?
Интегрированная среда моделирования для верификации и валидации программ управления подключенными и высокоавтоматизированными транспортными средствами
Highly automated and connected vehicles are gradually entering the market. Currently, solutions are being proposed that allow these technologies to be used for cooperative driving automation, which can significantly improve traffic safety. Such technologies and their software should be tested to ensure safety before being implemented in real systems. Verification and validation of vehicular control software in the real world are difficult. Therefore, simulation computer modeling is used for this purpose. The simulation of connected and automated vehicles requires the combined use of traffic flow models, vehicle dynamics models, and automotive communication network simulators. Modern tools exist in these areas, but they are difficult to combine together, or they do not fully cover the technology domain. This article analyzes the requirements for an integrated simulation environment for modeling connected and automated vehicles and cooperative driving automation with highly detailed consideration of the influence of surrounding objects. To this end, the existing problems and practices were analyzed. The tools CARLA, OpenCDA, SUMO, OMNeT++, Artery are considered. Taking into account the disadvantages of existing methods, the paper proposes the architecture of the CAVISE integrated modeling environment with full coverage of the subject area using the chosen open-source tools, which includes a CAPI (CAVISE API) interface between existing tools for modeling connected and automated vehicles and their software in a controlled environment. A detailed description of the developed interface and the results of its testing for the verification of connected automated vehicle software using cooperative perception algorithms are provided. The developed interface allows, for the first time among open-source tools, to simulate connected automated vehicles, simultaneously taking into account real algorithms and software for machine perception and information exchange over wireless communication channels. The findings can be further used in the research and development of technologies and software for connected and automated vehicles.