PhD Student in Assured Autonomous Systems (Two open positions)

Autonomous Cyber-Physical Systems (ACPS) have great potential to improve our ways of life, increasing mobility, cutting costs, and saving lives. Considering the complexity of the environments these systems operate in, ACPS increasingly rely on Machine Learning (ML) to perform a variety of challenging tasks in perception, planning, and control. While indispensable for autonomy, ML models are unpredictable; unanticipated changes in the environment may cause a neural network to produce faulty outcomes that could endanger the safety of the system. To assure the safe and reliable operation of ACPS, we need to have a comprehensive understanding of the behavior of the underlying ML components and capture the conditions of the environment in which these components will be trusted to behave correctly.  

Project Description

The goal of this PhD project is to develop techniques and tools for the construction of assured learning-enabled ACPS with a focus on runtime assurance. In this project, you will develop simulation-based methods for the construction of runtime monitors that identify under what conditions an ML component may risk the safe operation of the system. The task of the monitor is to recognize these scenarios and, if necessary, force the system to run using trustable components. A key challenge in developing such monitors is to handle noisy or missing data. You will build on the latest advances in formal methods and statistical learning theory, to develop methods that allow for the construction of monitors with guarantees on their correctness and reliability. 



Information about the division and the department

The Group for Safe and Trustworthy Autonomous Reasoning (STAR) is part of the Computing Science (CS) division in the Department of Computer Science and Engineering (CSE). Led by Dr. Hazem Torfah, the group works at the intersection of cyber-physical systems, formal methods, and artificial intelligence. The goal of the group is to develop theoretical foundations and techniques for the construction of safe, reliable, and secure autonomous cyber-physical systems.  

We are currently focused on developing methods and tools for: runtime assurance, the development of runtime monitoring approaches for the safe operation of ACPS; explainability, methods for synthesizing interpretations for learning-enabled components; specification, the design of specification languages for capturing properties of ML-based systems. We are particularly interested in the application domains of autonomous driving and aviation and interact closely with leading international academic and industrial groups working in these fields. In STAR we seek to create a vibrant and collaborative environment where students and postdocs are supported in their pursuit of challenging research questions. 

The CSE department is a joint department at Chalmers University of Technology and the University of Gothenburg, with activities on two campuses in the city of Gothenburg. The department is divided into four divisions, and employs around 270 people from over 30 countries. Research in the department has a wide span, from theoretical foundations to applied systems development. We provide high quality education at the Bachelor's, Master's and graduate levels, offering over 120 courses each year. We also have extensive national and international collaborations with academia, industry and society.


Major responsibilities
The employment is a fixed-term position limited to a maximum of five years, four years of studies and up to one year of departmental work, distributed over the whole employment period. The major responsibilities for a PhD student position in the division include conducting doctoral research and coursework. By the end of the PhD, students will be able to identify novel research directions and design the appropriate computational experiments to answer key research questions. Students are expected to effectively communicate the results of their research verbally and in writing, and will receive specific training towards building these skills. This position also includes teaching at Chalmers' undergraduate level, or performing other teaching and departmental duties corresponding to 20% of working hours. Furthermore, this position is funded by the WASP program, and as such comes with an additional set of coursework requirements and career development opportunities (https://wasp-sweden.org/)..


Qualifications
To qualify as a PhD student, applicants must have (or about to obtain) a master's level degree corresponding to at least 240 higher education credits in Computer Science, Electrical Engineering, or a related field. Previous coursework in cyber-physical systems, formal verification, computational logic, AI, or statistical methods would be beneficial but not required. The applicant should have strong background in mathematical foundations of computer science and experience in Python programming. Previous experience in deep learning, reinforcement learning, or explainable AI is a merit, but not required.

The position requires excellent verbal and written communication skills in English. If you are not familiar with Swedish, you are encouraged, but not required, to learn it during your employment. Chalmers offers Swedish courses.

Contract terms
Full-time temporary employment. The position is limited to a maximum of five years.

We offer
Chalmers offers a cultivating and inspiring working environment in the coastal city of Gothenburg. 
Read more about working at Chalmers and our benefits for employees.

Chalmers aims to actively improve our gender balance. We work broadly with equality projects, for example the GENIE Initiative on gender equality for excellence. Equality and diversity are substantial foundations in all activities at Chalmers.

Application procedure


Please click on this link to apply and see instructions for the applcation procedure.