Mobile Body Speed Estimation with Vibration Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to accurately estimate the moving speed of mobile bodies with periodic vibration components, leading to incorrect predictions of future positions due to vibration components not being adequately removed.
Innovation Solution
A moving speed estimation device that includes a position measuring unit, an instantaneous moving speed estimation unit, a vibration component information estimation unit, and a speed estimated value adjustment processing unit, which removes periodic vibration components from the estimated moving speed using band-stop filters and Kalman filters to suppress fluctuations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the time width for averaging instantaneous moving speed is short, then the responsiveness of moving speed estimation is improved, but the vibration components of instantaneous moving speed cannot be sufficiently removed
Solution Approach 1:
The patent segments the moving speed estimation process into multiple independent components: instantaneous moving speed estimation unit that calculates raw speed values, vibration component information estimation unit that identifies periodic vibration characteristics, and speed estimated value adjustment processing unit that removes vibration components. This segmentation allows each component to be optimized independently, resolving the contradiction between responsiveness and accuracy.
Solution Approach 2:
The patent extracts the vibration component from the instantaneous moving speed through spectral analysis (FFT). The vibration component information estimation unit identifies the frequency and amplitude of periodic vibrations, and the adjustment processing unit removes this extracted vibration component from the instantaneous speed, leaving only the true moving speed without vibration contamination.
2Measurement precision
If the time width for averaging instantaneous moving speed is long, then the vibration components are sufficiently removed, but the moving speed is estimated too late
Solution Approach 1:
Instead of using expensive, computationally intensive long-term averaging that delays estimation, the patent employs a more efficient approach: it uses spectral analysis (FFT) to identify vibration components and subtracts them directly from instantaneous speed measurements. This 'disposable' method of removing only the necessary vibration components achieves accuracy without the time penalty of extensive averaging.
3Speed
If instantaneous moving speed is used for future position prediction, then the prediction responsiveness is improved, but wrong prediction is easily made due to vibration components
Solution Approach 1:
The patent performs preliminary action by removing vibration components from the instantaneous moving speed before using it for future position prediction. The vibration component information estimation unit and speed estimated value adjustment processing unit preprocess the speed data to eliminate periodic vibrations, ensuring that the speed value used for prediction is clean and accurate, thus preventing wrong predictions while maintaining responsiveness.
Data Source
AI summary
A moving speed estimation device which includes a vibration component information estimation unit 4 which estimates the vibration cycle or vibration frequency of the periodic vibration component included in instantaneous moving speed of a mobile body A along with a periodic moving action of the mobile body A. The moving speed estimation device also includes a speed estimated value adjustment processing unit 5 which sequentially generates speed values as estimated values of the moving speed of the mobile body A, where the speed value is obtained by performing at least processing of removing a vibration component corresponding to the estimated value of the vibration cycle or vibration frequency for the time series of the instantaneous moving speed estimated from the time series of the measured value of the position of the mobile body A.