University of Pennsylvania - PRECISE

• Compositional Real-Time Scheduling Framework
  Real-time systems are ones in which correctness depends not only on logical correctness but also on timeliness. For complex real-time systems, it is necessary to develop them as hierarchical systems with real-time components with resource interfaces. This project is to develop abstraction and composition techniques for resource interface while ensuring compositional analysis.

• High-Confidence Medical Device Software Systems
  The development and production of medical device software and systems is a critical issue as medical device software is increasingly sophisticated and medical devices are networked. Of particular importance is how to ensure such medical device systems are safe.

• MaC (Run-Time Monitoring and Checking)
  As software becomes large and complex, it becomes more difficult to test or verify the correctness of a system. Continuously monitoring of a running system is a complementary approach to increase the assurance of correct execution, which is known as runtime verification. We have been developed a Monitoring and Checking (MaC) framework to monitor and check running systems against a formal requirement specification.

• QTM (Quantitative Trust Management)
  Cyber Physical Systems (CPS) consists of many interacting heterogeneous components that are distributed and evolve over time. As our lives depend on CPS, trustiness of CPS is important to ensure safe and correct operation of CPS. Quantitative Trust Management (QTM) provides a dynamic interpretation of authorization policies for access control decisions based on upon evolving reputations of the entities involved.

• Embedded Virtual Machines for Robust Wireless Control
  To address actuation in closed-loop wireless control systems there is a strong need to re-think the communication architectures and protocols for reliability, coordination and control. EVM is the distributed runtime system that maintains functional (e.g. control law) and para-functional invariants (e.g. timeliness, fault-tolerance) across physical node boundaries given an unreliable wireless substrate.

• Platforms for High-Confidence Networked Medical Systems
  Designing bug-free medical device software is difficult, especially in complex on-body and implantable devices that may be used in unanticipated contexts. This project is focused on the end-to-end systems software for medical devices and includes a disposable on-body Health-Strip, adaptive RTOS design for runtime programmable control and long-term context-based data interpretation.

• Real-Time Parallel Computing
  By 2020, embedded processors will sport 4,096 cores, server CPUs will have 512 cores and desktop chips could use 128 cores. This project’s goal is to devise a class of algorithms and complementary multi-core architectures for mission and safety-critical environments where unreliable software is not an option.

• Networked Automotive Cyber-Physical Systems
  The automobile of the future will be networked for network-based active safety and real-time traffic congestion probing and prediction. This project’s goal is to develop wireless protocols for vehicle-to-vehicle safety, real-time traffic congestion prediction and coordinated fleet driving.