Frame-to-Geospatial Coordinate Mapping for GPS-Denied Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for obtaining geospatial coordinates from sensor data, such as GPS, are inaccurate in environments where GPS signals are unavailable or unreliable, and fail to account for sensor operations like zooming or rotation.
Innovation Solution
A computing system that analyzes sensor data from successive frames to determine transformations like translation, rotation, and zooming, and maps these transformations to GPS coordinate changes, allowing accurate estimation of GPS coordinates in frames with missing or inaccurate data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS sensor is used to obtain geospatial coordinates, then coordinates can be obtained in open areas, but coordinates are unavailable or unreliable in tunnels and GPS-denied environments
Solution Approach 1:
The patent introduces an intermediary system consisting of visual inertial odometry and map matching algorithms that bridge the gap when GPS is unavailable. The system uses camera images and inertial measurement unit data as intermediate information sources to estimate position, then matches these estimates against pre-built maps to recover accurate geospatial coordinates in GPS-denied environments like tunnels.
Solution Approach 2:
The system performs preliminary actions by pre-building maps of the environment using GPS coordinates and visual data before the vehicle enters GPS-denied areas. This preliminary mapping allows the system to later match visual features against the pre-built map to recover position information when GPS is unavailable, effectively preparing the solution in advance.
2Productivity
If simple interpolation methods are used to estimate coordinates in GPS-denied frames, then computational cost is reduced, but accuracy of coordinate estimation deteriorates
Solution Approach 1:
The patent replaces simple mechanical interpolation methods with a more sophisticated system that substitutes visual feature matching and map comparison algorithms. Instead of merely extrapolating from previous GPS points, the system uses computer vision to identify landmarks and match them against the pre-built map, providing accurate position estimation without excessive computational burden.
Solution Approach 2:
The system employs a multi-functional approach where the same visual inertial odometry engine serves multiple purposes: it tracks position continuously, identifies landmarks for map matching, and provides orientation information. This universal engine handles both GPS-available and GPS-denied scenarios, maintaining accuracy across different operating conditions without requiring separate specialized systems.
3Device complexity
If transformations like zooming and rotation are not accounted for, then processing is simpler, but accuracy of coordinate mapping deteriorates
Solution Approach 1:
The patent applies dynamic transformation models that adapt to the actual camera operations performed between frames. Instead of assuming fixed or simple transformations, the system dynamically calculates the actual translation, rotation, and zoom factors by comparing visual features across frames, then applies these dynamic transformations to the map matching process to maintain accuracy despite complex camera movements.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Systems and methods are provided for one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the system to perform: receiving successive frames of sensor data, the successive frames comprising a first frame and a second frame; determining transformations, in sensor coordinates, between coordinates of corresponding elements in the successive frames; determining a mapping between the transformations in sensor coordinates and transformations in geospatial coordinates of the corresponding elements in the successive frames; and determining second geospatial coordinates of the corresponding elements of a third frame based on: a transformation between the second frame and the third frame, and the mapping.