Evolutionary Testing of Autonomous Software Agents

A system built in terms of autonomous agents may require even greater correctness assurance than one which is merely reacting to the immediate control of its users. Agents make substantial decisions for themselves, so thorough testing is an important consideration. However, autonomy also makes testing harder; by their nature, autonomous agents may react in different ways to the same inputs over time, because, for instance they have changeable goals and knowledge. For this reason, we argue that testing of autonomous agents requires a procedure that caters for a wide range of test case contexts, and that can search for the most demanding of these test cases, even when they are not apparent to the agents’ developers. In this paper, we address this problem, introducing and evaluating an approach to testing autonomous agents that uses evolutionary optimization to generate demanding test cases.

Knowledge mediation in software quality engineering

The risk of failure of the software development process remains high despite many attempts to improve the quality of software engineering. Contemporary approaches to process assurance, such as the capability maturity model have not prevented systemic failures, nor have project management methodologies provided guarantees of software quality. The paper proposes an approach to software quality assurance based on a knowledge mediated concurrent audit, which incorporates essential feedback processes. Through a tightly integrated approach to quality audit, programmers would be empowered to use any chosen methodology to advantage, supported by intelligent monitoring of the essential interactions which occur in the development process. An experimental application implementing some aspects of the proposal is described

Foundation software engineering practices for capstone projects and beyond

This paper outlines an introductory course in software engineering undertaken by computing students at Deakin University. The course focuses on areas of that are often not emphasized in introductory software engineering. These areas include formal problem solving, management of software development, communication, and working in groups. The approach includes the application of concepts in a collaborative group setting. Enhancements to the course have been made over a number of semesters and evidence suggests that our overall approach is working. The skills acquired by students taking this course can be applied to later courses in software engineering, particularly the capstone project course, and to professional life

Solar aided power generation from coal fired power stations: THERMSOLV software

Although the thermodynamic advantages of using solar energy to replace the bleed off steam in the regeneration system of Rankine cycle coal fired power stations has been proven theoretically, the practical techno/economic feasibility of the concept has yet to be confirmed relative to real power station applications. To investigate this concept further, a computer modelling software “THERMSOLV” was developed by Deakin University researchers, together with the support of the Victorian power industry and Australian Research Council (ARC). This newly developed software simulates the steam cycle to assess the techno/economic merit of the solar aided concept for various power station structures, locations and local electricity market conditions. Two case studies, one in Victoria Australia and one in Yunnan Province, China, have been carried out to show the application of the software. This paper reports the structure and functions of the software, and the results of the two case studies.

Automatic refinement of user requirements : a case study in software tool evaluation

This paper presents an assessment of system effectiveness in automatic requirements refinement by comparing results obtained from experts and novices with those achieved by the system. As the investigated system was a combination of a tightly inter-connected methods and a tool, the evaluation framework melded together a number of distinct methodological approaches structured into three empirical studies, which aimed at the construction of a case problem domain, calibrating the system using this defined domain elements and finally using the calibrated system to assess its effectiveness. In consequence, it was concluded that the evaluated methods and tools were effective in supporting requirements refinement.

Assessment management using software

Web based learning is providing new and innovative environments to facilitate the delivery of course material to increasing numbers students, but most of the effort is centered around the presentation of the learning material and creating environments where student interact with this material and each other. Despite the changes the web based learning environments have delivered, little attention has been given to improving the management and process of assessment. This paper introduces a new software application designed to take advantage of the web based environments and addresses the management overheads associated with assessment whilst improving the feedback to students. Initial investigations indicate significant benefits can be achieved by using software to manage the assessment process. A detailed description of the software is provided and two case studies presented that highlight the software's application and confirmed benefits when used to conduct and manage assessments.