Powered by
Conference Publishing Consulting

1st International Workshop on Control Theory for Software Engineering (CTSE 2015), August 31, 2015, Bergamo, Italy

CTSE 2015 – Proceedings

Contents - Abstracts - Authors
Twitter: https://twitter.com/FSEconf
Title Page

The Software Engineering community is pushing a significant effort on self-adaptive systems. Such systems are required to modify their behavior to maintain goals in response to unpredicted changes in their execution environment. Key challenges for self-adaptive systems include time-efficient diagnosis of requirements violation, fast decision making, and systematic procedures to assess their effectiveness and dependability. Further challenges arise in correlating local and global decision-making for larger-scale or distributed systems. Despite a variety of approaches has been proposed for self-adaptive software, only a few of them can provide formal guarantees about the quality of adaptation, mainly due to the difficulty of grounding the adaptation mechanisms within suitable theoretical frameworks.

SimCA vs ActivFORMS: Comparing Control- and Architecture-Based Adaptation on the TAS Exemplar
Stepan Shevtsov, M. Usman Iftikhar, and Danny Weyns
(Linnaeus University, Sweden)
Today customers require software systems to provide particular levels of qualities, while operating under dynamically changing conditions. These requirements can be met with different self-adaptation approaches. Recently, we developed two approaches that are different in nature - control theory-based SimCA and architecture-based ActivFORMS - to endow software systems with self-adaptation, providing guarantees on desired behavior. However, it is unclear which of the two approaches should be used in different adaptation scenarios and how effective they are in comparison to each other. In this paper, we apply SimCA and ActivFORMS to the Tele Assistance System (TAS) exemplar and compare obtained results, demonstrating the difference in achieved qualities and formal guarantees.

Publisher's Version Article Search
MORPH: A Reference Architecture for Configuration and Behaviour Self-Adaptation
Victor Braberman, Nicolas D'Ippolito, Jeff Kramer, Daniel Sykes, and Sebastian Uchitel
(University of Buenos Aires, Argentina; Imperial College London, UK)
An architectural approach to self-adaptive systems involves runtime change of system configuration (i.e., the system's components, their bindings and operational parameters) and behaviour update (i.e., component orchestration). Thus, dynamic reconfiguration and discrete event control theory are at the heart of architectural adaptation. Although controlling configuration and behaviour at runtime has been discussed and applied to architectural adaptation, architectures for self-adaptive systems often compound these two aspects reducing the potential for adaptability. In this paper we propose a reference architecture that allows for coordinated yet transparent and independent adaptation of system configuration and behaviour.

Publisher's Version Article Search
Adaptive Predictive Control for Software Systems
Konstantinos Angelopoulos, Alessandro Vittorio Papadopoulos, and John Mylopoulos
(University of Trento, Italy; Lund University, Sweden)
Self-adaptive software systems are designed to support a number of alternative solutions for fulfilling their requirements. These define an adaptation space. During operation, a self-adaptive system monitors its performance and when it finds that its requirements are not fulfilled, searches its adaptation space to select a best adaptation. Two major problems need to be addressed during the selection process: (a) Handling environmental uncertainty in determining the impact of an adaptation; (b) maintain an optimal equilibrium among conflicting requirements. This position paper investigates the application of Adaptive Model Predictive Control ideas from Control Theory to design self-adaptive software that makes decisions by predicting its future performance for alternative adaptations and selects ones that minimize the cost of requirement failures using quantitative information. The technical details of our proposal are illustrated through the meeting-scheduler exemplar.

Publisher's Version Article Search
Quo Vadis Cyber-Physical Systems: Research Areas of Cyber-Physical Ecosystems: A Position Paper
Christian Bartelt, Andreas Rausch, and Karina Rehfeldt
(TU Clausthal, Germany)
Many technological innovations from the research area of dynamic adaptive systems or IT ecosystems are already established in current software systems. Especially cyber-physical systems should benefit by this progress to provide smart applications in ambient environments of private and industrial space. But a proper and methodical engineering of cyber-physical ecosystems (CPES) is still an open and important issue. Traditional software and systems engineering facilities (system models, description languages, or process models) do not consider fundamental characteristics of these ecosystems as openness, uncertainty, or emergent constitution at runtime sufficiently. But especially these aspects let blur the line of system boundaries at design time. The diverse components of CPES have essential impacts on the engineering of CPES as well, concerning time synchronizing, execution control, and interaction structure. Self-balanced control in CPES promises new application possibilities, but also needs new engineering techniques concerning the overall engineering process, including requirements engineering and runtime verification. In this position paper we survey and summarize the dimensions of challenges in applying control theory for the engineering of cyber-physical ecosystems.

Publisher's Version Article Search
Robust Degradation and Enhancement of Robot Mission Behaviour in Unpredictable Environments
Nicolas D'Ippolito, Victor Braberman, Daniel Sykes, and Sebastian Uchitel
(University of Buenos Aires, Argentina; Imperial College London, UK)
Temporal logic based approaches that automatically generate controllers have been shown to be useful for mission level planning of motion, surveillance and navigation, among others. These approaches critically rely on the validity of the environment models used for synthesis. Yet simplifying assumptions are inevitable to reduce complexity and provide mission-level guarantees; no plan can guarantee results in a model of a world in which everything can go wrong. In this paper, we show how our approach, which reduces reliance on a single model by introducing a stack of models, can endow systems with incremental guarantees based on increasingly strengthened assumptions, supporting graceful degradation when the environment does not behave as expected, and progressive enhancement when it does.

Publisher's Version Article Search
Control Theory Meets Software Engineering: The Holonic Perspective
Luca Pazzi
(University of Modena and Reggio Emilia, Italy)
One of the main challenges towards a software-based theory of control consists in finding an effective method for decomposing monolithic event-based interactive applications into modules. The task is challenging since this requires in turn to decompose both the invariants to be maintained as well as the main control loop. We present a formalisms for gathering portion of behaviour by special units, called holons, which are both parts and wholes and which can be arranged into part-whole taxonomies. Each holon hosts a state machine and embodies different invariants which give semantics to its states. Control is achieved by both taking autonomously internal actions by the state machine in order to maintain such state invariants, as well as by having the the state machine move from one invariant to another by actions driven by external events. Such an approach requires to introduce non trivial solutions in order to allow communication among such modules, mainly by implementing control loops among couple of holons. The proposed model consists essentially in shaping each module in order to be both a controller and a controllable entity. Each module may control a definite number of modules and is controlled by a single module. Control is exercised by discrete events which travel through a communication medium. Control actions as well as feedback events travel thus from a module to the another, thus achieving local control loops which, taken globally, decompose the main control loop.

Publisher's Version Article Search

proc time: 0.38