Section: New Results

Facing ADAS validation complexity with usage oriented testing

Participant : Guy Fayolle.

Validating Advanced Driver Assistance Systems (ADAS) is a strategic issue, since such systems are becoming increasingly widespread in the automotive field.

But, ADAS validation is a complex issue, particularly for camera based systems, because these functions maybe facing a very high number of situations that can be considered as infinite. Building at a low cost level a sufficiently detailed campaign is thus very difficult.

The COVADEC project (type FUI/FEDER 15), which was aiming to provide methods and techniques to deal with these problems, was actually successfully completed in May 2017. The test cases automatic generation relies on a Model Based Testing (MBT) approach. The tool used for MBT is the software MaTeLo (Markov Test Logic), developed by the company All4Tec. MaTeLo is an MBT tool, which makes it possible to build a model of the expected behavior of the system under test and then to generate, from this model, a set of test cases suitable for particular needs. MaTeLo is based on Markov chains, and, for non-deterministic generation of test cases, uses the Monte Carlo methods. To cope with the inherent combinatorial explosion, we couple the graph generated by MaTeLo to an ad hoc random scan Gibbs sampler (RSGS), which converges at geometric speed to the target distribution. Thanks to these test acceleration techniques, MaTeLo also makes it possible to obtain a maximal coverage of system validation by using a minimum number of test cases. As a consequence, the number of driving kilometers needed to validate an ADAS is substantially reduced, see [53] and [54]. These methods do interest the French manufacturer Groupe PSA, who wishes to establish a contractual collaboration involving Armines-MINES ParisTech.