A real-time recognition based drilling strategy for lunar exploration

• Published in: 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems pp. 2375 - 2380
• Main Authors: Qiquan Quan, Junyue Tang, Shengyuan Jiang, Zongquan Deng, Hongwei Guo, Yihui Tao
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2014
• Subjects: Accuracy; Force; Media; Moon; Rocks; Support vector machines; Torque
• ISSN: 2153-0858; 2153-0866
• DOI: 10.1109/IROS.2014.6942884

Drilling & coring is considered as an effective way to acquire deep sample on the moon. Since the lunar regolith environment in depth is unknown, sampling drill should be developed to adapt to the undetermined drilling medium on the moon. Once mechanical system of sampling robot was finished, control strategy is a key to realize the high-efficiency drilling process. Since composition of lunar regolith is complicated, it's not easy to evaluate all the physical parameters to judge the drilling difficulty level. This paper proposes a novel idea of lunar regolith drillability which is established on the rate of penetration under the given standard terms. Drillability is selected to describe the drilling difficulty level which can be identified online by use of pattern recognition method of SVM. Control algorithm tunes the drilling parameters to adapt to the recognized medium. Experiments are conducted to verify the drillability online recognition based intelligent control strategy can make sampling robot adapt to complicated drilling media.