GNSS/INS Navigation Using Vision Tie Point Buffering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional GNSS/INS systems face accuracy issues when using partially occluded images for navigation, as they require a minimum number of tie points to provide accurate location updates, leading to instability and loss of navigation information in situations like multipath conditions or occlusions from large vehicles or geographic features.
Innovation Solution
A novel system that integrates a vision system with a GNSS/INS system, utilizing an error correction module to generate inertial system adjustment information from partially occluded images, allowing the system to provide accurate navigation even with fewer tie points by combining inertial and vision system location information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the vision system requires a minimum number of tie points for each image to provide navigation updates, then the measurement precision is improved, but the reliability deteriorates when images are occluded
Solution Approach 1:
The system performs preliminary actions by accumulating tie points from multiple images before a complete occlusion occurs. The buffer stores partially collected tie point data from images that were acquired before the occlusion event, allowing the system to maintain navigation updates even when subsequent images are completely blocked.
Solution Approach 2:
The buffer acts as a cushion that absorbs the impact of occluded images. By pre-storing tie point observations from non-occluded images, the system creates a buffer that compensates for the loss of data during occlusion events, preventing complete system failure and maintaining navigation reliability.
2Measurement precision
If the system uses more tie points from each image to resolve unknowns, then the measurement precision improves, but the loss of information increases when occlusions occur
Solution Approach 1:
The system performs preliminary data collection by gathering and storing tie points from multiple images in advance. This buffer of pre-collected information ensures that even when occlusions prevent new tie points from being acquired, the system still has sufficient historical data to maintain accurate navigation updates.
Solution Approach 2:
The system transitions from relying on spatial distribution of tie points within a single image to utilizing temporal distribution across multiple images. By buffering tie points from a sequence of images and using their temporal order, the system compensates for spatial occlusions and maintains information flow.
3Measurement precision
If the vision system requires complete images with sufficient tie points for each update, then the measurement precision is maintained, but the productivity decreases when occlusions occur
Solution Approach 1:
The system performs preliminary accumulation of tie points from multiple images before navigation updates are needed. This buffer of pre-collected data allows the system to generate navigation updates continuously even when new images are occluded, maintaining both precision and productivity.
Solution Approach 2:
The buffer enables continuous navigation updates by maintaining a supply of valid tie point data from previous images. Instead of interrupting the update process when occlusions occur, the system continuously draws from the buffer, ensuring uninterrupted navigation information flow and maintaining productivity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for augmenting a GNSS/INS system by using a vision system is provided. The GNSS system generates GNSS location Information and the INS system generates inertial location information. The vision system further generates vision system location information that is used as an input to an error correction module. The error correction module outputs inertial location adjustment Information that is used to update the inertial system's location information.