Multirotor Velocity Estimation With High-Pass Drift Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional localization techniques for multirotor systems, such as GPS, LIDAR, and IMU, face challenges in indoor environments and suffer from drift errors due to inexact modeling of drag forces, which impact the accuracy of velocity estimation and localization.
Innovation Solution
A processor-implemented method and system that utilize gyroscope data to compute acceleration based on multirotor dynamics, identify and filter out low-frequency drift caused by inexact drag force modeling using a high band pass filter, thereby estimating drift-free velocity and localizing the multirotor system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If IMU is used for localization, then high localization frequency is achieved, but localization accuracy drifts with time
Solution Approach 1:
The patent introduces an intermediary processing step between IMU acceleration data and velocity estimation. A band-pass filter is applied to remove low-frequency drift components caused by drag force modeling errors, while preserving the high-frequency useful velocity information. This intermediary filtering operation allows the system to maintain high localization frequency from IMU while eliminating the time-drifting accuracy degradation.
Solution Approach 2:
The patent changes the frequency domain parameters of the velocity signal by applying a band-pass filter with specific cutoff frequencies. The filter removes low-frequency components (drift) while preserving high-frequency components (useful velocity information). This parameter transformation in the frequency domain resolves the contradiction between maintaining high frequency response and eliminating drift-induced accuracy loss.
2Measurement precision
If vision-based localization algorithm is used, then metric-scale localization is provided, but localization frequency is limited by camera frame rate
Solution Approach 1:
The patent replaces the mechanical/optical measurement system (camera-based vision localization) with an inertial measurement system (IMU-based velocity estimation). The IMU provides high-frequency acceleration data at 200 Hz, substituting the camera's frame-rate-limited optical flow method. This substitution achieves both high frequency and acceptable accuracy by using inertial sensors instead of vision-based methods.
3Device complexity
If drag force is not modeled, then device complexity is reduced, but velocity estimation accuracy drifts due to inexact or non-modelling of drag force
Solution Approach 1:
The patent extracts and removes the harmful low-frequency drift component from the velocity estimation without requiring complex drag force modeling. By identifying the drift as a low-frequency component and applying a band-pass filter to eliminate it, the system achieves accurate velocity estimation without the complexity of precise drag force models. The harmful effect is separated and removed rather than compensated through complex modeling.
Data Source
AI summary
Embodiments of the present disclosure provide systems and methods to eliminate (or filter) drift for dynamics model based localization of multirotors. The dynamics equations require drag modelling, which is dependent on velocity, to generate vehicles' acceleration along the body axis. The present disclosure considers the drag contribution, at velocity level, as a low frequency component. Incorrect or nonmodelling of this low frequency component leads to drift at velocity level. This drift can then be removed through a high pass filter to obtain drift free velocity data for pose estimation and better localization thereof.