This paper examines a novel aerial platform utilizing fuzzy logic control with evolutionary tuning. First the new vehicle is introduced, then its dynamics are described. The development process and simulation techniques are then discussed. The fuzzy logic controller derived from the simulations is explained and then the physical vehicle built on the simulation model is examined. Finally, the testing results are presented and future directions are examined.

The objective of this paper is mainly to answer one question: "Why use a toolkit such as iOpt to solve a combinatorial optimization problem rather than hard-coding a solution from scratch?" To answer this question, we consider a well studied problem: the Vehicle Routing Problem. We explain in details how to make use of the modeling and solving facilities available in iOpt to tackle this problem. At each step of this building process, we discuss the benefits of using iOpt rather than starting building a solution from scratch. Then we exhibit some experiments comparing the results obtained using the best algorithm built using iOpt and the best known in the literature. The overall conclusion of this work is our toolkit allows the user to maximize reuse of his code, significantly reduce his development time, focus his attention on the design rather than the coding, and exchange problem models or algorithms in a very easy and simple way using XML files within his community. At last, algorithms built using iOpt appear to be very competitive compared to the best hard-wired algorithms found in the literature.

The profitability and morale of many organizations (such as factories, hospitals and airlines) are affected by their ability to schedule their personnel properly. Sophisticated and powerful constraint solvers such as ILOG, CHIP, ECLiPSe, etc. have been demonstrated to be extremely effective on scheduling. Unfortunately, they require non-trivial expertise to use. This paper describes ZDC-rostering, a constraint-based tool for personnel scheduling that addresses the software crisis and fills a void in the space of solvers. ZDC-rostering is easier to use than the above constraint-based solvers and more effective than Microsoft’s Excel Solver. ZDC-rostering is based on an open-source computer-aided constraint programming package called ZDC, which decouples problem formulation (or modelling) from solution generation in constraint satisfaction. ZDC is equipped with a set of constraint algorithms, including Extended Guided Local Search, whose efficiency and effectiveness have been demonstrated in a wide range of applications. Our experiments show that ZDC-rostering is capable of solving realistic-sized and very tightly-constrained problems efficiently. ZDC-rostering demonstrates the feasibility of applying constraint satisfaction techniques to solving rostering problems, without having to acquire deep knowledge in constraint technology.

Meta-heuristics are methods that sit on top of local search algorithms. They perform the function of avoiding or escaping a local optimum and/or premature convergence. The aim of this paper is to survey, compare and contrast meta-heuristics for local search. First, we present the technique of local search (or hill climbing as it is sometimes known). We then present a table displaying the attributes of all the different meta-heuristics. After this, we give a short description and discussion of each meta-heuristic with pseudo code. Finally, we describe why, in general, these techniques work and present some ideas of what is needed from the next generation of meta-heuristics.

Meta-heuristics are methods that sit on top of local search algorithms. They perform the function of avoiding or escaping a local optimum and/or premature convergence. The aim of this paper is to survey, compare and contrast meta-heuristics for local search. First, we present the technique of local search (or hill climbing as it is sometimes known). We then present a table displaying the attributes of all the different meta-heuristics. After this, we give a short description and discussion of each meta-heuristic with pseudo code. Finally, we describe why, in general, these techniques work and present some ideas of what is needed from the next generation of meta-heuristics.

In this paper, we show how an extended Guided Local Search (GLS) can be applied to the Quadratic Assignment Problem (Q AP). GLS is a general, penalty -based meta -heuristic, which sits on top of local search algorithms, to help guide them out of local minima. We present empirical results of applying several extended versions of GLS to the QAP, and show that these extensions can improve the range of parameter settings within which Guided Local Search performs well. Finally, we compare the results of running our extended GLS with some state of the art algorithms for the QAP.

Guided Local Search is a general penalty-based optimisation method that sits on top of local search methods to help them escape local optimum. It has been applied to a variety of problems and demonstrated effective. The aim of this paper is not to produce further evidence that Guided Local Search is an effective algorithm, but to present an extension of Guided Local Search that potentially has no parameter to tune. Compared to other algorithms, Guided Local Search is relatively easy to apply, as there is only one major parameter () to set. In some applications, performance of Guided Local Search is insensitive to the value of this parameter. Nevertheless, the value of this parameter can affect the performance of Guided Local Search in some problems. In this paper, we show how (a) an aspiration criterion and (b) random moves may be added to Guided Local Search to reduce the sensitivity of its performance to the parameter value. The extended Guided Local Search is tested on the SAT, weighted MAX-SAT and Quadratic Assignment Problems with positive results.

......................................................................................................................... 2 Acknowledgements......................................................................................................... 3

The constraint programming research community has accumulated a vast amount of experience in solving constraint satisfaction/optimisation problems. Unfortunately, applyi ng constraint technology to a particular problem requires expertise in the technology, which many potential users do not have. The CACP project attempts to provide a system that encompasses the entire process of applying constraint technology. It supports the tasks of problem formulation and entry, in addition to supplying pre -written solvers and aiding the user in choosing which of the available algorithms to apply. Users may specify their problems using a declarative language. The problem specification i s decoupled from the solvers, so the users may experiment with different problem formulations easily. Solvers supplied include a generalized Forward Checking solver, a Linear Programming solver and local search solvers implementing Guided Local Search, Tabu Search and Genetic Algorithms. A carefully designed interface is provided to guide users in understanding the technology.

