Verified Human Interfaces, Control, and
Learning for Semi-Autonomous Systems
VeHICaL is an NSF Cyber-Physical Systems (CPS) Frontier project that
is developing the foundations of verified co-design of interfaces
and control for human cyber-physical systems (h-CPS) ---
cyber-physical systems that operate in concert with human
operators. VeHICaL aims to bring a formal approach to designing both
interfaces and control for h-CPS, with provable guarantees. The
VeHICaL approach is bringing 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,
security and privacy, and human-machine interfaces.
The foundational work being pursued in the VeHICaL project is being
validated in two application domains: semi-autonomous ground
vehicles that interact with human drivers, and semi-autonomous
aerial vehicles (drones) that interact with human operators.
Using a Driver's Eye Data to Predict Accident-Causing Drowsiness Levels.
Alyssa Byrnes and Cynthia Sturton.
The 21st IEEE International Conference on Intelligent Transportation Systems, November 2018.
Drowsy Driver Dataset.
HindSight: Enhancing Spatial Awareness by Sonifying Detected Objects in Real-Time 360-Degree Video.
Eldon Schoop, James Smith, and Bjoern Hartmann. 2018.
In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (CHI '18)
Vazquez-Chanlatte, Marcell, et al. "Logical Clustering and
Learning for Time-Series Data." International Conference on
Computer Aided Verification. Springer, Cham, 2017.
Active Preference-Based Learning of Reward Functions
Dorsa Sadigh, Anca Dragan, S. Shankar Sastry, Sanjit A. Seshia
Robotics: Science and Systems (RSS), July 2017.
Stochastic Predictive Freeway Ramp Metering from Signal
Temporal Logic Specifications Negar Mehr, Dorsa Sadigh, Roberto
Horowitz, S. Shankar Sastry, Sanjit A. Seshia 2017 American Control
Conference (ACC), May 2017.
Information Gathering Actions over Human Internal State
Dorsa Sadigh, S. Shankar Sastry, Sanjit A. Seshia, Anca Dragan
IEEE/RSJ International Conference on Intelligent Robots and Systems
(IROS), October 2016.
Towards Trustworthy Automation: User Interfaces that
Convey Internal and External Awareness Tara Rezvani, Katherine
Driggs-Campbell, Dorsa Sadigh, S. Shankar Sastry, Sanjit A. Seshia,
Ruzena Bajcsy IEEE Intelligent Transportation Systems Conference
(ITSC), November 2016.
Safe and Interactive Autonomy: Control, Learning, and
Verification. Ph.D. Dissertation, EECS Department, University
of California, Berkeley, 2017.
Systematic Testing of Convolutional Neural Networks for
Autonomous Driving T. Dreossi, S. Ghosh, A. Sangiovanni-Vincentelli,
S. A. Seshia Reliable Machine Learning in the Wild (RMLW 2017)
Compositional Falsification of Cyber-Physical Systems with Machine Learning Components
T. Dreossi, A. Donzé, S. A. Seshia
NASA Formal Methods (NFM 2017)
The Visual-Acoustic Vehicle Dataset is a set of multimodal data collected from a Lincoln MKS car including vehicle sensors information, 360º image data collected from eight cameras on top of the car, and LiDAR information.
Visual-Acoustic Vehicle Dataset.