A nonlinear MPC based motion cueing strategy for a high performance driving simulator with active seat

Abstract

Dynamic driving simulators are nowadays a common tool in the automotive industry. Issues related to automated driving topic are moving the target user of these tools from professional drivers to non-professional ones. Unfortunately, these latter typically suffer more the virtual environment, mainly due to the unavoidable lack of low frequency sustained accelerations. Active Seat (AS) with integrated Active Belts (AB) is a promising add-on to dynamic platform to make the virtual experience more realistic. Its usage in dynamic simulators requires ad-hoc strategies to coordinate pressure/tension of AS/AB with platform motion. In the framework of Nonlinear Model Predictive Control (NMPC) we propose a Multi-Sensory Cueing Algorithm (MSCA) that computes coordinated references for the platform and for the AS/AB systems. A dynamic model of a seated human body is used to relate forces and contact pressures acting on the driver. Simulative results are reported, showing the advantages in using the approach comparing a compact simulator and a giant one.