Tufano, Francesco (2023) Automotive MiL/HiL Simulations: Methodologies for Check and Validation of Systems for Autonomous Driving. [Tesi di dottorato]

	Text Tufano_Francesco_35_COMPLETO.pdf Visibile a [TBR] Repository staff only Download (15MB)
Preview	Text Tufano_Francesco_35_PARZIALE.pdf Download (6MB) | Preview

Collection description

The innovation in the automotive field due to the Fourth Industrial Revolution pushed towards the integration in vehicles of mechatronic systems. Indeed, while safety systems, such as the electronic stability program, have pioneered the automation of cars, recent advances in the field of electronics have fuelled an increase in this trend. So intelligent transportation systems, advanced driver assistance systems, vehicle handling stability and active safety have increasingly been promoted. However, their implementation depends on accurate vehicle dynamics state information, including those such as the vehicle sideslip angle and the tire inflation pressure, that cannot be measured directly for technical and economical reasons. The thesis is dedicated to the development and testing of algorithms for the estimation of these variables. In particular, research activities have concerned the testing of a tire pressure estimation scheme in Hardware-In-the-Loop (HIL) environment, and the development, in Model-In-the-Loop (MIL) environment, of new algorithms for the estimation of vehicle sideslip angle and tire inflation pressure, respectively. Concerning the tire inflation pressure, several estimation schemes have been proposed to improve accuracy of the indirect Tire Pressure Monitoring Systems (iTPMS). The most common iTPMS actually used in many cars present on the market is based on wheel angular speed signal analysis. There are currently few studies that focus on investigating the possibility to execute the iTPMS integration tests in HIL environment. In particular, modelling and parameterization of simulation platform suitable for testing the iTPMS in HIL environment, especially for wheel speed signal frequency analysis, is a topic worthy of research, that have been addressed in the Thesis. The thesis also deals with the development of new estimation schemes for vehicle sideslip angle and tire inflation pressure. Indeed, even if estimation of these variables has been widely studied, moving to next-generation vehicle control and future autonomous driving require further advances in vehicle dynamics state estimation. Specifically, an innovative algorithm for vehicle sideslip estimation is proposed to deal with critical driving conditions of non-trivial scenarios. Based on the Interacting Multiple Model (IMM) filters, it not requires tire-road friction coefficient knowledge to give a reliable estimation of the sideslip angle, also when vehicle drives under critical road surface conditions. The IMM approach has been also adopted in a new estimation scheme to deal with the estimation of the tire inflation pressure on roads with highly uneven surface. The advantage of the presented algorithms is that they work only with CAN-BUS data coming from the sensors available on ordinary vehicles. The algorithms have been tested rigorously under all possible conditions in MIL simulation environment. To this purpose, a high-fidelity vehicle dynamics simulation platform has been developed, whose modelling ad validation is described in the Thesis.

Downloads

Actions (login required)

View Item