Harnessing Uncertainty in Autonomous Vehicle Safety

Stephen Thomas; Dirk Vandenberg

doi:10.56094/jss.v55i2.46

Authors

Stephen L. Thomas Uber ATG https://orcid.org/0000-0002-8536-8079
Dirk J. Vandenberg Uber ATG

DOI:

https://doi.org/10.56094/jss.v55i2.46

Keywords:

autonomous vehicle, uncertainty, bayesian, self-driving, safety

Abstract

Safely developing self-driving vehicles presents technical challenges. Among the key technical challenges are how to confidently demonstrate the safety of a self-driving vehicle when the number of permutations of operating conditions, scenarios, system inputs, etc. are complex, uncertain and potentially limitless. This paper provides a broad survey of the various types of uncertainty in the development of self-driving vehicles and outlines several possible strategies for handling uncertainty. Advantages and challenges of different approaches, including qualitative and quantitative methods, are also discussed.

Author Biographies

Stephen L. Thomas, Uber ATG

Stephen L. Thomas has over 25 years of experience as a system safety and control systems engineer. He is currently a currently a System Safety Architect for autonomous vehicles. He holds a BS degree in chemical engineering from Auburn University and a Masters degree reliability engineering from the University of Maryland. He is a certified functional safety expert (CFSE), certified reliability engineer (CRE), and a licensed professional engineer (PE). He authors a blog at FunctionalSafetyEngineer.com.

Dirk J. Vandenberg, Uber ATG

Dirk Vandenberg is a Sr. System Safety engineer for autonomous vehicles. He has over 20 years of experience in systems and safety engineering. He holds a BS degree in electrical and computer engineering from Carnegie Mellon.

References

Bloomfield, R. and P. Bishop. "Safety and Assurance Cases: Past, Present and Possible Future - An Adelard Perspective," Making Systems Safer: Proceedings of the Eighteenth Safety-Critical Systems Symposium, 51-67,Springer, London, 2010. https://doi.org/10.1007/978-1-84996-086-1_4 DOI: https://doi.org/10.1007/978-1-84996-086-1_4

Butler, R. W. and G. B. Finelli. "The Infeasibility of Quantifying the Reliability of Life-Critical Real-TimeSoftware," IEEE Transactions on Software Engineering, Vol.19, Issue 1, 3-12, 1993. https://doi.org/10.1109/32.210303 DOI: https://doi.org/10.1109/32.210303

Littlewood, B. and L. Strigini. "Validation of Ultra-High Dependability for Software-Based Systems," Predictably Dependable Computing Systems, 473-493, Springer, Berlin, Heidelberg, 1995. https://doi.org/10.1007/978-3-642-79789-7_27 DOI: https://doi.org/10.1007/978-3-642-79789-7_27

Bishop, P. G. and R. E. Bloomfield. "The SHIP Safety Case Approach," Safe Comp 95: The 14th InternationalConference on Computer Safety, Reliability and Security, 437-451, Springer, London, 1995. https://doi.org/10.1007/978-1-4471-3054-3_30 DOI: https://doi.org/10.1007/978-1-4471-3054-3_30

Habli, I. and T. Kelly. "Achieving Integrated Process and Product Safety Arguments," The Safety of Systems:Proceedings of the Fifteenth Safety-critical Systems Symposium, 55-68, Springer, London, 2007. https://doi.org/10.1007/978-1-84628-806-7_4 DOI: https://doi.org/10.1007/978-1-84628-806-7_4

Goodenough, J. B. and C. B. Weinstock. "Toward a Theory of Assurance Case Confidence," Report No. CMU/SEI-2012-TR-002, Carnegie-Mellon University Software Engineering Inst., Pittsburgh Pennsylvania, 2012. https://doi.org/10.21236/ADA609836 DOI: https://doi.org/10.21236/ADA609836

Koopman, P. and B. Osyk. "Safety Argument Considerations for Public Road Testing of Autonomous Vehi-cles," No. 2019-01-0123, SAE Technical Paper, 2019. https://doi.org/10.4271/2019-01-0123 DOI: https://doi.org/10.4271/2019-01-0123

Shalev-Shwartz, S., S. Shammah and A. Shashua. "On a Formal Model of Safe and Scalable Self-DrivingCars," arXiv:1708.06374, Cornell University, Ithaca, New York, 2017. https://doi.org/10.48550/arXiv.1708.06374

Henriksson, J., C. Berger, M. Borg, L. Tornberg, C. Englund, S. R. Sathyamoorthy and S. Ursing. "TowardsStructured Evaluation of Deep Neural Network Supervisors," arXiv:1903.01263, Cornell University, Ithaca,New York, 2019. https://doi.org/10.1109/AITest.2019.00-12 DOI: https://doi.org/10.1109/AITest.2019.00-12

Littlewood, B. and D. Wright. "A Bayesian Model that Combines Disparate Evidence for the QuantitativeAssessment of System Dependability," Safe Comp 95: The 14th International Conference on Computer Safety,Reliability and Security, 173-188, Springer, London, 1995. https://doi.org/10.1007/978-1-4471-3054-3_13 DOI: https://doi.org/10.1007/978-1-4471-3054-3_13

Droguett, E. L., F. J. Groen and A. Mosleh. "Bayesian Assessment of the Variability of Reliability Measures,"Pesquisa Operacional, Vol. 26, Issue 1, 109-127, 2006. https://doi.org/10.1590/S0101-74382006000100006 DOI: https://doi.org/10.1590/S0101-74382006000100006