The paper presents the activities ongoing in the framework of a two-year Italian National Research Project aimed at developing a distributed Driving and Steering (DaS) system for Electric Vehicles (EVs). The system is built up around an actuation module embedding a rotary-linear motor and including a power converter and a local Electronic Control Unit (ECU) to supply and control the motor. The actuation module is intended to make a wheel active so as to give it propelling and steering capabilities. The DaS system, in turn, includes a central ECU and a fieldbus to govern and network the system, and a power supply stage. The Project under consideration is focused on a DaS system with two active wheels for a light-duty EV.

The paper deals with some new results on the averaging theory approach for the analysis of feedback controlled pulsewidth modulation power electronic converters. It develops some considerations regarding the averaging approximation of time-variant differential systems proving an analytical relationship between the accuracy with which the solution of the averaged model converges to the solution of the original time-variant model (i.e., the switched model) and the switching frequency of the converters. This relation is valid both for stable and unstable systems. Moreover, this paper confirms, as corollary, that under suitable conditions (verified during the standard operations of the power electronic converters) the averaged model uniformly converges to the switched model.