Detector landing point positioning method based on spiking neural network

Abstract

The invention discloses a detector landing point positioning method based on a pulse neural network. The method comprises the following steps: preprocessing meteorite crater DEM data on the surface of a star; training a training set data by adopting a U-Net network architecture based on an artificial neural network model ANN; converting the UNET of the artificial neural network model ANN into a UNET model of a pulse neural network model SNN; coding the image, and carrying out test set detection by adopting the UNET model based on SNN; and carrying out meteorite crater matching and positioning through a template matching algorithm, and constructing a terrain landmark library to realize positioning. The spiking neural network is applied to the field of deep space exploration for the first time, and positioning of the landing point of the detector is achieved by identifying the meteorite crater.

Description

technical field [0001] The invention belongs to the technical field of vision-based descent segment autonomous navigation of spacecraft in deep space exploration missions, and in particular relates to a method for locating a landing point of a probe based on a pulse neural network. Background technique [0002] After entering the 21st century, the space powers have launched a series of deep space exploration missions. Among them, realizing the soft landing of the probe on the surface of the target star to ensure the safety of equipment and personnel is the basis for our country to explore deep space and develop aerospace technology. Usually, the detector positioning technology is through the ground measurement and control station or GNSS navigation technology. However, in the deep space environment, this technology has a large delay, and there are also problems such as star occlusion and signal transmission blockage. Therefore, autonomous safe landing technology is a necessa...

Images