M. Schillaci

• Polytechnic University of Turin

The μGP approach is centered around the evolutionary core. The tool is composed by about 50;000 lines of C++ code, comprising 118 classes in 310 files. The project is hosted by Sourceforge on http://ugp3.sourceforge.net/.

The idea of Evolutionary computation implies the existence of suitable tools to perform computations. Such tools have to be designed pondering the environment in which they will operate and the problems to which they will be applied, together with the chosen evolutionary technique. Every design process implies choices, some of which may not be imme...

μGP conceptual design is based on three macro blocks: the evolutionary core, the external evaluator, and the constraints library. The evolutionary core cultivates one or more populations of individuals. The external evaluator assigns a fitness value to individuals. The constraints library defines the appearance of individuals.

In this chapter, we describe how an evolutionary run must be configured in order to cope with the main requirements the μGP needs.

The μGP generates individuals following the indications of the constraint library. The constraints not only limit the possible structure and content of an individual, but also participate in the mapping of its genotype to a phenotype.

The parameters for every population in an evolutionary run can be set independently from the others. In this way different experiments can be performed, making changes local to a single file, without affecting global parameters. The user may even want to prepare several different files specifying the population parameters, and then switch between t...

This chapter contains a number of examples to illustrate the use of μGP. In the following each example is described completely, so that each experiment can be easily repeated. Every section contains detailed information about the preparation of the settings file, of the population settings file and of the constraints file. In addition, it contains...

During the evaluation phase the μGP core generates the phenotypes of the individuals to evaluate, launches the external evaluator and waits the results.

The μGP is meant to be used either as a stand-alone application or as an embedded tool, inside scripts or even called by other applications. It has to be portable across different hardware and OS platforms. To obtain the maximum possible flexibility and portablity the tool is run using a traditional command line.

Finding a good solution to a problem using evolution can be a lengthy process. Practical, real-world problems often have huge search spaces, so many candidate solutions exist. The fitness function, especially in the case of multi-objective optimization, may be deceptive, meaning that a large number of possibilities have to be explored in order to o...

Evolution is the theory postulating that all the various types of living organisms have their origin in other preexisting types, and that the differences are due to modifications inherited through successive generations. Evolutionary computation is the offshoot of computer science focusing on algorithms inspired by the theory of evolution. The defi...

Many of the options that the μGP recognizes on the command line are also available in the settings file. This file contains general settings for the evolutionary tool. The parameters specified in this file relate to the general architecture of the evolutionary run, such as the number and type of populations to evolve, and more administrative aspect...

A large part of microprocessor cores in use today are designed to be cheap and mass produced. The diagnostic process, which is fundamental to improve yield, has to be as cost effective as possible. This paper presents a novel approach to the construction of diagnosis-oriented software-based test sets for microprocessors. The methodology exploits ex...

Traditional test generation methodologies for peripheral cores are performed by a skilled test engineer, leading to long generation times. In this paper a test generation methodology based on an evolutionary tool which exploits high level metrics is presented. To strengthen the correlation between high-level coverage and the gate-level fault covera...

The car market shows a clear trend towards an increasing presence of electronic devices in engine control systems, due to a strong market drive towards high-performance control systems. Empirical evidence shows that a conventional screening approach only targeting system functionality is not enough to reach the desired high-quality targets. In this...

Traditional test generation methodologies for peripheral cores resort heavily to low-level descriptions of the circuit, leading to long generation times. Methodologies based on high-level descriptions can only be used if a clear relationship exists between the measured high-level coverage and the gate-level fault coverage. Even in medium complexity...

This paper details an evolutionary tool targeted at increasing the diagnostic power of a set of assembly programs. The underlying evolutionary scheme is quite peculiar in some aspect and present interesting characteristics The effectiveness of the generated set has recently been demonstrated. Here the use of the tool is further motivated through a...

Research community has not investigated as deeply as necessary the test generation problem of peripheral modules inside a system-on-a-chip (SoC), yet. Testing process for a peripheral core requires two distinct but highly correlated tasks: peripheral configuration and peripheral exercising. The configuration task is usually performed by an assembly...

Test of peripheral modules has not yet been deeply investigated by the research community. When embedded in a system on a chip, peripheral cores introduce new issues for post-production testing. A peripheral core embedded in a SoC requires a test set able to properly perform two different tasks: configure the device in different operation modes and...

The ever increasing usage of microprocessor devices is sustained by a high volume production that in turn requires a high production yield, backed by a controlled process. Fault diagnosis is an integral part of the industrial effort towards these goals. This paper presents a new methodology that significantly improves over a previous work. The goal...

Test of peripheral modules has not been deeply investigated by the research community. When embedded in a system on chip, however, peripherals pose accessibility problems that may make traditional test approaches ineffective. In this paper an evolutionary methodology, based upon coverage metrics at high- level, is described to automatically generat...

The paper describes the attempt to cultivate programs to climb the nano hill, a contest with exceptionally tight parameters of the game called corewar. An existing tool, called muGP, has been exploited. Two genetic operators were added to tackle the peculiarities of the objective. The generated programs compared favorably with others, either manual...

The widespread use of cheap processor cores requires the ability to quickly point out the manufacturing process criticalities in an effort to enhance the production yield. Fault diagnosis is an integral part of the industrial effort towards these goals. This paper describes an innovative application of evolutionary algorithms: iterative refinement...

This paper presents an evolutionary compaction method for microprocessor test programs originally written in the form of a loop. First it is shown that the loop form is redundant in the number of execution of a specific instruction; then a novel compaction method for these test programs is detailed. The effectiveness of the approach is finally prov...

The ever increasing usage of microprocessor devices is sustained by a high volume production that in turn requires a high production yield, backed by a controlled process. Fault diagnosis is an integral part of the industrial effort towards these goals. This paper presents a novel cost-effective approach to the construction of diagnostic software-b...

Microprocessor technology is increasingly used for many applications; the large market volumes call for cost containment in the production phase. Process yield for processor production is, however, far from ideal. To increase it fault diagnosis is an important means, since it can allow both process characterization and product repair by the usage o...

Checking if microprocessor cores are fully functional at the end of the productive process has become a major issue. Traditional functional approaches are not sufficient when considering modern designs. This paper describes new improvements for an existing evolutionary algorithm, called µGP, able to generate Turing-complete programs; these are expl...

This paper presents an extended specification and a new flexible architecture for a successful evolutionary tool. Architectural choices will be detailed and their conceptual significance underlined. Implementation detail issues will be discussed.