Intelligent perception of tandem flapping wings' flow field environment with sparse detectors

Abstract

Bionic flapping wing vehicles have great potential for civil and defense applications due to their flexibility and concealment at low Reynolds numbers. Since traditional flow field pattern recognition methods are difficult to identify effective information from the measured local flow field and deduce the state information of the moving body, this study uses an artificial intelligence method to establish the internal correlation between flow field pattern and state information. Specifically, a fully connected neural network is adopted to recognize the tandem flapping wings' flow field pattern by using different data acquisition methods and detector array distribution methods. Compared with the neural network based on time series data, the neural network based on spatial distribution data can realize the real-time judgment of flow field environment, which is closer to the real-time requirements in practical applications. In the paper, the intelligent perception of multi-flapping wings' flow field environment with sparse detectors is carried out and lays the theoretical foundation for autonomous navigation and obstacle avoidance of flapping wing aircrafts.