NSF has awarded (Award Nº 1545126) $4.6 million to a team led by UC Berkeley exploring human cyber-physical systems (h-CPS)—systems that operate in concert with human operators—with the aim of improving the interaction between humans, computers and the physical world. The research outcome of the project, called Verified Human Interfaces, Control, and Learning for Semi-Autonomous Systems, or VeHICaL, will have applications in emerging technologies such as semi-autonomous cars and autonomous aerial vehicles (drones).
The award was part of a total of $13 million NSF awarded to three five-year “Frontier” projects to advance cyber-physical systems (CPS). The other two projects are tackling monitoring and mitigating noise pollution in cities and quickly identifying and overcoming problems in manufacturing environments.
The VeHICaL project is grounded in a novel problem formulation that elucidates the unique requirements on h-CPS including not only traditional correctness properties on autonomous controllers but also quantitative requirements on the logic governing switching or sharing of control between human operator and autonomous controller, the user interface, privacy properties, etc. The project will make contributions along four thrusts:
- formalisms for modeling h-CPS;
- computational techniques for learning, verification, and control of h-CPS;
- design and validation of sensor and human-machine interfaces, and
- empirical evaluation in the domain of semi-autonomous vehicles.
VeHICaL will bring a conceptual shift of focus away from separately addressing the design of control systems and human-machine interaction and towards the joint co-design of human interfaces and control using common modeling formalisms and requirements on the entire system. This co-design approach is making novel intellectual contributions to the areas of formal methods, control theory, sensing and perception, cognitive science, and human-machine interfaces.
|Overview of the VeHICal process. Click to enlarge.|
The team, which includes colleagues from the California Institute of Technology and the University of North Carolina, Chapel Hill, will draw upon novel interdisciplinary research that combines ideas from several areas including formal methods, control theory, robotics and perception, cognitive science, machine learning, security and privacy, and human-machine interfaces.
The project will generate theory and tools for designers of human-cyber physical systems to create the next generation of verified intelligent systems that seamlessly collaborate with humans to perform complex tasks with provable guarantees on safety, privacy and performance.
VeHICaL aims to gain a deeper understanding of how humans and machines can collaborate to perform safety- and mission-critical tasks in society, such as semi-autonomous driving, shared control of drones, computer-assisted medicine and surgery, and advanced manufacturing, according to project leader Sanjit Seshia, a professor of electrical engineering and computer sciences at the University of California, Berkeley, and principal investigator of VeHICaL.
Our project aims to significantly impact the way humans collaborate and interact with automation, an area vital to the nation’s technological growth and societal well-being. As intelligent cyber-physical systems are deployed in critical sectors such as transportation, aerospace and healthcare, there is a pressing need to design for their interaction with humans so as to ensure that safety, security, privacy and performance objectives are met.—Sanjit Seshia
The researchers aim to demonstrate their work on practical testbeds and in collaboration with industry partners in a number of relevant domains, including the automotive sector. They will also provide recommendations to the agencies that regulate each type of system considered.
The ability to design complex systems that include both human and cyber-physical elements, especially in safety-critical applications, is very much in its infancy. This project will help expand the tools required to deliver on the promise of such systems.—co-principal investigator Richard Murray, a professor of control and dynamical systems and bioengineering at Caltech
Other team members are Ruzena Bajcsy, Tom Griffiths, Björn Hartmann, Shankar Sastry and Claire Tomlin, of UC Berkeley, and co-principal investigator Cynthia Sturton, of UNC.
Sanjit Seshia (2015) Verification by, for, and of Humans: Formal Methods for Cyber-Physical Systems and Beyond, ECE Department Colloquium, University of Illinois at Urbana-Champaign