IMU Calibration via Vision Sensor Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inertial measurement units (IMUs) suffer from accumulated error, leading to measurement drift in navigation systems, which results in an increasing gap between estimated and actual vehicle locations.
Innovation Solution
Integration of a vision sensor with the IMU to calibrate inertial sensors by estimating error values based on vision sensing input data, thereby adjusting subsequent inertial sensing input data to minimize drift.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an IMU is used for navigation or tracking purposes, then the system can measure linear accelerations and rotational velocities to track position and velocity, but the system suffers from accumulated error leading to measurement drift
Solution Approach 1:
The patent implements a feedback mechanism where vision sensor data continuously monitors the actual position and orientation of the apparatus, and this information is fed back to correct drift errors in the IMU measurements. The system compares IMU-based estimated position with vision-based actual position, calculates the deviation, and applies corrections to compensate for accumulated errors over time.
Solution Approach 2:
The patent introduces a vision sensor as an intermediary measurement source that serves as a reference for calibrating the IMU. The vision sensor captures images to determine actual position and orientation, acting as a mediator that bridges the gap between the drifting IMU measurements and the true physical state of the apparatus, enabling error estimation and correction.
2Measurement precision
If a vision sensor is integrated to calibrate the IMU, then measurement drift is reduced and accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent makes the vision sensor serve multiple functions: it acts as both a navigation aid for determining position and orientation, and as a calibration reference for correcting IMU drift. By utilizing the same vision sensor for both navigation and calibration purposes, the system avoids adding separate calibration hardware, thereby reducing overall device complexity while maintaining measurement accuracy.
Data Source
AI summary
A method is provided for calibrating an inertial sensing unit of a device utilizing a vision sensing unit integral to the device. The method includes receiving inertial sensing input data from the inertial sensing unit, receiving vision sensing input data from the vision unit, and determining when the received vision sensing input data represents a predetermined input state of the vision sensing unit. The method includes estimating an error value in the inertial sensing input data received from the inertial sensing unit based on the received vision sensing input data upon determination that the received vision sensing input data represents the predetermined vision sensing input state. The method further includes adjusting first subsequent received inertial sensing input data from the inertial sensing unit based on the estimated error value, thereby calibrating the inertial sensing unit.