Runtime Verification - 17 Years Later

Havelund, Klaus; Roşu, Grigore

doi:10.1007/978-3-030-03769-7_1

“…Runtime assurance is essential in the deployment of safety-critical (cyberphysical) systems [12,29,45,49,50]. Monitors observe system behavior and indicate when the system is at risk to violate system specifications.…”

Section: Introductionmentioning

confidence: 99%

Runtime Monitors for Markov Decision Processes

Junges

¹

,

Torfah

²

,

Seshia

³

2021

Computer Aided Verification

View full text Add to dashboard Cite

We investigate the problem of monitoring partially observable systems with nondeterministic and probabilistic dynamics. In such systems, every state may be associated with a risk, e.g., the probability of an imminent crash. During runtime, we obtain partial information about the system state in form of observations. The monitor uses this information to estimate the risk of the (unobservable) current system state. Our results are threefold. First, we show that extensions of state estimation approaches do not scale due the combination of nondeterminism and probabilities. While exploiting a geometric interpretation of the state estimates improves the practical runtime, this cannot prevent an exponential memory blowup. Second, we present a tractable algorithm based on model checking conditional reachability probabilities. Third, we provide prototypical implementations and manifest the applicability of our algorithms to a range of benchmarks. The results highlight the possibilities and boundaries of our novel algorithms.

show abstract

“…Path checking is generally much easier than model checking [3], particularly for infinite state models, for which the model checking problem is usually undecidable. Hence runtime verification, a popular approach to formal system verification [4], abandons model checking in favour of path checking, at the expense of exhaustive coverage. Path checking algorithms for discrete words are well studied [1,3,5].…”

Section: Introductionmentioning

confidence: 99%

Sampling Polynomial Trajectories for LTL Verification

Selvaratnam,

Cantoni,

Davoren

et al. 2021

Preprint

0

View full text Add to dashboard Cite

This paper concerns the verification of continuous-time polynomial spline trajectories against linear temporal logic specifications (LTL without 'next'). Each atomic proposition is assumed to represent a state space region described by a multivariate polynomial inequality. The proposed approach is based on sampling the trajectories in a manner that captures all region transitions, to yield a discrete word called a trace, which is amenable to established formal methods for path checking. The continuous-time trajectory is shown to satisfy the specification if and only if the trace does. General topological conditions on the sample points are derived that ensure a trace is recorded for arbitrary continuous paths, given arbitrary region descriptions. Using techniques from computer algebra, a trace generation algorithm is developed to satisfy these conditions when the path and region boundaries are defined by polynomials. The proposed PolyTrace algorithm has polynomial complexity in the number of atomic propositions, and is guaranteed to produce a trace of any polynomial path. Its performance is demonstrated via numerical simulations.

show abstract

“…Runtime assurance is essential in deployment of safety-critical (cyber-physical) systems [44,12,29,48]. Monitors observe system behavior and indicate when the system is at risk to violate system specifications.…”

Section: Introductionmentioning

confidence: 99%

Runtime Monitoring for Markov Decision Processes

Junges,

Torfah,

Seshia

2021

Preprint

0

View full text Add to dashboard Cite

We investigate the problem of monitoring partially observable systems with nondeterministic and probabilistic dynamics. In such systems, every state may be associated with a risk, e.g., the probability of an imminent crash. During runtime, we obtain partial information about the system state in form of observations. The monitor uses this information to estimate the risk of the (unobservable) current system state. Our results are threefold. First, we show that extensions of state estimation approaches do not scale due the combination of nondeterminism and probabilities. While exploiting a geometric interpretation of the state estimates improves the practical runtime, this cannot prevent an exponential memory blowup. Second, we present a tractable algorithm based on model checking conditional reachability probabilities. Third, we provide prototypical implementations and manifest the applicability of our algorithms to a range of benchmarks. The results highlight the possibilities and boundaries of our novel algorithms.

show abstract

Runtime Verification - 17 Years Later

Cited by 13 publications

References 52 publications

Runtime Monitors for Markov Decision Processes

Runtime Monitors for Markov Decision Processes

Sampling Polynomial Trajectories for LTL Verification

Runtime Monitoring for Markov Decision Processes

Contact Info

Product

Resources

About