Autonomous Neural Network Almanac for Spacecraft Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current space navigation systems rely on ground-based control methods that are inefficient due to outdated navigation data and long communication lag, leading to increased uncertainty in spacecraft state and delayed course corrections.
Innovation Solution
An autonomous neural network navigation system that uses an onboard computer to determine the spacecraft's current navigation state, read neural network model parameters from an almanac, and execute the corresponding neural network model to provide thrust commands, enabling real-time navigation and immediate course corrections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ground-based control methods are used for space navigation, then navigation commands can be generated with comprehensive calculation capabilities, but communication lag and outdated navigation data lead to delayed course corrections and increased uncertainty in spacecraft state
Solution Approach 1:
The spacecraft navigation system performs autonomous navigation operations using onboard neural network models. The system determines its own navigation state, selects appropriate neural network models from the almanac based on current epoch, and generates thrust commands independently without requiring ground-based control intervention, thereby eliminating communication lag and enabling immediate course corrections
Solution Approach 2:
Multiple neural network models are pre-trained and stored in the almanac for different epochs and navigation scenarios before the spacecraft mission. This allows the onboard system to quickly retrieve and execute the appropriate pre-trained model without requiring real-time ground-based calculations, thus reducing time loss while maintaining navigation accuracy
2Measurement precision
If multiple neural network models are stored in the almanac for different epochs, then navigation accuracy is improved through model selection, but memory requirements and system complexity increase
Solution Approach 1:
The navigation problem is segmented into multiple epochs, with each epoch having its own dedicated neural network model stored in the almanac. The system divides the trajectory into discrete time segments and selects the appropriate model based on the current epoch, improving navigation accuracy while keeping each individual model relatively simple and manageable
Solution Approach 2:
The system manages complexity by changing the parameter of model selection based on epoch rather than maintaining a single complex model. The almanac stores models with different parameters optimized for specific epochs, allowing the system to switch between simpler specialized models rather than using one complex general model, thus improving accuracy without proportionally increasing overall system complexity
Data Source
AI summary
An almanac method for autonomous neural network navigation of a spacecraft includes performing steps via a computer onboard the spacecraft for determining a current navigation state of the spacecraft and determining a current target epoch for the spacecraft based on the current navigation state. The method also includes reading neural network model parameters from an almanac for the current target epoch. The almanac contains a plurality of neural network models, with each model being valid for an epoch corresponding to a portion of a trajectory. The method further includes executing the neural network model corresponding to the current target epoch to provide one or more thrust commands for the spacecraft.