Speaker info
Elena Troubitsyna is a Full Professor in Computer Science, KTH – Royal Institute of Technology, Sweden, School of Electrical Engineering and Computer Science. She conducts research in the area of engineering dependable systems. In particular, Elena focuses on the design of dependability-assurance solutions for autonomous systems, including the models and algorithms for surveillance, diagnostics and adaptation.
By using rigorous mathematical modelling and verification techniques, Elena’s research contributes to the development of methods and technologies that enable autonomous systems with formally proven safety and reliability. Elena’s research also includes the design of AI algorithms that will empower systems to learn and adapt in real time. She has published over 150 peer-reviewed articles. Prof. Troubitsyna was a chair of several key conferences in the area of dependability. She has also been leading several research projects in the area of engineering dependable autonomous systems.
abstract
Autonomous systems such as self-driving cars, drones, various servo-robots are intended to operate without human intervention over prolonged time periods, perceive their operating environment, adapt to changes while pursuing the defined goals or generating the new ones. Both perception and planning rely on artificial intelligence (AI) and machine learning (ML). Often autonomous systems implement safety-critical functions, which might cause immediate or indirect safety hazards. However, ensuring safety of systems that employ AI and ML-based components is an open engineering challenge. In this talk, I will discuss some advances in ensuring safe autonomy as well as point out open research challenges.