The traditional support vector machines only deals with the binary classification. It has difficulty in solving the multi-class classification problem like the shift decision for the automatic transmission of the engineering vehicle. A shift decision algorithm that based on SVM-binary tree was presented. This method distributed classifier to nodes that constituted multi-class SVM. The number of SVM classifier and duplicated training samples could be reduced. The optimal shifting gear could be decided by the proposed approach, and the requirement of the engineering vehicle to the automatic shifting could be satisfied in time and accurately. The experiment showed that the support vector machines based on binary tree achieved better results than RBF neural network with genetics.
參考文獻(xiàn)
相似文獻(xiàn)
引證文獻(xiàn)
引用本文
韓順杰,趙丁選.多類分類SVM在工程車輛自動(dòng)變速擋位決策中的應(yīng)用[J].農(nóng)業(yè)機(jī)械學(xué)報(bào),2007,38(2):10-12.[J]. Transactions of the Chinese Society for Agricultural Machinery,2007,38(2):10-12.