Research And Realization Of UAV Control Law Test Scenario Based On Monte Carlo Method

The control law is regarded as the core of the UAV flight control system,and its quality is considered to be the key to ensure the normal operation of the aircraft.Therefore,testing the flight control law is an important means to ensure its safety and stability.In the previous airborne software testing,people paid more attention to the testing of the software itself,that is,testing the functions and performance of the software,but less research on actual flight scenario.Therefore,it is of great practical significance to study the automatic generation technology of test cases of flight scenarios.The testing of aviation software is used as the research background.Software Considerations in DO-178 B Airborne Systems and Equipment Certification are used as a standard for testing airborne control law software.The goal of software testing is to achieve the highest use case coverage with the fewest test cases.Firstly,the different flight scenarios of the UAV are divided according to the flight stage,and the flight parameters are analyzed and processed;Secondly,in order to simulate the randomness of the input data during the flight process,the generation method of the random number is studied and the quality of the random number is tested;Finally,the existing combined test case generation methods are compared,the Ant Colony Optimization（ACO）is selected and improved to generate combined test cases.The MC/DC logic coverage test case is added.Based on the test case generation algorithm,an automatic test case generation tool is designed and applied to the coverage verification of the actual flight control law model.The main research contents are as follows:（1）The flight scenarios are divided into reference flight scenarios,special flight scenarios and invalid flight scenarios.The input parameters of control laws in different scenarios are analyzed and processed,and the random number generation process is analyzed to provide data for the generation of combined test cases.enter.（2）The combined use case generation algorithm is compared,and a Greedy-ACO is proposed.Dynamic heuristic information is added to the ant colony search process to enhance the local search ability of the ants and to improve the quality of the use case generated by the algorithm.Experiments show that the improved ant colony algorithm（Greedy-ACO）is more suitable for parameter combination testing of aviation software.The generation method of MC/DC logic coverage test case is studied.（3）The Greedy-ACO is developed as a test case automatic generation tool.The flight control law of a certain type of fixed-wing UAV is used as the test object,and verify through multiple Monte Carlo simulations that the test case generated by the improved algorithm has a higher performance.coverage.