Research On Testing Methods Of Aerospace Visualization Software

With the rapid development of China's software industry,software quality problems emerge one after another.In the actual project development process,software testing has not been paid attention to,and the hidden defects in the software have become security risks.Aerospace software is more complex than general software and has the heavy responsibility of completing major national space missions.It has strict requirements on safety,reliability and real-time.This means that general software testing methods are not suitable for aerospace software.This paper mainly studies the MDOP testing method,aiming at obtaining the testing method suitable for space visualization software.The main research contents of this subject include the following three parts.In the first part,the common code defects of C/C++ which is the mainstream programming language of aerospace software are analyzed.Combined with the focus of common static testing tools,C++Test and PC-Lint are selected to detect runtime errors,structural errors,security failures and code logic of the MDOP.The results of the two tests are compared and supplemented to form a perfect static test report.The second part is functional testing.Design and write test cases according to functional requirements and test case design method(common test case design method,test case generation method based on UML activity diagram model,test case generation method based on finite state machine model).Data-driven test script development is used for each test case.For each test case,we use UFT(Unified Functional Testing)to create test scripts,enhance scripts(add transactions,parameterization,iteration,etc.),execute test scripts,and analyze test results.The product requirements,use cases,execution results and other information are input into Zen Dao project management software,and functional test reports are automatically generated,which standardizes requirements management,project management and test management.The third part is performance testing.Select test tools Silent Test Runner and Load Runner according to MDOP performance requirements.Import the enhanced test script into Silent Test Runner to generate a concise transaction summary report,and import it into LoadRunner to create and execute scenarios.Delay detection of image display,response time analysis and system resource analysis were carried out.The corresponding performance index data are obtained and the performance requirements are verified.Through the above research,the test methods suitable for the static testing,functional testing and performance testing of aerospace visualization software are obtained,which improves the level of automated testing.The test results also verify the practicability and feasibility of this study,and provide a reference for the future practice of space engineering software testing engineering in China.