Abstract
We address one of the key challenges in assuring safety of autonomous cyber-physical systems that rely on learning-enabled classification within their environmental perception: How can we achieve confidence in the perception chain, especially when dealing with percepts safe-guarding critical manoeuvres? We present a methodology which allows to mathematically prove that the risk of misevaluating a safety-critical guard conditions referring to environmental artefacts can be bounded to a considerably lower frequency than the risk of individual misclassifications, and can thereby be adjusted to a value less than a given level of societally accepted risk.
M. Fränzle—Supported by the State of Lower Saxony within the Zukunftslabor Mobilität as well as by Deutsche Forschungsgemeinschaft under grant no. DFG FR 2715/5-1.
M. Swaminathan—Contribution while employed at the University of Oldenburg.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Baig, Q., Perrollaz, M., Laugier, C.: A robust motion detection technique for dynamic environment monitoring: a framework for grid-based monitoring of the dynamic environment. IEEE Robot. Automat. Mag. 21(1), 40–48 (2014)
Dreossi, T., Donzé, A., Seshia, S.A.: Compositional falsification of cyber-physical systems with machine learning components. In: Barrett, C., Davies, M., Kahsai, T. (eds.) NFM 2017. LNCS, vol. 10227, pp. 357–372. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-57288-8_26
Fawcett, T.: ROC graphs: notes and practical considerations for researchers. Mach. Learn. 31(1), 1–38 (2004)
Fawcett, T.: An introduction to ROC analysis. Pattern Recogn. Lett. 27(8), 861–874 (2006)
Galar, M., Fernandez, A., Barrenechea, E., Bustince, H., Herrera, F.: A review on ensembles for the class imbalance problem: bagging-, boosting-, and hybrid-based approaches. IEEE Trans. Syst. Man Cybernet. Part C (Appl. Rev.) 42(4), 463–484 (2011)
Geirhos, R., Janssen, D.H.J., Schütt, H.H., Rauber, J., Bethge, M., Wichmann, F.A.: Comparing deep neural networks against humans: object recognition when the signal gets weaker. CoRR abs/1706.06969 (2017). http://arxiv.org/abs/1706.06969
Hammer, P.L., Rudeanu, S.: Pseudo-Boolean programming. Oper. Res. 17(2), 233–261 (1969). https://doi.org/10.1287/opre.17.2.233
Junges, S., Jansen, N., Katoen, J.-P., Topcu, U., Zhang, R., Hayhoe, M.: Model checking for safe navigation among humans. In: McIver, A., Horvath, A. (eds.) QEST 2018. LNCS, vol. 11024, pp. 207–222. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-99154-2_13
Khreich, W., Granger, E., Miri, A., Sabourin, R.: Iterative Boolean combination of classifiers in the ROC space: an application to anomaly detection with HMMs. Pattern Recogn. 43(8), 2732–2752 (2010). https://doi.org/10.1016/j.patcog.2010.03.006
Levinson, J., Montemerlo, M., Thrun, S.: Map-based precision vehicle localization in urban environments. In: Proceedings of Robotics: Science and Systems. Atlanta, GA, USA, June 2007. https://doi.org/10.15607/RSS.2007.III.016
Levinson, J., Thrun, S.: Robust vehicle localization in urban environments using probabilistic maps. In: IEEE International Conference on Robotics and Automation, pp. 4372–4378 (2010)
Moras, J., Cherfaoui, V., Bonnifait, P.: Moving objects detection by conflict analysis in evidential grids. In: IEEE Intelligent Vehicles Symposium (IV 2011), pp. 1120–1125 (2011)
Păsăreanu, C.S., Gopinath, D., Yu, H.: Compositional verification for autonomous systems with deep learning components. In: Yu, H., Li, X., Murray, R.M., Ramesh, S., Tomlin, C.J. (eds.) Safe, Autonomous and Intelligent Vehicles. UST, pp. 187–197. Springer, Cham (2019). https://doi.org/10.1007/978-3-319-97301-2_10
Petrovskaya, A., Thrun, S.: Model based vehicle detection and tracking for autonomous urban driving. Auton. Robots 26(2–3), 123–139 (2009)
Powers, D.: Evaluation: From precision, recall and f-measure to ROC, informedness, markedness & correlation. J. Mach. Learn. Technol. 2(1), 37–63 (2011)
