Unmanned aerial vehicle electric quantity prediction device based on multi-parameter environment perception

Abstract

The invention discloses an unmanned aerial vehicle electric quantity prediction device based on multi-parameter environment perception. The device comprises a data acquisition module, a data preprocessing module, a time-varying data extraction module and an electric quantity prediction module. By integrating the data acquisition module, the data preprocessing module, the time-varying data feature extraction module and the electric quantity prediction module, the problems that an existing prediction technology is low in accuracy and cannot adapt to different environments are solved, and high-precision prediction of the electric quantity of the unmanned aerial vehicle is achieved. According to the method, multi-parameter environment information and an advanced deep learning algorithm are fully integrated, a channel attention mechanism is introduced to introduce different proportions to the extracted features, higher prediction accuracy is realized in combination with the swing-transform and the LSTM, and an effective guarantee is provided for task planning and safe flight of the unmanned aerial vehicle.

Images