MEMS gyroscopes error compensation method for micro satellite based on integration nerval net
A micro-satellite and neural network technology, applied in the inertial field, can solve the problems of low neural network prediction accuracy, poor stability, and low result accuracy, so as to improve stability and prediction accuracy, improve overall generalization error, and enhance generalization ability Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0027] The realization process of micro-satellite MEMS gyro error compensation method of the present invention is as shown in Figure 1, mainly comprises following four steps:
[0028] (1) Set the network scale and establish a neural network model
[0029] As shown in Figure 2, D is all learning samples, D i (i=1, 2,..., k) are not exactly the same multiple sets of learning samples, φ i (x)(i=1, 2, ..., k) is the learning sample D i The individual neural network after training, φ(x) is the integrated neural network model obtained after integrating the individual network. In the integrated neural network strategy, each base network is obtained by training with different self-service sample sets, and the final integrated prediction model is formed by aggregating network base predictor sets. Using the aggregation method of calculating the average value of all prediction results, each single base network obtained by training contains the transition rules from digital states to e...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com