Online Monitoring of Aviation Software

Safety for Aviation Automation Systems

This work was generously supported by the National Science Foundation through grants CNS-1329341 and CNS-1836942.

As aircraft automation becomes increasingly sophisticated, new challenges arise in maintaining our national air transportation system’s exceptional safety record. It is particularly difficult to maintain high safety standards for sophisticated automation systems in modern manned and unmanned aircraft. Our lab is investigating bug-monitoring concepts for application to flight management systems (FMS) for automated aviation. Recent work in using machine learning for FMS bug detection has the potential to streamline the increasingly costly and time-consuming activity of verifying system safety.

CNS-1836942 focused on novel approaches to modeling software components of automation systems. To date, analysis of combined software and physical systems (also known as cyberphysical systems) has imposed severe constraints on programmers. This project has investigated methods for deriving models of code with the goal that those models can automatically be constructed from arbitrary programs. The results of this work have been published in several conferences and journals. Funding has supported two PhD students and a postdoctoral scholar, the last whom has recently entered academia as a professor at UCF.

Return to Main Lab Research Page >

Recent Publications Related to this Project:

C. Eniyoha and J. Rife (accepted). A tool for modeling cyber-physical system software, AIAA SciTech Forum 2020.

H. Huang (2019). Detecting Semantic Bugs in Autopilot Software by Classifying Anomalous Variables. Ph.D. Thesis, Dept. of Computer Science, Tufts University.

H. Huang, S.Z. Guyer, and J.H. Rife (submitted, 2019). Detecting semantic bugs in autopilot software by classifying anomalous variables, submitted to AIAA Journal of Aerospace Information Systems.

J. Larson, D. Gebre-Egziabher, and J.H. Rife (2019). Multivariate error overbounding with a Gaussian-pareto model, AIAA J. Aerospace Information Systems. doi: 10.2514/1.I010675

J. Larson, D. Gebre-Egziabher, and J.H. Rife (2019). Gaussian-pareto overbounding of DGNSS pseudoranges from CORS, NAVIGATION, 66(1), pp. 139-150, doi: 10.1002/navi.276.

J.H. Rife, H. Huang, and S.Z. Guyer (2019). Applying sensor integrity concepts to detect intermittent bugs in aviation software, accepted to NAVIGATION, 66(3), pp. 603-619, doi: 10.1002/navi.322. First published in Proc. ION Global Navigation Satellite Systems+ (ION GNSS+ 2018), Miami, Florida.

H. Huang, S.Z. Guyer, and J.H. Rife (2017). Improving run-time bug detection in CPS software using program slicing, Proc. IEEE-Cyber 2017, Waikiki Beach, HI, doi: 10.1109/CYBER.2017.8446584. PDF (Accepted).

H. Huang, S. Guyer, and J. Rife (2016). Applying machine learning for run-time bug detection in aviation software. Proc. AIAA Infotech @ Aerospace, AIAA Science and Technology Forum and Exposition 2016, San Diego, CA, doi: 10.2514/6.2016-0482.