User-Centered Development of a Visual Exploration System for In-Car Communication

Modern premium automobiles are equipped with an increasing number of Electronic Control Units (ECUs). These ECUs are interconnected and form a complex network to provide a wide range of advanced vehicle functionality. Analyzing the flow of messages in this network and tracking down problems has become a major challenge for automotive engineers. By observing their working practices, we found that the tools they currently use are mostly text-based and largely fail to provide correlations among the enormous amount of data. We established requirements for a more appropriate (visual) tool set. We followed a user-centered approach to design several visualizations for in-car communication processes, each with a clear purpose and application scenario. Then we used low-fidelity prototypes to evaluate our ideas and to identify the “working” designs. Based on this selection, we finally implemented a prototype and conducted an expert evaluation which revealed the emergence of a novel mental model for thinking about and discussing in-car communication processes.