Modelling adaptive controllers with evolving logic programs

Proceedings of the 5th International Conference on Pervasive and Embedded Computing and Communication Systems, 2015

The paper deals with a new Object Oriented Solution for Adaptive Control Systems having as name CHAMELEON. The CHAMELEON system provides more flexibility and a gain in memory compared to the Classical Oriented Object Approach. Firstly, we give a formalization of the solution. Then, we propose a CHAMELEON meta-model and a sequence diagram to emphasize the dynamic aspects. Finally, an evaluation is done in order to highlight the performance of the proposed approach.

Modern software systems need to be continuously available under varying conditions. Their ability to dynamically adapt to their execution context is thus increasingly seen as a key to their success. Recently, many approaches were proposed to design and support the execution of Dynamically Adaptive Systems (DAS). However, the ability of a DAS to evolve is limited to the addition, update or removal of adaptation rules or reconfiguration scripts. These artifacts are very specific to the control loop managing such a DAS and runtime evolution of the DAS requirements may affect other parts of the DAS. In this paper, we argue to evolve all parts of the loop. We suggest leveraging recent advances in model-driven techniques to offer an approach that supports the evolution of both systems and their adaptation capabilities. The basic idea is to consider the control loop itself as an adaptive system.

Journal of Parallel and Distributed Computing, 1991

This paper presents an ASOCS (Adaptive Self-Organizing Concurrent System) model for massively parallel processing of incrementally defined rule systems in such areas as adaptive logic, robotics, logical inference, and dynamic control. An ASOCS is an adaptive network composed of many simple computing elements operating asynchronously and in parallel. This paper focuses on Adaptive Algorithm 2 (AA2) and details its architecture and learning algorithm. AA2 has significant memory and knowledge maintenance advantages over previous ASOCS models. An ASOCS can operate in either a data processing mode or a learning mode. During learning mode, the ASOCS is given a new rule expressed as a boolean conjunction. The AA2 learning algorithm incorporates the new rule in a distributed fashion in a short, bounded time. During data processing mode, the ASOCS acts as a parallel hardware circuit.

2000

Fuzzy logic controllers have been applied to a wide range of control problems, but are very difficult to build for situations where the environment changes quickly and there is a lot of uncertainty. This work investigates a new method of creating fuzzy controllers, in the form of reactive agents, for such environments. The framework for this investigation is the RoboCup

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2009

Logic Programming Update Languages were proposed as an extension of logic programming that allows modeling the dynamics of knowledge bases where both extensional (facts) and intentional knowledge (rules) may change over time due to updates. Despite their generality, these languages do not provide a means to directly access past states of the evolving knowledge. They are limited to so-called Markovian change, i.e. changes entirely determined by the current state.

Proc. of the 34th International Conference on Software Engineering (ACM Student Research Competition), 2012

Complexity is now one of the major challenges for the IT industry [1]. Systems might become too complex to be managed by humans and, thus, will have to be self-managed: Self-configure themselves for operation, self-protect from attacks, self-heal from errors and self-tune for optimal performance [2]. (Self-)Adaptive systems evaluate their own behavior and change it when the evaluation indicates that it is not accomplishing the software's purpose or when better functionality and performance are possible [3]. To that end, we need to monitor the behavior of the running system and compare it to an explicit formulation of requirements and domain assumptions [4]. Feedback loops (e.g., the MAPE loop [2]) constitute an architectural solution for this and, as proposed by past research [5], should be a first class citizen in the design of such systems. We advocate that adaptive systems should be designed this way from as early as Requirements Engineering and that reasoning over requirements is fundamental for run-time adaptation. We therefore propose an approach for the design of adaptive systems based on requirements and inspired in control theory [6]. Our proposal is goal-oriented and targets softwareintensive socio-technical systems [7], in an attempt to integrate control-loop approaches with decentralized agents inspired approaches [8]. Our final objective is a set of extensions to state-of-the-art goal-oriented modeling languages that allow practitioners to clearly specify the requirements of adaptive systems and a run-time framework that helps developers implement such requirements. In this 2-page abstract paper, we summarize this approach.

Current research initiatives like Organic Computing (OC) and Intelligent Environments (IEs) aim at building technology-driven embedded systems that are aware of the conditions in the real world, derive plans of how to act, and proactively manipulate their environment to serve the user. As a result, novel systems according to these initiatives are characterised by a high degree of adaptivity, user-sensitivity, robustness and reliability. Adaptivity means that for such systems we defer a part of the design process from design time to runtime. Therefore we need a runtime infrastructure, which takes care of runtime modifications. And we need a meta-design process responsible for the parametrisation of the runtime infrastructure. This article presents such a meta-design process to develop such systems in a unified way in order to handle the growing complexity in technical systems by focusing on adaptation and self-optimisation capabilities. A promising application for organic concepts and a large-scale testbed for IEs is the control of road traffic signals in urban areas. This article exemplarily applies the proposed design process to a traffic scenario and evaluates it in a realistic setting modelling the (football) stadium environment in Hannover, Germany. Thereby, the significant benefit of the adaptation component developed through the design process is demonstrated in terms of traffic-relevant metrics like delays and energy consumption.

The growing interest in developing adaptive systems has led to numerous proposals for approaches aimed at supporting the development of such systems. Some approaches define adaptation mechanisms in terms of architectural designs, consisting of concepts such as components, connectors and states. Other approaches are requirements-based, thus concerned with goals, tasks, contexts and preferences as concepts in terms of which adaptation is defined. By considering only a partial view of software systems (either the problem space or the solution space), such proposals are limited in specifying the adaptive behavior of a software system. In this paper we present ongoing work towards deriving architectural models in order to support the design and runtime execution of software adaptation both at a requirements and architectural level.

International Journal of Intelligent Information and Database Systems, 2017

Evolution is inherent in software systems because of the rapid improvement of technologies. As metamodels are the cornerstone of model driven engineering and they evolve iteratively, their evolution affects the rest of artefacts involved in a development process, e.g., models or transformation rules. Therefore, co-evolution tools for models and other artefacts are indispensable. We present in this paper an intelligent approach to adapt models by means of state-based and operator-based techniques. The defined co-evolution process consists of four phases. Initially, changes between two metamodel versions are detected. After that, evolution scenario is reconstructed using logic programming method. Evolution scenario is first calculated from difference model and represented by a set of primitive evolution operations, after that it is transformed using an inference engine to include eventual composite evolution operations; from this scenario adaptation solutions are generated according to the evolution operator impact in model level. Finally, migration scenario is applied on an input model conforming to the old metamodel version to remain compliant with its metamodel.