){kind=avatar}
Biography
I am Shreyas Ramakrishna, a senior safety architect at NVIDIA Corporation. My work involves applying functional safety analysis as per the ISO 26262 safety standard for assuring the safety of the automotive hardware and software developed by NVIDIA. I received my Ph.D. in electrical engineering from Vanderbilt University, where I was advised by Professor Abhishek Dubey, and I worked at SCOPE Lab, which is located in Institute for Software Integrated Systems. My research focused on developing dynamic safety assurance techniques for autonomous cyber-physical systems. I designed a safety assurance framework that combines runtime recovery procedures along with machine learning-based assurance monitors (e.g., Out-of-Distribution detector, risk assessors) and design-time safety assurance case to assure the safety of an autonomous system during operation.
I’ve published over 15 articles and abstracts in renowned international journals like ACM Transactions on Cyber-physical systems, Elsevier’s Journal of system architecture and conferences like ICCPS, SEAMS, EMSOFT, Safecomp, ITSC, and DESTION. I am also serving as a reviewer for journals published in Elsevier, SAE International, PHM Society, and Springer.
Download my [CV] [Research Slides]
- Cyber-Physical Systems
- Autonomous Vehicles
- Machine Learning
- Safety Engineering
PhD in Electrical Engineering, 2022
Vanderbilt University, USA
Masters in Electrical Engineering and Computer Science, 2015
Technical Univerity Kaiserslautern, Germany
Bachelors in Electronics and Communication Engineering, 2012
Visvesvaraya Technological University, India
Experience
PHD Thesis
Advisor: Abhishek Dubey Rest of Thesis Committee: Janos Sztipanovits, Gabor Karsai, Xenofon Koutsoukos, Arun Ramamurthy, and Ayan Mukhopadhyay
Abstract: Cyber-Physical Systems (CPSs) are ubiquitous through our interactions with applications such as smart homes, medical devices, avionics, and automobiles. However, the ever-increasing complexity, domain interdependence, and dynamic nature of operations have raised safety concerns in using them for safety-critical applications. Typically, safety assurance techniques such as an assurance case are used at design time to design a safety argument supported by evidence intended to demonstrate that the system will satisfy its assurance guarantees. The argument is subject to certain assumptions about the behavior of components and the system’s operating environment. However, on deployment, the evolving nature of the system potentially invalidates the design-time assumptions, thereby defeating the safety argument. This problem of safety assurance is exacerbated by using data-driven Learning Enabled Components (e.g., Deep Neural Networks) to design autonomous CPS. Despite having performed well, the closed-world assumptions under which these components are trained often get invalidated when deployed in open-world scenarios (Out-of-Distribution data problem). The invalidation of these design-time assumptions could potentially increase the system’s risk of consequence at runtime. Therefore, there is a need to continuously monitor these assumptions at runtime and quantify the risk posed to the system. This is not possible by the existing design-time safety assurance techniques and requires a dynamic safety assurance approach. For this, the thesis proposes a dynamic safety assurance framework, which is a synergy of the design-time and runtime assurance approaches. This thesis also discusses the research challenges in implementing the framework components and provides a solution approach for each identified problem. The effectiveness of the proposed framework is demonstrated with two different autonomous CPS testbeds.
Download my [PHD Thesis]