Terrain Classification with Crawling Robot Using Long Short-Term Memory Network
Terrain classification is a crucial feature for mobile robots operating across multiple terrains. One way to learn a terrain classifier is to use a stream of labeled proprioceptive data recorded during a terrain traversal. In this paper, we propose a new terrain classifier that combines a feature extraction from a data stream with the long short-term memory (LSTM) network. Features are extracted from the information-sparse data stream by applying a sliding window computing three central moments. The feature sequence is continuously classified by the LSTM network into multiple terrain classes. Furthermore, a modified bagging method is used to deal with a limited and unbalanced training set. In comparison to the previous work on terrain classifiers for a hexapod crawling robot using only servo-drive feedback, the proposed classifier provides continuous classification with the F1 score up to 0.88, and thus provide better results than SVM classifier learned on the same input data.
KeywordsOnline classification Proprioception Recurrent neural networks
The presented work has been supported by the Czech Science Foundation (GAČR) under research project No. 18-18858S. The support of grant No. SGS16/235/OHK3/3T/13 to Rudolf Szadkowski is also gratefully acknowledged. Access to computing and storage facilities owned by parties and projects contributing to the National Grid Infrastructure MetaCentrum provided under the programme “Projects of Large Research, Development, and Innovations Infrastructures” (CESNET LM2015042), is greatly appreciated.
- 1.Bartoszyk, S., Kasprzak, P., Belter, D.: Terrain-aware motion planning for a walking robot. In: 2017 11th International Workshop on Robot Motion and Control (RoMoCo), pp. 29–34 (2017)Google Scholar
- 2.Best, G., Moghadam, P., Kottege, N., Kleeman, L.: Terrain classification using a hexapod robot. In: Australasian Conference on Robotics and Automation (2013)Google Scholar
- 5.Gers, F.: Long short-term memory in recurrent neural networks. Unpublished Ph.D. dissertation, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland (2001)Google Scholar
- 8.Mrva, J., Faigl, J.: Feature extraction for terrain classification with crawling robots. Inf. Technol. Appl. Theory 1422, 179–185 (2015)Google Scholar
- 9.Mrva, J., Faigl, J.: Tactile sensing with servo drives feedback only for blind hexapod walking robot. In: 10th International Workshop on Robot Motion and Control (RoMoCo), pp. 240–245 (2015)Google Scholar
- 12.Otte, S., Weiss, C., Scherer, T., Zell, A.: Recurrent neural networks for fast and robust vibration-based ground classification on mobile robots. In: IEEE International Conference on Robotics and Automation (ICRA), pp. 5603–5608 (2016)Google Scholar
- 13.Rebula, J.R., Neuhaus, P.D., Bonnlander, B.V., Johnson, M.J., Pratt, J.E.: A controller for the littledog quadruped walking on rough terrain. In: IEEE International Conference on Robotics and Automation (ICRA), pp. 1467–1473 (2007)Google Scholar
- 14.Sasaki, Y., et al.: The truth of the F-measure. Teach. Tutor. Mater 1(5), 1–5 (2007)Google Scholar
- 16.Tieleman, T., Hinton, G.: Lecture 6.5-rmsprop: divide the gradient by a running average of its recent magnitude. COURSERA Neural Netw. Mach. Learn. 4(2), 26–31 (2012)Google Scholar
- 18.Walas, K., Kanoulas, D., Kryczka, P.: Terrain classification and locomotion parameters adaptation for humanoid robots using force/torque sensing. In: IEEE-RAS 16th International Conference on Humanoid Robots, pp. 133–140 (2016)Google Scholar
- 19.Walas, K., Nowicki, M.: Terrain classification using laser range finder. In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 5003–5009 (2014)Google Scholar