In this paper, we show how Guided Local Search (GLS) can be applied to the SAT problem and show how the resulting algorithm can be naturally extended to solve the weighted MAX-SAT problem. GLS is a general, penalty-based meta-heuristic, which sits on top of local search algorithms to help guide them out of local minima. GLS has been shown to be successful in solving a number of practical real-life problems, such as the traveling salesman problem, BT"s workforce scheduling problem, the radio link frequency assignment problem, and the vehicle routing problem. We present empirical results of applying GLS to instances of the SAT problem from the DIMACS archive and also a small set of weighted MAX-SAT problem instances and compare them with the results of other local search algorithms for the SAT problem.

Guided Local Search (GLS) has been shown to be successful in solving a number of practical real life problems, such as the travelling salesman problem, radio link frequency assignment problem and the vehicle routing problem. GLS is a penalty-based metaheuristic, which works by augmenting the objective function of a local search algorithm with penalties, to help guide them out of local minima. Our aim is to show that adding GLS to Local Search algorithms generally enhances the performance of such algorithms, which do not include some similar meta-heuristic already. The SAT problem is a class of NP-complete problems. It is known to be important in mathematical logic, constraint satisfaction, VLSI engineering and computing theory. It has recently been the focus of much research on local search algorithms, for example, GSAT and WalkSAT. In this paper, we show progress in applying GLS to local search algorithms along similar lines to GSAT. Results so far show that GLS can reduce the amount of computational effort required to find a solution, when added to such local search algorithms and can also improve the success rate in finding solutions for the local search algorithms.

Research on children's causal thinking has emphasized the perception of temporal and spatial contiguity between cause and effect. However, our causal judgements often involve a contrast between a perceived sequence (A, then B) and a counterfactual case (in the absence of A, then not B). In three experiments, children's capacity for such counterfactual thinking was assessed. In Experiment 1, children aged 3–5 years observed a sequence such as A causing B. Subsequently, they replied quite accurately to a question about a counterfactual sequence, for example: “What if A had not occurred, then B or not B?”. In Experiment 2, children were asked about two counterfactual antecedents, one that would not have caused B, and one that (like the actual antecedent) would also have caused B. Children differentiated between the two types of antecedent. Finally, in Experiment 3, children heard stories in which the protagonist chose a course of action that led to a minor mishap (e.g., drawing with a black pen and getting inky fingers), having rejected an option that would have prevented it in experimental stories (e.g., using a pencil) or an option that would have led to an equivalent outcome in control stories (e.g., using a blue pen). Children aged 3 and 4 years often cited the failure to adopt another course of action as the cause of the mishap and, particularly in experimental stories, they focused on the rejection of the alternative option. Children's use of counterfactual thinking is discussed in relation to contemporary accounts of causal reasoning.

This technical report describes a new language for describing constraint satisfaction problems, called EaCL. Traditional Constraint Programming Languages have been built on top of host languages such as Prolog, Lisp, C++. This means that the user must have reasonable knowledge of the syntax and semantics of the host language before being able to use the constraint technology effectively. On top of this, the user may also be required to specify the heuristics and, or algorithm to solve the constraint problem. This leads to a bottleneck in the amount of people who have the necessary expertise in both constraint programming and the host language to implement practical systems, which use constraint satisfaction techniques. Our language is designed to abstract out as many of these details as possible, to produce a high level language, where the problem specification is the focus. With this in mind, our language is designed to be simple, high level, intuitive and declarative (the order in which constraints are specified has no significance).

Personnel scheduling is a very practical problem. It is widely studied because solutions to it can be generalized to many other problems. This paper describes ZDC-Rostering, a powerful constraint-based tool for personnel scheduling. ZDC-Rostering is based on a computer-aided constraint programming package called ZDC, which decouples problem formulation (or modelling) from solution in constraint satisfaction problems, and provides a set of constraint satisfaction algorithms, including complete and incomplete algorithms, to users who are not required to learn how these algorithms work. ZDC allows us to focus on constraint modelling in the rostering problem, which is expressed by a simple declarative language called EaCL. The simplicity of EaCL allows users with basic training in programming and problem specification to add new constraints easily. Solvers supplied in ZDC include a generalized Forward Checking solver, a Linear Programming solver and local search solvers implementing Guided Local Search, Tabu Search and Genetic Algorithms. Our experiments show that Guided Local Search is capable of solving realistic and very tightly constrained problems efficiently.

This technical report describes a new language for describing constraint satisfaction problems, called EaCL. Traditional Constraint Programming Languages have been built on top of host languages such as Prolog, Lisp, C++. This means that the user must have reasonable knowledge of the syntax and semantics of the host language before being able to use the constraint technology effectively. On top of this, the user may also be required to specify the heuristics and, or algorithm to solve the constraint problem. This leads to a bottleneck in the amount of people who have the necessary expertise in both constraint programming and the host language to implement practical systems, which use constraint satisfaction techniques. Our language is designed to abstract out as many of these details as possible, to produce a high level language, where the problem specification is the focus. With this in mind, our language is designed to be simple, high level, intuitive and declarative (the order in which constraints are specified has no significance). This technical report subsumes previous technical reports on EaCL.