Grid-Based Terrain Map Coding for Robust Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for terrain navigation using digital elevation models (DEMs) are limited by their reliance on Cartesian coordinates, which are not robust for error correction and do not effectively integrate sensory information into phase-space representations, essential for accurate localization.
Innovation Solution
A method and system that encode DEMs into grid cells parameterized by spatial scale, orientation, and 2D offset, aggregating them into modules to generate phase codes, which are then used to create a coincidence map for accurate location estimation based on sensor data along a terrain trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Cartesian coordinates are used for terrain navigation, then the representation is simple and intuitive, but the system lacks robustness for error correction and cannot effectively integrate sensory information into phase-space representations
Solution Approach 1:
The patent transforms the coordinate representation from simple Cartesian coordinates to a phase-space parameterized by spatial scale, orientation, and 2D offset. This parameter transformation enables the system to achieve robust error correction capabilities while maintaining computational tractability through the structured phase-space formulation.
Solution Approach 2:
The invention introduces phase-space as an additional dimensional framework beyond traditional 2D terrain coordinates. By parameterizing grid cells in phase-space with dimensions including spatial scale, orientation, and offset, the system achieves more robust error correction and sensory integration capabilities.
2Measurement precision
If grid cells are aggregated into modules with multiple parameters, then the phase-space representation becomes more robust for localization, but the computational complexity increases
Solution Approach 1:
The patent segments the phase-space parameterization into distinct modular components: spatial scale, orientation, and 2D offset. Each grid module is characterized by these separate parameters, allowing independent processing and aggregation while maintaining overall localization accuracy. This segmentation reduces computational complexity by enabling modular computation.
Solution Approach 2:
The grid modules serve multiple functions simultaneously: they encode spatial information, provide error correction through their structured parameterization, and enable efficient sensory data integration. The same phase-space representation framework handles multiple localization tasks without requiring separate computational systems.
3Measurement precision
If traditional Cartesian coordinate methods are used, then the system is easier to implement, but it cannot effectively decode sensor data into accurate location estimates
Solution Approach 1:
The patent introduces phase-space as an intermediary representation between raw sensor data and final location estimates. The phase-coded terrain map serves as a mediator that transforms elevation data into a format optimized for decoding sensor measurements, improving location estimation accuracy while maintaining reasonable implementation complexity through structured phase-space operations.
Data Source
AI summary
Encoding terrain maps is provided. The method comprises receiving a digital elevation model (DEM) of a terrain and encoding the DEM into grid cells parameterized by spatial scale, orientation, and 2D offset. Grid cells with shared scale and orientation are aggregated into grid modules. Locations from the DEM that correspond to a given elevation produce a contour line of locations that fall within the given elevation. 2D phase codes are calculated for each grid module to produce a phase candidate dictionary, wherein a subset of phase codes comprises phase candidates corresponding to locations from the contour line. When sensor data is received along a trajectory over the terrain, the phase candidate dictionary is queried. Phase candidates are corrected for relative displacement from a reference point and summed to produce a coincidence map over the DEM that identifies a current location estimate over the terrain.