Authors: Bin Shi; Jie Hu; Li Xu
Addresses: College of Electrical Engineering, Zhejiang University, Zheda road #38, Hangzhou 310027, China ' College of Electrical Engineering, Zhejiang University, Zheda road #38, Hangzhou 310027, China ' College of Electrical Engineering, Zhejiang University, Zheda road #38, Hangzhou 310027, China
Abstract: Gear shifting strategy may affect fuel economy and driving ability, therefore, gear prediction algorithm becomes an important research topic for automatic transmission. In this work, we propose to predict the gear changes based on historical vehicle test data and driver's driving style. Firstly, a set of vehicle data that affect gear shifting are obtained using a dimension reduction algorithm, e.g., principal component analysis. This will greatly reduce the computational complexity of the system. Secondly, the dimension reduced data are applied to obtain the personalised transmission gear model (PTGM). This is accomplished by using the locally designed neural network, i.e., CMAC in this work. Finally, the driver style evaluation index is applied here as an auxiliary parameter to achieve the accuracy of the predictions. Simulations are conducted to verify the effectiveness of the proposed scheme.
Keywords: personalised transmission gear modelling; PTGM; vehicle test data; principal component analysis; PCA; CMAC; driving style; vehicle speed; throttle position; neural networks; automatic transmission; gear prediction; personalisation; gear shifting strategy; gear changes; simulation.
International Journal of Vehicle Systems Modelling and Testing, 2015 Vol.10 No.4, pp.356 - 365
Available online: 12 Nov 2015 *Full-text access for editors Access for subscribers Purchase this article Comment on this article