Range Image Aided Inertial Navigation Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inertial navigation systems (INS) suffer from integration drift, leading to cumulative errors in position and orientation over time, necessitating external aiding data for correction.
Innovation Solution
A navigation system that integrates an Inertial Measurement Unit (IMU) with a Range Image (RI) sensor and a processor to perform a priori transformations of RI data using INS solutions, constructing delta pose registration cost gradients to determine INS error corrections and update the INS solution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If INS uses dead reckoning to calculate position and orientation, then navigation can operate without external sources, but integration drift causes cumulative errors to accumulate over time
Solution Approach 1:
The patent implements a feedback mechanism by using range image data to detect and correct INS errors. The system continuously compares expected range image measurements with actual measurements, generates innovation signals from the differences, and feeds these corrections back to adjust the INS state estimates, thereby preventing error accumulation over time
Solution Approach 2:
The patent introduces range image data as an intermediary aiding source between the INS and external environment. This intermediary provides independent measurement information about position and orientation that does not rely on GNSS or other traditional external sources, enabling error correction through a novel measurement channel
2Reliability
If INS is aided by external sources like GNSS, then measurement errors can be corrected, but the system requires line-of-sight to satellites which is not always available
Solution Approach 1:
The patent makes the navigation system universally applicable across diverse environments by using range image data as a universal aiding source that functions independently of satellite availability. The same range image-based correction mechanism works in urban canyons, indoor environments, and outdoor areas, providing consistent error correction capability regardless of location
Solution Approach 2:
The patent changes the fundamental parameter of what external aiding data is used - transitioning from electromagnetic satellite signals (GNSS) to optical/range image measurements. This parameter change enables operation in environments where satellite signals are blocked but visual/range information is still available
3Measurement precision
If traditional INS/GNSS integration is used, then navigation accuracy can be maintained when satellites are visible, but the system fails in urban canyons and indoor environments
Solution Approach 1:
The patent introduces range image data as an intermediary that bridges the gap between INS and the environment in situations where GNSS cannot serve as the intermediary. The range images provide environmental information that can be used to correct INS errors even when satellite-based intermediaries are unavailable
Solution Approach 2:
The patent performs preliminary transformation of range image data into a format suitable for INS correction before the actual navigation update. By pre-processing the range images and transforming them into the INS reference frame, the system prepares correction data in advance that can be immediately applied when needed, ensuring continuous accuracy
Data Source
AI summary
A navigation system includes an IMU, a navigation estimator configured to estimate a current navigation solution based on (i) a previous navigation solution, and (ii) a specific force vector and an angular rate vector measured by the IMU, an RI sensor, an RI data preprocessor configured to perform an a priori transformation of RI data acquired by the RI sensor using the current navigation solution to obtain transformed RI data, an RI map database configured to retrieve a valid keyframe map based on the transformed RI data, and an RI filter manager (RFM) configured to construct a map registration cost gradient (MRCG) measurement based on (i) the transformed RI data, and (ii) the known position and the known orientation of the valid keyframe map. The navigation estimator is further configured to determine an absolute navigation solution based on at least (i) the current navigation solution, and (ii) the MRCG measurement.