The Research And Application Of Self-Health Management Technology For Spacecraft

With space technology advancing, the number of in-orbit spacecrafts ever-increasing and new aerospace applications continuing to emerge, the traditional operation mode for spacecraft missions has been difficult to satisfy the current development of the aerospace industry. To satisfy the demand for new aerospace applications, improve the survival and anti-destruction capability, lower data transfer overhead between aerospace and ground station and lower expenses of artificial maintenance, the development of autonomous spacecraft is imperative. Spacecraft autonomy management includes autonomous planning and scheduling, self-health management and autonomous payload data processing. Self-health management is the safeguards of spacecraft's autonomous, safe and stabe orbiting.In order to realize a new framework SASF (Spacecraft Autonomy Software Framework) with self-health management support, we firstly introduce spacecraft technology and focus on the analysis of self-health management technology; secondly analyze the technology of the existing components and frameworks and their applications in spacecraft fields. Based on this, we propose the framework which uses design patterns, interfaces, components and other technology to provide framework level software development support for spacecraft autonomous planning and scheduling and self-health management. The paper outlines the main architecture of the framework, and focuses on the composition, design and implementation of self-health management part. The self-health management part's design and realization provides framework level support for error detection, fault location, self-decision-making and failure recovery which reduces the coupling between the various modules of the software and the complexity of software realization. SASF improves the efficiency of spacecraft software development and reusability.Faced with component health problems caused by instantaneous hardware fault, based on the idea of software implemented hardware fault-tolerance, we suggest a component-based data flow model, define the data flow graph and analyze the model. Then in order to lower dual redundant fault tolerance system's great overhead, based on the data flow model, we propose two error detection algorithms, a fault location algorithm and a kind of self-decision-making algorithm and expound the principles of these algorithms. The two error detection algorithms lower system operating overhead, as well as ensure the error detection coverage while reducing the number of redundant nodes; fault location algorithm lowers fault tolerance latency by reducing fault location nodes; self-decision-making algorithm by realizing error forecast handling and dynamic adjustment of redundant nodes lowers fault tolerance latency and imporves error detection flexibility. Finally, we implement the data flow model and its associated algorithms based on SASF.