Target tracking: a real-time object oriented design experiment

Developers of modern computer based systems are employing object oriented (OO) design methodologies and supporting tool sets with the goal of reducing life cycle costs and improving software quality. A variety of OO methodologies are emerging, with some more suited to information system design, others focused on real-time applications, and the rest of a general purpose nature. The Real-Time Object Oriented Modeling (ROOM) technique has been optimized for application specifically to real-time, distributed applications. Its notation has been shaped by a decade of design experience in the telecommunications industry, and is rapidly gaining credibility across the entire spectrum of real-time system application domains. Unlike other OO techniques the ROOM notation has a formal notation which renders executable models at all levels of abstraction and achieves a substantial level of mechanically generated source code. The purpose of this paper is to expose the reader to the fundamentals of the ROOM design methodology and its supporting ObjecTime tool set, and to characterize the strengths of this technique when applied to the analysis and design of state-intensive, real-time, distributed computer-based systems. A sample system implementation is presented which is based on a simplified sonar tracking system. This sample problem has been limited in scope, but has all the design elements which challenge the system design process, including distributed concurrent processing, fault tolerance, and real-time performance. We conclude with an analysis of several inherent strengths and benefits of ROOM methodology applicable to the development of real-time, distributed computer systems