On provably correct decision-making for automated driving

The introduction of driving automation in road vehicles can potentially reduce road traffic crashes and significantly improve road safety. Automation in road vehicles also brings several other benefits such as the possibility to provide independent mobility for people who cannot and/or should not drive. Many different hardware and software components (e.g. sensing, decision-making, actuation, and control) interact to solve the autonomous driving task. Correctness of such automated driving systems is crucial as incorrect behaviour may have catastrophic consequences. Autonomous vehicles operate in complex and dynamic environments, which requires decision-making and planning at different levels. The aim of such decision-making components in these systems is to make safe decisions at all times. The challenge of safety verification of these systems is crucial for the commercial deployment of full autonomy in vehicles. Testing for safety is expensive, impractical, and can never guarantee the absence of errors. In contrast, formal methods, which are techniques that use rigorous mathematical models to build hardware and software systems can provide a mathematical proof of the correctness of the system. The focus of this thesis is to address some of the challenges in the safety verification of decision-making in automated driving systems. A central question here is how to establish formal verification as an efficient tool for automated driving software development.


  • English

Media Info

  • Pagination: 142p

Subject/Index Terms

Filing Info

  • Accession Number: 01781135
  • Record Type: Publication
  • Source Agency: Swedish National Road and Transport Research Institute (VTI)
  • Files: ITRD, VTI
  • Created Date: Sep 1 2021 2:26PM