TY - GEN
T1 - On the Effectiveness of Signal Rescaling in Hybrid System Falsification
AU - Zhang, Zhenya
AU - Lyu, Deyun
AU - Arcaini, Paolo
AU - Ma, Lei
AU - Hasuo, Ichiro
AU - Zhao, Jianjun
N1 - Funding Information:
This work is supported in part by JSPS KAKENHI Grant No. 20H04168, 19K24348, 19H04086, and JST-Mirai Program Grant No. JPMJMI18BB, Japan. Paolo Arcaini and Ichiro Hasuo are supported by ERATO HASUO Metamathematics for Systems Design Project (No. JPMJER1603), JST.
Publisher Copyright:
© 2021, Springer Nature Switzerland AG.
PY - 2021
Y1 - 2021
N2 - Hybrid system falsification employs stochastic optimization to search for counterexamples to a system specification in Signal Temporal Logic (STL), guided by quantitative STL robustness. The scale problem could arise when the STL formula is composed of sub-formulas concerning signals having different scales (e.g., speed [km/h] and rpm): the performance of falsification could be negatively affected because different scales can mask each other’s contribution to robustness. A natural solution consists in rescaling the signals to the same order of magnitude. In this paper, we investigate whether this “basic” approach is always effective, or better rescaling strategies could be devised. Experimental results show that basic rescaling is not always the best strategy, and sometimes “unbalanced” rescalings work better. We investigate the reasons of this, and we identify future research directions based on this observation.
AB - Hybrid system falsification employs stochastic optimization to search for counterexamples to a system specification in Signal Temporal Logic (STL), guided by quantitative STL robustness. The scale problem could arise when the STL formula is composed of sub-formulas concerning signals having different scales (e.g., speed [km/h] and rpm): the performance of falsification could be negatively affected because different scales can mask each other’s contribution to robustness. A natural solution consists in rescaling the signals to the same order of magnitude. In this paper, we investigate whether this “basic” approach is always effective, or better rescaling strategies could be devised. Experimental results show that basic rescaling is not always the best strategy, and sometimes “unbalanced” rescalings work better. We investigate the reasons of this, and we identify future research directions based on this observation.
UR - http://www.scopus.com/inward/record.url?scp=85111284410&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85111284410&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-76384-8_24
DO - 10.1007/978-3-030-76384-8_24
M3 - Conference contribution
AN - SCOPUS:85111284410
SN - 9783030763831
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 392
EP - 399
BT - NASA Formal Methods - 13th International Symposium, NFM 2021, Proceedings
A2 - Dutle, Aaron
A2 - Muñoz, César A.
A2 - Moscato, Mariano M.
A2 - Titolo, Laura
A2 - Perez, Ivan
PB - Springer Science and Business Media Deutschland GmbH
T2 - 13th International Symposium on NASA Formal Methods, NFM 2021
Y2 - 24 May 2021 through 28 May 2021
ER -