Detection of car abnormal vibration using machine learning

One difficulty of car-sharing services is trouble among users related to damage of the automobile. With the rapid growth of car-sharing services, demand exists for a new approach for automatic detection of the occurrence of damage during crewless operation. In this study, we propose a method for automatic detection of abnormal vibration that damages an automobile that occurs driving and stopping. After using multiple piezoelectric sensors installed in the automobile to acquire vibration, we applied supervised learning to extract features from vibrations and to detect abnormal vibrations. The proposed method achieved recall of 0.96 for stopped automobiles and recall of 0.82 for driving automobiles.