervised learning -- Clustering -- Data-driven modeling -- Slip -- MSL rover -- LATUV rover; Trafficability -- Periodicals; Praticabilité (Routes) -- Périodiques; Trafficability; Periodicals; 629.222
• DOI: 10.1016/j.jterra.2017.09.001
• Languages: English
• ISSNs: 0022-4898

Abstract

Highlights: An unsupervised method for detecting the slip level of planetary exploration rovers. It is based on aggregating signals into features, clustering, and Bayesian tracking. Results show improved accuracy, enabling real-time autonomous tracking of slip events. Abstract: This paper introduces an unsupervised method for the classification of discrete rovers' slip events based on proprioceptive signals. In particular, the method is able to automatically discover and track various degrees of slip (i.e. low slip, moderate slip, high slip). The proposed method is based on aggregating the data over time, since high level concepts, such as high and low slip, are concepts that are dependent on longer time perspectives. Different features and subsets of the data have been identified leading to a proper clustering, interpreting those clusters as initial models of the prospective concepts. Bayesian tracking has been used in order to continuously improve the parameters of these models, based on the new data. Two real datasets are used to validate the proposed approach in comparison to other known unsupervised and supervised machine learning methods. The first dataset is collected by a single-wheel testbed available at MIT. The second dataset was collected by means of a planetary exploration rover in real off-road conditions. Experiments prove that the proposed method is more accurate (up to 86% of accuracy vs. 80% for K-means) in discovering various levels of slip while being fully unsupervised (no need for hand-labeled data for training).