Radtke, P.V., Granger, E., Sabourin, R., Gorodnichy, D.O.: Skew-sensitive Boolean combination for adaptive ensembles – an application to face recognition in video surveillance. Inf. Fus. 20, 31–48 (2014). https://doi.org/10.1016/j.inffus.2013.11.001
Sagi, O., Rokach, L.: Ensemble learning: a survey. WIREs Data Min. Knowl. Discovery 8(4), e1249 (2018). https://doi.org/10.1002/widm.1249
Schumann, J., Liu, Y. (eds.): Applications of Neural Networks in High Assurance Systems, Studies in Computational Intelligence, vol. 268. Springer, Cham (2010). https://doi.org/10.1007/978-3-642-10690-3
Scott, M.J.J., Niranjan, M., Prager, R.W.: Realisable classifiers: improving operating performance on variable cost problems. In: Proceedings of the British Machine Vision Conference, pp. 31.1–31.10. BMVA Press (1998)
Thrun, S., Burgard, W., Fox, D.: Probabilistic Robotics (Intelligent Robotics and Autonomous Agents). The MIT Press, Cambridge (2005)
U.S. Department of Transportation, N.H.T.S.A.: Automated driving systems 2.0. a vision for safety (2017). www.nhtsa.gov/sites/nhtsa.dot.gov/files/documents/13069a-ads2.0_090617_v9a_tag.pdf
Acknowledgements
The research reported herein has been supported by the State of Lower Saxony within the Zukunftslabor Mobilität as well as by Deutsche Forschungsgemeinschaft under grant no. DFG FR 2715/5-1 “Konfliktresolution und kausale Inferenz mittels integrierter sozio-technischer Modellbildung”. It furthermore benefit from technical discussions with Jan Peleska, and we dedicate it to him on the occasion of his 65th anniversary.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Proofs
A Proofs
Proof of
Lemma 1. We use induction over n to show that the inequality
holds for any positive integer n and fractions \(\frac{a_1}{b_1}, \dots , \frac{a_n}{b_n}\) with real nominators \(a_1,\dots ,a_n\) and positive real denominators \(b_1,\dots ,b_n\). The base case \(n=1\) is trivial and the case \(n=2\) follows immediately from the mediant inequality \(\frac{a}{b} \le \frac{c}{d} \implies \frac{a}{b} \le \frac{a + c}{b + d} \le \frac{c}{d}\) for real numbers a, b and positive real numbers b, d. Assume that the induction hypothesis holds for \(n-1\). For any fraction \(\frac{a_n}{b_n}\) with real \(a_n,b_n>0\) at least one of the inequalities (i) \(\frac{\sum _{i=1}^{n-1} a_i}{\sum _{i=1}^{n-1} b_i} \le \frac{a_n}{b_n}\) or (ii) \(\frac{a_n}{b_n} \le \frac{\sum _{i=1}^{n-1} a_i}{\sum _{i=1}^{n-1} b_i}\) holds. From case (i) it follows
and from case (ii) it follows
where \((*)\) denotes the application of the mediant inequality. \(\square \)
Proof of
Lemma 2. We have to show that the inequalities
hold for all disjoint events \(B_1,\dots ,B_n\). Using the identity , an application of Lemma 1 yields
which finally rewrites to the asserted inequalities. \(\square \)
Proof of
Lemma 3. We have to show that the identity
is equivalent to the identity
for all positive integers n, atoms \(A_1,\dots A_n\), and \(i\ne j\), \(i\le n\), \(j\le n\). Note that we implicitly assume the well-definedness of Eq. (3) and Eq. (4). I.e., both identities stipulate \(P(A_1^{\pi _1},\dots ,A_n^{\pi _n})>0\) and Eq. (3) additionally stipulates \(P(A_j^{\lambda _j} \mid A_1^{\pi _1},\dots ,A_n^{\pi _n}) > 0\). The equivalent transformation from Eq. (3) to Eq. (4) is obtained by multiplying both sides of Eq. (3) with \(P(A_j^{\lambda _j} \mid A_1^{\pi _1},\dots ,A_n^{\pi _n})\). The equivalent transformation from Eq. (4) to Eq. (3) by division is valid as long as the stronger stipulation \(P(A_j^{\lambda _j} \mid A_1^{\pi _1},\dots ,A_n^{\pi _n}) > 0\) imposed by Eq. (3) holds. Finally, note that for \(P(A_j^{\lambda _j} \mid A_1^{\pi _1},\dots ,A_n^{\pi _n}) = 0\) the identity Eq. (4) does not contain any deeper findings, as it degenerates to the trivial identity \(0=0\) in this case. \(\square \)
Proof of
Lemma 4. We have to show that the identities
hold for the limit probabilities \(\underline{P}\) and \(\overline{P}\) for all \(i\ne j\). To see this consider the following chain of rewritings.
