Model-based testing of a vehicle instrument cluster for design validation using machine vision

Abstract

This paper presents an advanced testing system, combining model-based testing and machine vision technologies, for automated design validation of a vehicle instrument cluster. In the system, a hardware-in-the-loop (HIL) tester, supported by model-based approaches, simulates vehicle operations in real time and dynamically provides all essential signals to the instrument cluster under test. A machine vision system with advanced image processing algorithms is designed to inspect the visual displays. Experiments demonstrate that the system developed is accurate for measuring the pointer position, bar graph position, pointer angular velocity and indicator flash rate, and is highly robust for validating various functionalities including warning lights status, symbol and text displays. Moreover, the system developed greatly eases the task of tedious validation testing and makes onerous repeated tests possible.