Event-Based Electric Vehicle Mass and Grade Estimation

Abstract

This work investigates an event-based electric vehicle mass and grade estimation using a Recursive Least Squares (RSL) with variable forgetting factors method. Given the vehicle speed and electric power consumption, we proposed a two-layer identification architecture in which the first layer provides acceleration and cruise motion periods, whereas the second layer is responsible for mass and grade parameter estimations. The forgetting factors are updated based on the vehicle acceleration values. The proposed method does not require torque measurements from the propulsion system. Therefore, it can be used for different type of vehicles. The preliminary comparative study suggests that the proposed method is efficient and can provide satisfactory results even in presence of noisy measurements.