Though Object-Oriented Analysis, Design, and languages have become the dominant practices in many, or most, domains of software engineering, concerns about complexity, size, and performance in the embedded, real-time software domain have led to a prevalent view that OO technology is not suitable for the domain. We challenge this view through a successful application of OOA, OOD, and C++ (including STL) in the embedded, real-time flight software in an Earth-orbiting science instrument named Aquarius (see [1]). We've found that OOA and OOD with UML actually enhance communication with systems and hardware engineers. We also found that C++, thoughtfully used, need not lead to code bloat, and that its performance is every bit as good as that of C. We begin with an overview of the requirements and describe our overall use of UML modeling, followed by a discussion of the use of UML for Object-Oriented Analysis with use cases. Then the application of UML for high-level and detailed design, the use of frameworks supporting a component architecture and multi-platform execution, and code generation from UML detailed design are described. We also present the use of UML for organizing and designing and documenting our verification and test environment and scenarios, and using HTML, generated by our UML tool, for all documentation and for requirement traceability. Finally, we discuss the use of C++ as the implementation language, and give an overview of status and work metrics.

Machine learning is being applied to almost all corners of our society today. The inherent power of large amount of empirical data coupled with smart statistical techniques makes it a perfect choice for almost all prediction tasks of human life. Information retrieval is a discipline that deals with fetching useful information from a large number of documents. Given that today millions, even billions, of digital documents are available, it is no surprise that machine learning can be tailored to this task. The task of learning-to-rank has thus emerged as a wellstudied domain where the system retrieves the relevant documents from a document corpus with respect to a given query. To be successful in this retrieving task, machine learning models need a highly useful set of features. To this end, meta-heuristic optimization algorithms may be utilized. The aim of this work is to investigate the applicability of a notable meta-heuristic algorithm called simulated annealing to select an effective subset of features from the feature pool. To be precise, we apply simulated annealing algorithm on the well-known learning-torank datasets to methodically select the best subset of features. Our empirical results show that the proposed framework achieve gain in accuracy while using a smaller subset of features, thereby reducing training time and increasing effectiveness of learningto-rank algorithms.

Missions involving robotic space flight typically have a way to change the software that controls the flight system, or some part of it, such as an instrument, after launch. Usually this is accomplished by uplinking small sets of binary machine instructions and writing them to known locations in memory. We present an approach, used on the Aquarius mission, that involves replacing running components of, or adding components to, the running software at a higher logical level, specifically at the software architecture level, and on the C++ rather than machine-language level. This approach provides significant advantages in flexibility, robustness, reliability, and testability. We present the component-based flight software (FSW) design features that enable these capabilities. We then discuss the approach used to verify the robustness and reliability of these techniques, and finally describe usages to date.

No abstract