Data Mining Approach for the Identification of the Intelligent Gear Decision Strategy for the Stepped Automatic Transmission

The primary goal of automatic transmission vehicle design and optimization is to ensure that drivers can operate automatic transmission vehicles with the same driving experience as skilled drivers driving manual transmission vehicles. However, the existing gear decision strategies of stepped automatic transmission vehicles have poor adaptability to different driving intentions and complex driving environments. When operating manual transmission vehicles, skilled drivers can make intelligent gear decision to satisfy their driving intentions and adapt to different driving environments. Therefore, this study's objective is to establish the design and formulation process for the identification of the intelligent gear decision strategy for the stepped automatic transmission that adapts to different driving intentions and complex driving environments. This is achieved using the data mining method to identify skilled drivers’ gear decision strategy from their driving data. To this end, the driving data of skilled drivers operating manual transmission vehicles are collected. The shift boundary points of each gear under the target driving condition are obtained through data preprocessing, data cleaning, outlier processing, and boundary points extraction. Thereafter, the gear decision strategy of skilled drivers is established by fitting the shift boundary points. Taking the flat-straight road condition as an example, the gear decision strategy is established and experimental comparisons are made to verify the effectiveness of proposed gear decision design method.