Improving Vehicle Diagnostics through Wireless Data Collection and Statistical Analysis

Abstract

Collection and analysis of diagnostic data from electronic control units is of paramount importance in the automotive industry, both from a life cycle support perspective, and as a tool in the product development. For pre-series test vehicles, access to diagnostic data is crucial in order to be able to track problems as early as possible in the development process, preventing serious faults to pass undetected into production vehicles. However, systematic collection of diagnostic data from test vehicles is complicated by the fact that pre-series vehicles are frequently unavailable for diagnostic read-outs, while away on test expeditions in rural areas, or being operated by taxi companies, or being otherwise inaccessible. In response, we propose a system for wireless read-out of diagnostic data, based on an embedded system installed in the test vehicles and a server component at the automotive company. Moreover, we present a number of computer-based tools for statistical analysis of diagnostic trouble codes, to aid engineers in processing the potentially very large amount of diagnostic data generated by the automated collection system. We strongly believe that wireless diagnostic collection systems, in combination with statistical analysis tools, will be a key component in meeting future demands on reduced development time, while the complexity of automotive electronics continue to grow. Initial results from our pilot trials with the tools described in this paper support this claim.