• Video
  • 21-Mar-2012 09:50 EDT

Integrating Formal Model Checking with the RTEdge™ AADL Microkernel

00:26:21
Length:

Purchase Required to View Video

Short Preview Below

Edgewater Computer Systems Inc. product RTEdge Platform 1.2 is a software toolset supporting proof based engineering, implementation and deployment of software components, built using the RTEdge AADL Microkernel modeling subset. This is a small subset of the AADL component model and execution semantics, covering threads and thread-groups communicating solely through asynchronous event ports and through explicitly shared data ports. Threads behavior is expressed as state machines and dispatch run time semantics is encoded in a Run-time Executive, enforcing pre-emptive priority dispatch based on statically assigned event priorities, with ceiling priority protocol access to shared data. This simple AADL microkernel semantic core can support all dispatch policies, communication and synchronization mechanisms of a fully fledged AADL run time environment, permitting the systematic use of the RTEdge static analysis tools for AADL compliant software components. The objective of our work was to integrate RTEdge Platform 1.2 with the Promela / Spin model checker, for the purpose enabling formal verification of user defined formal properties, expressed on software application models built and executed on the RTEdge AADL Microkernel. In order to deal with the state space explosion problem, Model Checking relies on abstractions. Human intervention is required in an error prone process of defining the right abstractions, implementing an equivalent abstracted model in the model checker formalism, proving or disproving a property by model checking and applying the findings to the implementation. Assume-Guarantee Compositional Verification provides both a state space divide and conquer strategy and a model checking automation strategy, in a reverse engineering process applied to existing component implementations, as an iterative process of assumption refinement by model checking. The RTEdge concepts of Protocols, Capsule Interfaces and Composite Capsules open the possibility of using Assume-Guarantee contracts as a specification mechanism, in a forward engineering manner, and of verifying their consistence and conformance by the systematic use of Model Checking. Such an approach would allow verifying very large state spaces by using hierarchical component abstractions formally maintained in the RTEdge AADL component model via the extends and implements refinement relationships. To that end, Protocols, Capsule Interfaces and Composite Capsules must have a compact and convenient way of specifying sets of legal sequences of Signal Events. We developed a set of requirements for adapting IEEE1850 Property Specification Language (PSL) to formally express RTEdge software component temporal properties. The adapted PSL will use RTEdge Protocols Signals and Data as the underlying vocabulary of the PSL Boolean layer.

Presenter
Serban Gheorghe, Edgewater Computer Systems Inc.

Buy
Select
Price
List
Purchase to View
$19.00
Learn More
Technical Paper / Journal Article
2011-10-18
TECH PPR 2011 AEROTECH
Share
HTML for Linking to Page
Page URL
Grade
Rate It
No ratings yet

View More Video

Video
2015-12-22
“Spotlight on Design” features video interviews and case studies, focusing on technology breakthroughs, hands-on testimonials, and the importance of fundamentals. Viewers are virtually taken to industry labs and research centers to learn how design engineers solve real-life problems. These challenges include enhancing product performance, reducing cost, improving quality and safety, while decreasing environmental impact, and achieving regulatory compliance. In the episode “Engine Development for Performance and Efficiency” (22:00), engineers from Ricardo and General Motors explain the importance of lessons learned on the race tracks, how using the appropriate design and simulation tools expedite development, and present the new testing frontiers now available with 3D printing.
Video
2012-02-15
Virtual testing is a method that simulates lab testing using multi-body dynamic analysis software. The main advantages of this approach include that the design can be evaluated before a prototype is available and virtual testing results can be easily validated by subsequent physical testing. The disadvantage is that accurate specimen models are sometimes hard to obtain since nonlinear components such as tires, bushings, dampers, and engine mounts are hard to model. Therefore, virtual testing accuracy varies significantly. The typical virtual rigs include tire and spindle coupled test rigs for full vehicle tests and multi axis shaker tables for component tests. Hybrid simulation combines physical and virtual components, inputs and constraints to create a composite simulation system. Hybrid simulation enables the hard to model components to be tested in the lab.
Video
2012-05-25
In support of the U.S Department of Energy's Vehicle Technologies Program, numerous vehicle technology combinations have been simulated using Autonomie. Argonne National Laboratory (Argonne) designed and wrote the Autonomie modeling software to serve as a single tool that could be used to meet the requirements of automotive engineering throughout the development process, from modeling to control, offering the ability to quickly compare the performance and fuel efficiency of numerous powertrain configurations. For this study, a multitude of vehicle technology combinations were simulated for many different vehicles classes and configurations, which included conventional, power split hybrid electric vehicle (HEV), power split plug-in hybrid electric vehicle (PHEV), extended-range EV (E-REV)-capability PHEV, series fuel cell, and battery electric vehicle.
Video
2012-02-15
Zero-dimensional, one-dimensional, and quasi-dimensional models for simulation of SI and CI engines with respect to: engine breathing and boosting; SI combustion and emissions; CI combustion and emissions; fundamentals of engine thermodynamics; thermal management; mechanical and lubrication systems; system level models for controls; system level models for vehicle fuel economy and emissions predictions. Presenter Fabio Bozza, Universita di Napoli

Related Items

Article
2017-07-26
Training / Education
1997-05-29
Technical Paper / Journal Article
2011-05-17
Event
2018-04-10