Steering feeling characterization by means of a hardware in the loop approach

Nowadays, the steering systems dramatically affect several aspects of passenger vehicles: performance, safety, driving pleasure and others suffer this influence. Additionally, differently from other systems, the driver’s perception of the vehicle is greatly possible thanks to the steering wheel; therefore, it explains the more and more interest in their development, having the human factor as the central point. Moreover, the growing diffusion of electric power steering systems, mainly for reasons of efficiency, has contributed to sustain this interest. Indeed, removing the hydraulic circuit in favour of an electric actuation, the energy losses are reduced and consequently the global efficiency of the vehicle increases. In addition, specific control logics can be deployed for different functioning conditions, allowing a perfect integration with ADAS – Advanced Driving Assistance Systems. Unfortunately, all these benefits are counterbalanced by a longer and more demanding process of tuning. Typical developing approaches that make use of test drives have been confined to the last phases of the verification, gradually introducing alternative methodologies, whose core are driving simulators. The reasons lie in the greater complexity, in the higher costs, in the lower level of repeatability and in the time-consuming process of the classic approaches. Driving simulators are complex systems created with the main purpose to exploit and combine the features of advanced software for the characterization of the vehicle dynamics, with the human factor. Typically, the steering behaviour is transmitted to the driver using a so-called feed-back unit, which is a mechatronic device capable of reproducing the resisting torque using an electric motor. To guarantee this result, an advanced steering model must be used because, although the apparent structural simplicity of the steering systems, the inertial and hysteresis effects complicate the modelling phase. In this context it is placed the proposed project, which aims at designing and realizing an experimental test bench for steering systems, capable to introduce a real steering unit in the loop of simulation. The main purpose of this apparatus is the reproduction of the force profiles generated by the tires in contact with the road surface at the tie rods, allowing also the rotation of the uprights around their vertical steering axis. In parallel, a specific steering model has been created, to define a reference point for comparisons to verify the actual benefits brought by the proposed test rig. Furthermore, it was born with a second aim: considering the difficulties in replacing a physical part of the steering rack installed on the test rig, a novel procedure was conceived to speed up the process. It makes use of the model to vary some steering features with a software procedure, which changes the request of tie rods forces without replacing any physical part. It represents a brand new approach for the state of the art.