Chris J. Hinde

A new extension of fuzzy sets using rough sets: R-fuzzy sets

This paper presents a new extension of fuzzy sets: R-fuzzy sets. The membership of an element of a R-fuzzy set is represented as a rough set. This new extension facilitates the representation of an uncertain fuzzy membership with a rough approximation. Based on our definition of R-fuzzy sets and their operations, the relationships between R-fuzzy sets and other fuzzy sets are discussed and some examples are provided.

Inference of fuzzy relational tableaux from fuzzy exemplifications

The paper describes a method for deriving fuzzy relational tableaux from fuzzy exemplifications. In particular the work by Thole, Zimmerman and Zysno [4] is cited as they derive an operator for AND using data derived from human subjects. The method results in an operator or operators which is a maximal restriction of the operators implied by the data. Results showing the process are described and a value of AND is derived using data presented in [4].

Using neural networks as a tool for constructing rule based systems

Abstract The paper presents a method for deriving rules from a neural network and abstracting meaning from them. This produces a connectionist rule induction system that combines some of the useful attributes of connectionist and traditional symbolic systems. A tutorial example of the noise immunity enjoyed by neural networks is presented together with examples drawn from the field of intelligent control. Finally, the limits of the technique are explored by applying the technique to a large scale neural network drawn from visual character recognition.

Feature recognition within a truth maintained process planning system

Abstract A major problem for computer aided process planning (CAPP)is the difference in data representation between the design and planning stages. Computer aided design (CAD) uses a data representation which is suited to solid modelling but is impractical for process planning. A partial solution to the non-unique nature of the set-theoretic solid model is detailed. A method of feature recognition for the set-theoretic solid model, together with its implementation in a process planning system is described. It is essentially a rule-based conversion of the CSG string into a set of manufacturing features. One or more alternative ways of viewing the component in terms of features are produced, and user interaction may be required to determine the best set of features to process plan with.

Experience of the Application of Intelligent Control Paradigms to Real Manufacturing Processes

The application of conventional and intelligent control paradigms to the dispensing of adhesive in the manufacture of mixed technology printed circuit boards, a real process with real time constraints and error conditions, is outlined in this paper. A description of the process and its various operation and error states is followed by a brief discussion of the application of conventional (system identification and incremental proportional control) and intelligent (knowledge-based, fuzzy logic and neural network) controllers to the system. Comments on the control and timing performance of each paradigm are given.

Inconsistency and semantic unification

Abstract Baldwin et al. in their work Fril—fuzzy and evidential reasoning [Artificial Intelligence, Research Studies Press Ltd, Taunton, UK, 1995] present a semantic unification function for mass assignment logic. Its purpose is to evaluate the truth that a claim X is supported by evidence Y . This paper exposes some flaws with how the original unification function handles belief, possibility, and inconsistency. By making changes to the function this paper allows both interval and point-value semantic unification to be used in a more consistent and intuitive way.

An improved representation for evolving programs

A representation has been developed that addresses some of the issues with other Genetic Program representations while maintaining their advantages. This combines the easy reproduction of the linear representation with the inheritable characteristics of the tree representation by using fixed-length blocks of genes representing single program statements. This means that each block of genes will always map to the same statement in the parent and child unless it is mutated, irrespective of changes to the surrounding blocks. This method is compared to the variable length gene blocks used by other representations with a clear improvement in the similarity between parent and child. In addition, a set of list evaluation and manipulation functions was evolved as an application of the new Genetic Program components. These functions have the common feature that they all need to be 100% correct to be useful. Traditional Genetic Programming problems have mainly been optimization or approximation problems. The list results are good but do highlight the problem of scalability in that more complex functions lead to a dramatic increase in the required evolution time.

Allocating Railway Platforms Using A Genetic Algorithm

This paper describes an approach to automating railway station platform allocation. The system uses a Genetic Algorithm (GA) to find how a station’s resources should be allocated. Real data is used which needs to be transformed to be suitable for the automated system. Successful or ‘fit’ allocations provide a solution that meets the needs of the station schedule including platform re-occupation and various other constraints. The system associates the train data to derive the station requirements. The Genetic Algorithm is used to derive platform allocations. Finally, the system may be extended to take into account how further parameters that are external to the station have an effect on how an allocation should be applied. The system successfully allocates around 1000 trains to platforms in around 30 seconds requiring a genome of around 1000 genes to achieve this.

A Consensus-Based Grouping Algorithm for Multi-agent Cooperative Task Allocation with Complex Requirements

Abstract This paper looks at consensus algorithms for agent cooperation with unmanned aerial vehicles. The foundation is the consensus-based bundle algorithm, which is extended to allow multi-agent tasks requiring agents to cooperate in completing individual tasks. Inspiration is taken from the cognitive behaviours of eusocial animals for cooperation and improved assignments. Using the behaviours observed in bees and ants inspires decentralised algorithms for groups of agents to adapt to changing task demand. Further extensions are provided to improve task complexity handling by the agents with added equipment requirements and task dependencies. We address the problems of handling these challenges and improve the efficiency of the algorithm for these requirements, whilst decreasing the communication cost with a new data structure. The proposed algorithm converges to a conflict-free, feasible solution of which previous algorithms are unable to account for. Furthermore, the algorithm takes into account heterogeneous agents, deadlocking and a method to store assignments for a dynamical environment. Simulation results demonstrate reduced data usage and communication time to come to a consensus on multi-agent tasks.

Semantic transfer and contradictory evidence in intuitionistic fuzzy sets

The relationship between object level intuitionistic fuzzy sets and predicate based intuitionistic fuzzy sets is explored. Mass assignment uses a process called semantic unification to evaluate the degree to which one set supports another, the inverse function is semantic separation. Intuitionistic fuzzy sets are mapped onto a mass assignment framework and the semantic unification operator is generalised to support both mass assignment and intuitionistic fuzzy sets, as is semantic separation. Transfer of inconsistent and contradictory evidence are also dealt with.