The corresponding identity for \(\overline{P}\) follows analogously. \(\square \)
Proof of
Thm. 1. We have to show that
holds for any formula \(\phi \) with label \(\phi ^{\lambda _\phi }\) and truth value \(\phi ^{\pi _\phi }\). In Eq. (1) we already showed
As all involved probabilities are nonnegative, monotonicity of \(\le \) yields
Combining the inequalities yields the asserted estimation. \(\square \)
Proof of
Thm. 2. Let \(\phi \) be an arbitrary formula with given label \(\phi ^\lambda _\phi \) and truth value \(\phi ^{\pi _\phi }\). Further let \(m=(A_1^{\pi _1},\dots ,A_n^{\pi _n})\) be a truth assignment and \(l=(A_1^{\lambda _1},\dots ,A_n^{\lambda _n})\) a label assignment for all atoms \(A_i\). In order to establish the theorem we show
The bounds
can be obtained without any further assumption and have already been established in Eq. (1). In Eq. (2) we already argued that the bounds
can be derived under the Independent Labelling Assumption 1. Finally, Assumption 2 allows us to bound each term of the form \(P(A_i^{\lambda _i} \mid m) = P(A_i^{\lambda _i} \mid A_1^{\pi _1},\dots ,A_n^{\pi _n})\) by its respective lower and upper bound \(\underline{P}({A_i}^{\lambda _i} \mid {A_i}^{\pi _i})\) and \(\overline{P}({A_i}^{\lambda _i} \mid {A_i}^{\pi _i})\). As all involved terms are nonnegative, we obtain
\(\square \)
Proof of
Thm. 4. We show the identities for the lower limiting rates \(\underline{\textrm{TPR}}_{\phi \wedge \psi }=\underline{\textrm{TPR}}_\phi \underline{\textrm{TPR}}_\psi \) and \(\underline{\textrm{FNR}}_{\phi \wedge \psi }=\underline{\textrm{FNR}}_\phi + \underline{\textrm{FNR}}_\psi - \underline{\textrm{FNR}}_\phi \underline{\textrm{FNR}}_\psi \). The identities for the upper limiting rates follow analogously. The remaining estimations for \(\underline{\textrm{TNR}}_{\phi \vee \psi }\), \(\underline{\textrm{FPR}}_{\phi \vee \psi }\), \(\overline{\textrm{TNR}}_{\phi \vee \psi }\), and \(\overline{\textrm{FPR}}_{\phi \vee \psi }\) can be obtained from the identity \(\phi \vee \psi \equiv \lnot (\lnot \phi \wedge \lnot \psi )\) and Thm. 3.
We decompose the conditional probabilities into proper assignments using Lemma 3 and Thm. 1.
\(\square \)
Proof of
Theorem 5. We show the inequalities for the lower limiting rates only. The corresponding inequalities for the upper limiting rates follow analogously. Note that for the conditional probabilities \(\underline{\textrm{TPR}}_{\phi \vee \psi }=\underline{P}((\phi \vee \psi )^+\mid (\phi \vee \psi )^\top )\) and \(\underline{\textrm{FNR}}_{\phi \vee \psi }=\underline{P}((\phi \vee \psi )^-\mid (\phi \vee \psi )^\top ))\) the conditioning event \((\phi \vee \psi )^\top \) can be decomposed into disjoint models yielding
Lemma 2 allows us to infer a lower estimate of the rates:
We decompose the labelling of \((\phi \vee \psi )^+\) and \((\phi \vee \psi )^-\) of each term in the minimum and maximum expression individually, where conjunctive events can further be decomposed into products using Lemma 3 E.g., the first term of the estimate for \(\underline{\textrm{TPR}}_{\phi \vee \psi }\) is rewritten as follows:
and the second term as follows:
After rewriting all terms accordingly, the lower bounds for \(\underline{\textrm{TPR}}_{\phi \vee \psi }\) and \(\underline{\textrm{FNR}}_{\phi \vee \psi }\) are established. The upper bounds for \(\overline{\textrm{TPR}}_{\phi \vee \psi }\) and \(\overline{\textrm{FNR}}_{\phi \vee \psi }\) follow analogously, and the remaining bounds for for \(\underline{\textrm{TNR}}_{\phi \wedge \psi }\), \(\overline{\textrm{TNR}}_{\phi \wedge \psi }\), \(\underline{\textrm{FPR}}_{\phi \wedge \psi }\), and \(\underline{\textrm{FPR}}_{\phi \wedge \psi }\) are obtained using De Morgan’s law. \(\square \)
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Fränzle, M., Hagemann, W., Damm, W., Rakow, A., Swaminathan, M. (2023). Safer Than Perception: Assuring Confidence in Safety-Critical Decisions of Automated Vehicles. In: Haxthausen, A.E., Huang, Wl., Roggenbach, M. (eds) Applicable Formal Methods for Safe Industrial Products. Lecture Notes in Computer Science, vol 14165. Springer, Cham. https://doi.org/10.1007/978-3-031-40132-9_12
Download citation
DOI: https://doi.org/10.1007/978-3-031-40132-9_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-40131-2
Online ISBN: 978-3-031-40132-9
eBook Packages: Computer ScienceComputer Science (R0)