自动程序修复技术对软件调试和软件维护具有重要的理论和实际应用价值。本项目从蜕变测试技术的角度入手来研究自动程序修复问题，旨在形成基于蜕变测试技术的自动程序修复方法体系，以扩展自动程序修复技术的应用领域以及推动其在实际应用中的影响力。通过利用蜕变测试技术的不同特点，主要研究内容包括：1）提出基于蜕变关系的输入机制来为程序修复提供丰富的输入信息；2）提出基于蜕变关系的补丁评价机制来对程序修复过程中所产生的候选补丁进行合理评价和划分；3）提出基于蜕变测试技术的综合程序修复算法以获得更好的修复效率。本项目的研究成果可以提高自动程序修复技术的实际应用性以及其修复效率。进一步，本项目的理论研究成果以及应用数据能够为后续相关研究工作提供一定的借鉴。

Automated program repair (APR) is of important theoretical significance and great application value to both software debugging and software maintenance. This project studies automated program repair from the perspective of metamorphic testing (MT), aiming at establishing methodologies of MT based APR so as to extend the application domain of APR as well as to increase the effects of APR in real applications. By leveraging different characteristics of MT, this project proposes: 1) an metamorphic relations based input mechanism in order to provide rich input information for APR; 2) an metamorphic relations based patch evaluation mechanism in order to reasonably evaluate and classify candidate patches generated during the repair process; 3) an MT based program repair algorithm in order to achieve better repair effectiveness. The research results of this project will increase the applicability and effectiveness of APR. Furthermore, the theoretical results and experimental data will provide guidance for related studies in future.