A fuzzy Q-learning based assisted power management method for comfortable riding of pedelec

Abstract

In this study, a fuzzy logic controller with parameters tuned by Q-learning is proposed for the assisted power management of a pedelec. The pedelec is a human-electric hybrid vehicle driven by mainly the rider's pedal force with the assisted power from the electric motor. The proposed assisted power management (APM) method adaptively provides an appropriate assisted power (action) according to the environmental changes via the fuzzy inference coordinated Q-learning, i.e. fuzzy Q-learning. Simulations of the proposed method, hereafter abbreviated as FQLAPM, on a pedelec are performed, and the results exhibit better performance in comparing with other existent assisted power methods.