A new integrated navigation system for the indoor unmanned aerial vehicles (UAVs) based on the neural network predictive compensation

Abstract

Aiming at the problem that the reliability of data fusion in the unmanned aerial vehicle navigation system will be drastically reduced when the environmental characteristic changes, this paper proposes a new algorithm to address the problem based on the prediction and compensation of neural network. First, the Extended Kalman Filter and particle filter are used for the data fusion of laser and optical flow sensor. And then a Radial Basis Function (RBF) Neural Network is used to estimate the error of the particle filter. When the laser data is reliable, RBF Neural Network converts into the learning mode to train the model, and when the laser data is interrupted or unreliable, the system is compensated by using the trained model. The experimental results show that the RBF neural network model can effectively improve the reliability of the UAV navigation information when the environment characteristic changes, which prove the validity of the algorithm, proposed in this paper.