Timing for Cyber-Physical Systems

 

TIME IN CYBER-PHYSICAL SYSTEMS

Time is a foundational aspect of Cyber-Physical Systems (CPS). Correct timing of system events is critical to optimize responsiveness to the environment, in terms of timeliness, accuracy, and precision in the knowledge, measurement, prediction, and control of CPS behavior (DAC Special Session 2016).

In order to design more resilient and reliable CPS, first and foremost, there should be a way to specify the timing constraints that a constructed Cyber-Physical System must meet. Only then, we can seek systematic approaches to check if all timing constraints are being met, and develop correct-by-construction methodologies. In this regard, we have developed a logic, Timestamp Temporal Logic (TTL) to specify the timing constraints on a distributed CPS (TTL-EMSOFT-TECS-2017). Designers can specify the timing requirement that a CPS must satisfy in a succinct and intuitive manner in TTL. For example, they can express that some two events on two different parts of the system must occur within 1 millisecond of each other. Further, we designed an FPGA-based testbed that can hook up to a CPS and take in these timing constraints specified in TTL and verify if the timing constraints are being met (CPS Timing Testbed, ReConfig 2015). One of the great features of using TTL to express timing constraints is that the time monitoring logic becomes very simple. TTL logic does not need to compute whether the constraint is being met at each and every instance of time but it re-evaluates a constraint only when there is an event that can affect the outcome. This enables our approach, TMA to perform online timing monitoring of CPS (TMA-DAC 2018) for less required computation and resources. Furthermore, we have come up with the minimum design parameters of the timing CPS that are required to enable testing the timing of CPS. For example, a system that is sampling at milliseconds level cannot test a timing constraint that has a requirement of accuracy to the level of microseconds (Testbed-DAC-2017).

We have built several CPS applications to test the need and effectiveness of our approaches. We have built a i) flying paster – a system needed in the printing press to continue the paper feed from another roller when the current roller finishes. ii) We have been able to align the phase of different motors up to milliseconds even when they are connected through the internet. iii) Synchronize the image capture time from different cameras, so that a 3D image reconstruction can be done with minimal blurring, and iv) developed a time-sensitive traffic intersection design for autonomous vehicles (Crossroads-DAC-2017).

 

In the future, we are interested in expanding our vision to build a Timing Health Monitoring and Reasoning System, (THMRS). THMRS would be able to, not only test the timing constraints but will be able to argue about them also. For example, it will be able to do some root cause analysis. If a timing constraint of a safety-critical CPS fails, that should not mean that the system cannot operate. It may be possible to reason and guarantee that the system can operate at lower levels of performance by reconfiguring the system. For example, if the rotating LIDAR motor slows down on an autonomous car due to some reason, then the timing constraint for object detection and braking will no longer be met. However, it should certainly be possible to operate safely at lower speeds. Therefore, graceful degradation of performance of CPS will be possible using THMRS. For CPS design also, a reasoning system could be helpful. For example, a complex timing constraint can be proven to be met, if few simple timing constraints are met. If these things can be found out, then designers can focus only on meeting the simple timing constraints. Furthermore, a timing reasoning system will bring us closer to our dream of designing systems that are correct by construction. The idea here is that if the system can be divided into subcomponents, and each of them has some timing constraints, then, if all the parts can be built so that they meet the timing constraints, then the built system will be guaranteed to meet the timing constraints.

 

Our Publications
Videos
Software