Supervised traversability learning for robot navigation
2011 (English)In: 12th Annual Conference on Towards Autonomous Robotic Systems, TAROS 2011, Springer Berlin/Heidelberg, 2011, Vol. 6856 LNAI, 289-298 p.Conference paper (Refereed)
This work presents a machine learning method for terrain's traversability classification. Stereo vision is used to provide the depth map of the scene. Then, a v-disparity image calculation and processing step extracts suitable features about the scene's characteristics. The resulting data are used as input for the training of a support vector machine (SVM). The evaluation of the traversability classification is performed with a leave-one-out cross validation procedure applied on a test image data set. This data set includes manually labeled traversable and non-traversable scenes. The proposed method is able to classify the scene of further stereo image pairs as traversable or non-traversable, which is often the first step towards more advanced autonomous robot navigation behaviours.
Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2011. Vol. 6856 LNAI, 289-298 p.
, Lecture Notes in Computer Science, ISSN 03029743
machine learning, robot navigation, stereo vision, SVM, traversability classification, v-disparity image, Autonomous robot navigation, Data sets, Depth Map, Leave-one-out cross validations, Machine learning methods, Processing steps, Stereo image pairs, Test images, Traversability, Classification (of information), Learning systems, Navigation, Navigation systems, Robotics, Robots, Statistical tests, Support vector machines
Computer Vision and Robotics (Autonomous Systems)
IdentifiersURN: urn:nbn:se:kth:diva-50984DOI: 10.1007/978-3-642-23232-9_26ScopusID: 2-s2.0-80052803673Archive number: 978-364223231-2OAI: oai:DiVA.org:kth-50984DiVA: diva2:463181
12th Annual Conference on Towards Autonomous Robotic Systems, TAROS 2011. Sheffield. 31 August 2011 - 2 September 2011
QC 20111212. Language of Original Document: English2011-12-082011-12-082011-12-12Bibliographically approved