The Optimization of Electric Vehicles Gearshift Controller Parameters Using Reinforcement Learning Algorithm

In order to improve the efficiency of gearshift controller parameter optimization and obtain a good control effect, this article proposes an method for electric vehicle transmission controller, selects dual clutch as research object that establishes 11-degree-of-freedom dynamics model feedforward-feedback model. The feedforward is chosen from target trajectory given by Legendre pseudo-spectral approach, feedback Gaussian kernel radial basis function neural network controller. performs Probabilistic Inference Learning Control (PILCO) reinforcement learning algorithm strategy matches actual conditions. By comparing how well master-slave moving disk can follow during various iterations, it verified requires only few experiments complete optimization, optimized Radial Basis Function Neural Network (RBFNN) has better effect results different iterations. Applying learned slope load circumstances yields data demonstrate all have obvious with robustness. Additionally, technique suggested in be used assist engineers technicians increasing effectiveness their Research Development.