2019
DOI: 10.48550/arxiv.1911.01523
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Counterexample-Guided Synthesis of Perception Models and Control

Abstract: We consider the problem of synthesizing safe and robust controllers for real world robotic systems like autonomous vehicles, which rely on complex perception modules. We propose a counterexample-guided synthesis framework which iteratively learns perception models that enable finding safe control policies. We use counterexamples to extract information relevant for modeling the errors in perception modules. Such models then can be used to synthesize controllers robust to errors in perception. If the resulting p… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2021
2021
2022
2022

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(3 citation statements)
references
References 19 publications
0
3
0
Order By: Relevance
“…Furthermore, four papers study the use of models as the initial step for derivation of other artefacts (model-driven engineering). Examples are the development of big data ML software [109] and the incorporation of safe and robust control policies in ML models [70]. The other two from the same authors [198,199] target the derivation of testing frameworks for evaluating properties of an autonomous driving system with ML components.…”
Section: Software Engineering Models and Methods (38 Studies)mentioning
confidence: 99%
“…Furthermore, four papers study the use of models as the initial step for derivation of other artefacts (model-driven engineering). Examples are the development of big data ML software [109] and the incorporation of safe and robust control policies in ML models [70]. The other two from the same authors [198,199] target the derivation of testing frameworks for evaluating properties of an autonomous driving system with ML components.…”
Section: Software Engineering Models and Methods (38 Studies)mentioning
confidence: 99%
“…Furthermore, four papers study the use of models as the initial step for derivation of other artefacts (model-driven engineering). Examples are the development of big data ML software [108] and the incorporation of safe and robust control policies in ML models [68]. The other two from the same authors [196,197] target the derivation of testing frameworks for evaluating properties of an autonomous driving system with ML components.…”
Section: Software Engineering Models and Methods (38 Studies)mentioning
confidence: 99%
“…These adversarial scenarios are valuable for understanding the shortcomings of the controller at early design stages, which may be hard to expose by random simulations. Moreover, once found, these adversarial scenarios can be used to improve the design, e.g., see [19], [21], [43].…”
Section: Introductionmentioning
confidence: 99%