Estimation of Distance to Empty of Small Commercial EVs for BYOD (Bring Your Own Device) Application using EKF and ANN

The acronym BYOD stands for Bring Your Own Device. The Basic concept behind BYOD is to replace the instrument cluster of a small commercial electric vehicle with the driver’s mobile phone and thereby monitor vehicle parameters in the user’s device itself with the help of an android based application. This in turn leads to cost reduction in the long run. Different vehicle parameters like speed, odometer, State of Charge etc. is fetched directly from the IVN (In Vehicle network) using J1939 protocol and send to the android application in the user’s device through a secured link with the help of a CAN- Bluetooth Converter called ELM-327. However, the Distance to Empty of the battery of an EV is not directly available in the CAN. Therefore, this paper focuses on estimation of Distance to empty by using two methods i.e. Extended Kalman filter (EKF) and Artificial neural network (ANN) and compares them by estimating the root mean square error in predicting DTE.