EVOLVABLE ADAS: H-GQM S.M.A.R.T.E.S.T. APPROACH

Abstract

The introduction to the mass market of vehicles with an ADAS 3+ level of automation is expected in the early 2020s. Currently, the vast majority of automakers conduct research in this field, a fairly large number of prototypes, pre-production and production systems have already been demonstrated. ADAS (advanced driver assistance systems) are complex hardware & software systems, the feature of which is that the core hardware platform remains unchanged for one or even several generations of vehicles (5–7 years). At the same time, the system should be able to transform and evolve to correct errors and expand functionality, especially due to active development of sensory peripheral systems and software algorithms. The GQM methodology and its modifications are used to support the development process of complex systems and evaluate them. However, these methodologies are limited exclusively to software products. Also, authors of these methodologies are not addressing explicitly the issues of applying the GQM methodology for analyzing and tracking the process of evolution of complex technical systems. This paper presents HGQM (Hardware GQM) methodology for controllable evolution of complex automotive hardware & software systems. The H-GQM methodology is based on GQM and is aimed at hardwaresoftware systems with a monolithic hardware core, a modifiable software core and atomic peripherals. Entity harmonization process is described to prove the applicability of the GQM for software- and-hardware systems analysis. S.M.A.R.T.E.S.T goal-setting concept is proposed for choice of evolutionary goals. This concept is based on S.M.A.R.T. criteria for the setting objectives of business processes and extended with harmonization and evolvability restrictions. The formulation of the H-GQM plan framework is provided using ADAS as an example. Within the framework of the proposed methodology, an ADAS-specific scalable target template has been formed.