Occupancy Grid Range Finding for GPS-Denied Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing positioning systems, such as GNSS and cellular networks, face challenges in providing accurate positioning for autonomous or semi-autonomous vehicles, especially in environments where satellite positioning is unavailable or unreliable.
Innovation Solution
An apparatus and method that utilize an occupancy grid of reference areas to process range finding data from radar, lidar, or other sources. This involves allocating data items to reference areas, summing magnitudes within each area, and applying a threshold to identify landmark candidates for navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If range finding data is processed using traditional positioning systems (GNSS, cellular networks), then positioning information can be obtained, but accuracy deteriorates in environments where satellite positioning is unavailable or unreliable
Solution Approach 1:
The patent introduces an occupancy grid as an intermediary data structure that maps range finding data to spatial locations. This grid serves as a mediator between raw sensor data and navigation decisions, allowing the system to process and interpret environmental information without relying on satellite positioning systems.
Solution Approach 2:
The patent replaces satellite-based electromagnetic positioning systems with a local sensor-based occupancy grid system. By substituting GNSS/cellular network dependency with radar/lidar-based environmental mapping, the system achieves reliable positioning in GPS-denied environments through local spatial awareness rather than external satellite signals.
2Loss of information
If all range finding data items are processed individually, then detailed information is retained, but processing time and computational complexity increase
Solution Approach 1:
The patent merges multiple range finding data items that map to the same occupancy grid cell by summing their magnitudes. This consolidation reduces the total number of data items that need to be processed individually, decreasing computational complexity and processing time while retaining essential information through aggregated magnitude values that indicate the presence and strength of reflections.
Solution Approach 2:
The patent segments the continuous spatial environment into discrete occupancy grid cells of specific size and shape. By dividing the environment into manageable reference areas, the system can efficiently process range finding data by allocating items to specific grid cells rather than handling all data points in a continuous space, thereby reducing computational burden.
3Reliability
If threshold is applied to select reference areas with integrated magnitudes, then spurious reflections are filtered out, but some potentially useful data may be excluded
Solution Approach 1:
The patent changes the parameter representation of range finding data by transforming individual reflection magnitudes into integrated magnitudes summed across multiple data items within the same occupancy grid cell. This parameter transformation allows the application of magnitude thresholds to filter spurious reflections while preserving genuine landmarks, as legitimate landmarks generate consistently high integrated magnitudes across multiple measurements.
Data Source
AI summary
According to an example aspect of the present invention, there is provided an apparatus configured at least to store information defining a set of reference areas, each reference area having a location, shape and geographic size, obtain range finding data which comprises plural data items, each data item having a location and a magnitude, allocate at least a part of the data items to the set of reference areas such that magnitudes of data items with locations in a same reference area are summed, and apply a threshold to select from among the reference areas a second set of reference areas with integrated magnitudes in excess of the threshold.