Formalized analysis of structural characteristics of large complexsystems

Macrostructural modelization is paramount to the development of large complex systems (LCS). The paper explores the macrostructural modelization of LCS in terms of a block diagram based model and a grammar based model. Firstly, the macrostructural modelization problem of LCS is formulated. Secondly, a block diagram based model is proposed and established for LCS. Specifically, two general-purpose information-processing modules are proposed and constructed, called perception cube and decision spheroid. Thirdly, a grammar based model is proposed and established for LCS through applying formal language theory to the block diagram based model. Specifically, perception cube and decision spheroid are visually represented as context-free grammars, named fusion grammar and synthesis grammar, respectively. Through a stratified constructive linkup between a stream of bottom-up growing fusion grammars and a stream of top-down growing synthesis grammars, a level of LCS is constructively defined and accordingly represented as a context-free grammar, named level grammar. Then, a whole LCS is represented as a context-free grammar through a compounding of all level grammars. Finally, a case study is presented to demonstrate the potential usability of the proposed and established models of LCS     

Critic Systems – Towards Human–Computer Collaborative Problem Solving

Human–computer collaboration is extremely necessary for solving ill-structured problems and critic systems can effectively facilitate human–computer collaborative problem solving. This paper conducts a systematic study on critic systems. First, the concepts of critic systems are presented. Then, a literature review is presented on critic systems. Afterwards, a generic architecture is put forward for critic systems, with its important aspects being analyzed. Finally, two case studies are given to illustrate critic systems.     

Structuring of Large-scale Complex Hybrid Systems: from Illustrative Analysis toward Modelization

System structuring is paramount to the development of large-scale complex hybrid systems (LCHS). However, there is no well-established and effective methodology for the structuring of LCHS. Using the approach of illustrating and abstracting, this paper investigates the structuring of LCHS based on a wide variety of applications. First, intrinsic attributes of LCHS are explored. Then, a unified framework-based analysis is made of six typical examples of LCHS, i.e. human brain-body systems, autonomous robot systems, autonomous fed-batch reactor systems, human-car driving systems, autonomous urban traffic systems, and autonomous production (manufacturing/industrial process) systems. In this way, common characteristics among typical examples, e.g., perception–decision link, distributing, nesting, hierarchy, multiple gradation and hybrid dynamics over a spectrum of time drivers, and, incidentally, wide application orientations of LCHS, are illustrated. Furthermore, after basic concepts for the modelization of LCHS are formulated, two novel general-purpose information-processing modules are proposed and constructed, called perception cube and decision spheroid. Based on them, a novel block diagram based model is originally proposed and established for LCHS.     

Sequencing via explanation‐based learning

Like planning and scheduling, sequencing is an unstructured problem in enterprises. Knowledge‐based approaches are necessary in many cases. Domain‐specific searching heuristics are very effective for increasing the searching efficiency of sequencing. However, it is rather difficult to acquire and especially to recognize, acquire, and compile automatically. In this paper, after the sequencing problem is illustrated and formally formulated, a new learning‐based method is developed for sequencing by applying explanation‐based learning, the latter of which is best at acquiring domain‐specific searching heuristics. Prolog‐based representation and logic charts are created to illuminate the basic principle of the sequencing via explanation‐based learning. Moreover, an example study shows that this method is feasible and effective. © 2001 John Wiley & Sons, Inc.     

An efficient network coded ARQ for multisource multidestination relay networks over mixed flat fading channels

This paper proposes a new reliable automatic repeat request (ARQ) transmission protocol for wireless multisource multidestination relay networks over mixed fading channels. Conventional application of ARQ protocols to retransmit lost or erroneous packets in relay networks can cause considerable delay latency with a significant increase in the number of retransmissions when networks consist of multiple sources and multiple destinations. To address this issue, a new ARQ protocol based on network coding (NC) is proposed where the relay detects packets from different transmission sources, then uses NC to combine and forward lost packets to their destinations. An efficient means for the retransmission of all lost packets is proposed through two packet-combination algorithms for retransmissions at the relay and sources. The paper derives mathematical formulation of transmission bandwidth for this new NC-based ARQ protocol and compares analytical and simulation results with some other ARQ protocols over both mixed Rayleigh and Rician flat fading channel. The mixed fading model permits investigation of two typical fading scenarios where the relay is located in the neighbourhood of either the sources or the destinations. The transmission bandwidth results show that the proposed NC-based ARQ protocol demonstrates superior performance over other existing ARQ schemes.     

Robot action planning via explanation-based learning

Domain-specific searching heuristics is greatly influential upon the searching efficiency of robot action planning (RAP), but its computer-realized recognition and acquisition, i.e., learning, is difficult. This paper makes an exploration into this challenge. First, a problem formulation of RAP is made. Then, by applying explanation-based learning, which is currently the only approach to acquiring domain-specific searching heuristics, a new learning based method is developed for RAP, named robot action planning via explanation-based learning (RAPEL). Finally, an example study demonstrates the effectiveness of RAPEL     

Self-adaptive differential evolution algorithm with α-constrained-domination principle for constrained multi-objective optimization

Real-world problems are inherently constrained optimization problems often with multiple conflicting objectives. To solve such constrained multi-objective problems effectively, in this paper, we put forward a new approach which integrates self-adaptive differential evolution algorithm with α-constrained-domination principle, named SADE-αCD. In SADE-αCD, the trial vector generation strategies and the DE parameters are gradually self-adjusted adaptively based on the knowledge learnt from the previous searches in generating improved solutions. Furthermore, by incorporating domination principle into α-constrained method, α-constrained-domination principle is proposed to handle constraints in multi-objective problems. The advantageous performance of SADE-αCD is validated by comparisons with non-dominated sorting genetic algorithm-II, a representative of state-of-the-art in multi-objective evolutionary algorithms, and constrained multi-objective differential evolution, over fourteen test problems and four well-known constrained multi-objective engineering design problems. The performance indicators show that SADE-αCD is an effective approach to solving constrained multi-objective problems, which is basically enabled by the integration of self-adaptive strategies and α-constrained-domination principle.