Modeling And Simulation Of Virtual Test Scene For Intelligent Vehicle Based On UC-win/Road

Autonomous vehicle integrates the most advanced software and hardware technology,and the simulation and verification of its safety and stability is the focus of current research.At present,the test work on autopilot is still in the primary stage,that is,there are no scene,large-scale simulation test methods,lack of the overall framework and system,so it is necessary to establish an integrated and systematic simulation process.It also provides a high safety,large-scale simulation test and verification method for autonomous vehicle for typical extreme scenarios.First of all,in order to ensure the authenticity and convenience of the simulation system,this study has carried on the secondary development to the existing scene modeling and simulation software UC-win/Road,endows the experimental vehicle with more abundant and real autopilot function.More software characteristics and the need of simulation scenarios put forward a new modeling process,improve the modeling efficiency,and lay the foundation for large-scale extreme scene modeling.Secondly,the flow chart of virtual test system for autonomous vehicle is designed.According to the characteristics of the autonomous vehicle,the testing process of unit testing,integration testing and system testing is designed to ensure the efficiency and comprehensiveness of the test.According to the characteristics of extreme condition scenarios,the modeling elements are re-classified to highlight the generation elements of extreme conditions,and the impact of scene elements on vehicles is briefly analyzed.Thirdly,the traffic scene is complex and changeable,and its components are complex.in order to simplify the workload of scene modeling,it is necessary to highlight the main factors that affect the decision-making control of autonomous vehicle.Therefore,this paper uses the grey relational analysis method,taking the number of injuries and deaths as the characteristic variables,finds out the scene elements which have a significant impact on the characteristic variables,and establishes a simple typical extreme working condition scene database.Finally,the scene model,vehicle model and control model are integrated for joint simulation.In this paper,a typical extreme scenario,"ghost probe",is designed to test the automatic emergency braking function of autonomous vehicle.The performance of the self-driving vehicle is evaluated by comparing the speed,acceleration,braking strength and other parameters produced under the two driving modes of self-driving and manned driving.