Multi-IMU Calibration Using Rigid-Body Motion Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor calibration techniques for autonomous systems, such as autonomous vehicles, are time-consuming, require extensive infrastructure, and do not account for the specific use cases or detection of sensor anomalies after installation, leading to potential safety issues due to inaccurate sensor data.
Innovation Solution
The implementation of a method using rigid body kinematics to calibrate multiple inertial measurement units (IMUs) quickly and efficiently, allowing for real-time validation and adjustment of sensor data without the need for specialized equipment, by determining relative rotations and translations between IMUs and the vehicle, using a combination of first and second calibration models that reduce computational expense.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calibration techniques using infrastructure (fiducial markers) are used, then calibration accuracy can be achieved, but system downtime increases and safety risks arise due to the need to transport the system to specialized locations
Solution Approach 1:
The system performs self-calibration using its own sensors (IMUs, GPS, wheel encoders) and motion data, eliminating the need for external infrastructure like fiducial markers. The calibration is done in-situ at the operational location, allowing the system to calibrate itself without being transported to specialized facilities, thus reducing downtime while maintaining accuracy
Solution Approach 2:
The system collects sensor data and performs calibration computations in advance during normal operation or scheduled maintenance windows, so that calibration is completed before the system needs to be deployed. This preliminary calibration action ensures readiness without requiring downtime during critical operations
2Measurement precision
If traditional calibration techniques are used, then calibration can be performed, but the process is tedious and time-consuming because only a single sensor can be calibrated at a given time
Solution Approach 1:
The system merges the calibration processes of multiple IMUs into a single unified computation. By using the known rigid-body relationships between IMUs and incorporating data from multiple sensors simultaneously (GPS, wheel encoders, other IMUs), the system calibrates all sensors in parallel through one integrated optimization process, dramatically increasing calibration throughput
Solution Approach 2:
The calibration system is designed to handle multiple sensor types and configurations universally. The same computational framework calibrates individual IMUs, multiple IMUs, and integrates data from auxiliary sensors like GPS and wheel encoders, making the process versatile and efficient across different sensor arrangements without requiring separate calibration procedures for each sensor
3Loss of time
If existing calibration techniques that attempt to mitigate drawbacks are used, then calibration time may be reduced, but computational expense increases significantly
Solution Approach 1:
The system changes the calibration parameters by using relative transformations between sensors instead of absolute calibration for each sensor. By expressing IMU calibrations in terms of relative poses with respect to a reference frame and using incremental updates based on sensor comparisons, the computational complexity is reduced while maintaining calibration speed
Solution Approach 2:
The system performs partial calibration by focusing on determining relative transformations between sensors rather than full absolute calibration for each sensor. This partial action approach calibrates the sensor network sufficiently for operational purposes without the excessive computational burden of complete individual calibration of every sensor parameter
Data Source
AI summary
Systems and methods for calibrating multiple inertial measurement units on a system include calibrating a first of the inertial measurement units relative to the system using a first calibration model, and calibrating the remaining inertial measurement unit(s) relative to the first inertial measurement unit using a second calibration model. The calibration of the remaining inertial measurement unit(s) to the first inertial measurement unit can be based on a rigid body model by aligning a rotational velocity of the first inertial measurement unit with a rotational velocity of the remaining inertial measurement unit(s).