This website requires JavaScript.

Rethinking Certification for Higher Trust and Ethical Safeguarding of Autonomous Systems

Dasa KusnirakovaBarbora Buhnova
Mar 2023
With the increasing complexity of software permeating critical domains suchas autonomous driving, new challenges are emerging in the ways the engineeringof these systems needs to be rethought. Autonomous driving is expected tocontinue gradually overtaking all critical driving functions, which is addingto the complexity of the certification of autonomous driving systems. As aresponse, certification authorities have already started introducing strategiesfor the certification of autonomous vehicles and their software. But even withthese new approaches, the certification procedures are not fully catching upwith the dynamism and unpredictability of future autonomous systems, and thusmay not necessarily guarantee compliance with all requirements imposed on thesesystems. In this paper, we identified a number of issues with the proposedcertification strategies, which may impact the systems substantially. Forinstance, we emphasize the lack of adequate reflection on software changesoccurring in constantly changing systems, or low support for systems'cooperation needed for the management of coordinated moves. Other shortcomingsconcern the narrow focus of the awarded certification by neglecting aspectssuch as the ethical behavior of autonomous software systems. The contributionof this paper is threefold. First, we discuss the motivation for the need tomodify the current certification processes for autonomous driving systems.Second, we analyze current international standards used in the certificationprocesses towards requirements derived from the requirements laid on dynamicsoftware ecosystems and autonomous systems themselves. Third, we outline aconcept for incorporating the missing parts into the certification procedure.