arXiv subjects

Kengo Kido

Publications and source records attributed to Kengo Kido.

3 recordsLinked to original sources

arXiv2017

Bounding Errors Due to Switching Delays in Incrementally Stable Switched Systems (Extended Version)

Time delays pose an important challenge in networked control systems, which are now ubiquitous. Focusing on switched systems, we introduce a framework that provides an upper bound for errors caused by switching delays. Our framework is based on approximate bisimulation, a notion that has been previously utilized mainly for symbolic (discrete) abstraction of state spaces. Notable in our framework is that, in deriving an approximate bisimulation and thus an error bound, we use a simple incremental stability assumption (namely {\delta}-GUAS) that does not itself refer to time delays. That this is the same assumption used for state-space discretization enables a two-step workflow for control synthesis for switched systems, in which a single Lyapunov-type stability witness serves for two different purposes of state discretization and coping with time delays. We demonstrate the proposed framework with a boost DC-DC converter, a common example of switched systems.

eess.SY↗

arXiv2017

Sharper and Simpler Nonlinear Interpolants for Program Verification

Interpolation of jointly infeasible predicates plays important roles in various program verification techniques such as invariant synthesis and CEGAR. Intrigued by the recent result by Dai et al.\ that combines real algebraic geometry and SDP optimization in synthesis of polynomial interpolants, the current paper contributes its enhancement that yields sharper and simpler interpolants. The enhancement is made possible by: theoretical observations in real algebraic geometry; and our continued fraction-based algorithm that rounds off (potentially erroneous) numerical solutions of SDP solvers. Experiment results support our tool's effectiveness; we also demonstrate the benefit of sharp and simple interpolants in program verification examples.

cs.LO↗

arXiv2015

Abstract Interpretation with Infinitesimals: Towards Scalability in Nonstandard Static Analysis (Extended Version)

We extend abstract interpretation for the purpose of verifying hybrid systems. Abstraction has been playing an important role in many verification methodologies for hybrid systems, but some special care is needed for abstraction of continuous dynamics defined by ODEs. We apply Cousot and Cousot's framework of abstract interpretation to hybrid systems, almost as it is, by regarding continuous dynamics as an infinite iteration of infinitesimal discrete jumps. This extension follows the recent line of work by Suenaga, Hasuo and Sekine, where deductive verification is extended for hybrid systems by 1) introducing a constant dt for an infinitesimal value; and 2) employing Robinson's nonstandard analysis (NSA) to define mathematically rigorous semantics. Our theoretical results include soundness and termination via uniform widening operators; and our prototype implementation successfully verifies some benchmark examples.

cs.PL↗