Paper Info
Title
A system identification based Oracle for control-CPS software fault localization
Abstract
Control-CPS software fault localization (SFL, aka bug localization) is of critical importance as bugs may cause major failures, even injuries/deaths. To locate the bugs in control-CPSs, SFL tools often demand many labeled ("correct"/"incorrect") source code execution traces as inputs. To label the correctness of these traces, we must judge the corresponding control-CPS physical trajectories' correctness. However, unlike discrete outputs, the boundaries between correct and incorrect physical trajectories are often vague. The mechanism (aka oracle) to judge the physical trajectories' correctness thus becomes a major challenge. So far, the ad hoc practice of "human oracles" is still widely used, whose qualities heavily depend on the human experts' expertise and availability. This paper proposes an oracle based on the well adopted autoregressive system identification (AR-SI). With proven success for controlling black-box physical systems, AR-SI is adapted by us to identify the buggy control-CPS as a black-box. We use this identification result as an oracle to judge the control-CPS's behaviors, and propose a methodology to prepare traces for control-CPS debugging. Comprehensive evaluations on classic control-CPSs with injected real-life and artificial bugs show that our proposed approach significantly outperforms the human oracle approach in SFL accuracy (recall) and latency, and in oracle false positive/negative rates. Our approach also helps discover a new real-life bug in a consumer-grade control-CPS.
Year
DOI
Venue
2019
10.1109/ICSE.2019.00029
Proceedings of the 41st International Conference on Software Engineering
Keywords
Field
DocType
cyber-physical system, debug, oracle, testing
Autoregressive model,Source code,Computer science,Physical system,Correctness,Oracle,Real-time computing,Cyber-physical system,Artificial intelligence,System identification,Machine learning,Debugging
Conference
ISSN
ISBN
Citations 
0270-5257
978-1-7281-0870-4
0
PageRank 
References 
Authors
0.34
39
6
Name
Order
Citations
PageRank
Zhijian He1443.12
yao chen2249.82
Enyan Huang300.34
Qixin Wang450339.57
Pei Yu5445.20
Haidong Yuan600.34