Dynamic Testing for the Verification of Autonomous Systems2018-05-18T15:41:56+00:00

Verification Futures 2018

Conference:Verification Futures 2018 (click here to see full programme)
Speaker:Anas Shrinah & Nyasha Masamba, University of Bristol
Presentation Title:Dynamic Testing for the Verification of Autonomous Systems
Abstract:The development of autonomous intelligent systems is bringing new intellectual challenges to the verification field. The Dynamic Testing Workshop identified challenges such as safety assurance, definition of coverage metrics, regulation concerns and utilisation of artificial intelligence in the verification process. Addressing these challenges provides the basis for an autonomous system testing framework, ultimately leading to safer and more trustworthy autonomous systems.

Top 3 Challenges:

  1. Metrics for verification completeness and quality – How do we know we are done, and what assurances can we give?
  2. Scalability – Exhaustive simulation of autonomous systems is not tractable. How can we leverage artificial intelligence with existing verification techniques to make them scale better?
  3. System integrity – How can we continuously ensure that an autonomous system (which is constantly adapting) is operating as safely and capably as it initially did, considering that this depends on several factors such as incoming data quality, in-field learning and various emergent properties?
Speaker Bio:Anas Shrinah has a first-class honours BEng in Computer and Automation Engineering. He spent two years working on different automation projects in industry, followed by five years working as instrumentation and control engineer for Petrofac in the oil and gas industry. In 2017, Anas was awarded an MSc in Robotics with Distinction from the University of Bristol. Currently, he is a PhD student in the Trustworthy Systems Laboratory at the University of Bristol. Anas’ research is focused on applying verification techniques to planning-based industrial processes, sponsored by Schlumberger Gould Research.

Nyasha Masamba is a research student at the University of Bristol. After obtaining his BSc Computing degree, he spent six years in industry working in software engineering and analytics roles. Thereafter, Nyasha gained his MSc in Advanced Computing (Machine Learning, Data Mining and High Performance Computing) during which he got involved in robotics verification. Nyasha is currently part of the Trustworthy Systems Laboratory. In his PhD he collaborates with Infineon Technologies, investigating the use of artificial intelligence to further automate coverage driven test generation.