Movement Data Error Correction for Nonlinear Object Motion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor technologies in autonomous driving systems, such as LiDAR, image, radar, and ultrasonic sensors, suffer from errors due to environmental factors like temperature, humidity, diffuse reflection, and diffraction, leading to inaccuracies in vehicle control and potential accidents.
Innovation Solution
An error correction method using an artificial intelligence-based model to identify and correct errors in movement data by grouping and normalizing sensing values, determining error values based on slope deviations, and correcting them using non-error values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional error correction methods (ATM or KF) are used, then some error correction is achieved, but they cannot handle non-linear errors or high-speed irregular movements completely
Solution Approach 1:
The patent changes the mathematical parameters of error correction by using a sliding window approach with dynamic threshold calculation. Instead of fixed linear models, the system calculates thresholds based on the standard deviation of measurements within a sliding window, allowing adaptation to non-linear patterns and high-speed movements. This parameter change enables the system to handle complex motion patterns that conventional linear methods cannot process effectively.
2Measurement precision
If temperature and humidity measurement methods are used to correct sensor errors, then tiny errors from environmental factors are detected, but errors from diffraction or diffuse reflection caused by object movements cannot be corrected
Solution Approach 1:
The patent applies self-service by using the sensor's own measurement data to detect and correct errors. The sliding window algorithm analyzes the temporal patterns of the sensor measurements themselves, calculating thresholds based on the standard deviation of recent measurements. This allows the system to automatically detect anomalies and correct errors without requiring external reference data or additional sensors, enabling it to handle object movement-induced errors like diffraction and diffuse reflection.
3Measurement precision
If KF method is used to achieve higher error correction rate, then linear function errors are corrected well, but non-linear characteristics and irregular high-speed movements cannot be completely corrected
Solution Approach 1:
The patent introduces dynamics by using a sliding window that moves through the measurement data, continuously adapting the correction threshold based on current measurement patterns. The window size and threshold calculation dynamically adjust to match the actual motion characteristics of the object. This dynamic approach allows the system to adapt to non-linear movements and high-speed irregular patterns, overcoming the limitations of static linear models like the KF method.
Data Source
AI summary
An error correction method and an error correction device for correcting an error in movement data of an object are provided. The error correction method includes collecting sensing values corresponding to movement of the object for each data frame, based on a predetermined time interval, grouping continuously collected data frames among data frames corresponding to the collected sensing values, normalizing the sensing values based on a minimum value among the sensing values of each of the grouped data frames, using an error detection model, determining whether a previous sensing value is an error value, based on a current sensing value and the previous sensing value collected at a time earlier than the current sensing value among the normalized sensing values, and when the previous sensing value is the error value, correcting the error value based on normal values that are not the error value among the sensing values.