Gravity Estimation for Motion Recognition Using Accelerometer and Gyroscope Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for determining motion in mobile devices, such as GPS-based and accelerometer-based systems, face challenges with high power consumption, indoor inaccuracy, and poor accuracy in distinguishing transportation modes due to the influence of Earth's gravity, which affects the accuracy of motion recognition.
Innovation Solution
A method and apparatus that estimate Earth's gravity by determining the mean of acceleration data, interpolating between stable periods, and adjusting a threshold value based on stability values to account for changes in orientation, using a combination of acceleration and gyroscope sensors to mitigate the impact of centripetal forces and improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS-based systems are used to determine motion information, then positioning information can be obtained, but power consumption increases and accuracy is limited
Solution Approach 1:
The patent extracts and removes the gravity component from accelerometer measurements by estimating it separately using gyroscope data and orientation information. This allows the system to use only accelerometer data for motion detection without relying on GPS, thereby reducing power consumption while maintaining accuracy in distinguishing transportation modes
Solution Approach 2:
The patent introduces gyroscope sensors as an intermediary device to measure orientation changes. This intermediary measurement enables accurate estimation and removal of the gravity component from accelerometer data, allowing motion information to be derived without GPS while improving accuracy in detecting transportation modes
2Use of energy by moving object
If accelerometer sensors are used to detect motion, then power consumption is reduced, but measurement accuracy deteriorates due to earth's gravity influence
Solution Approach 1:
The patent extracts the gravity component from the total acceleration measured by the accelerometer. By estimating gravity using gyroscope orientation data and removing it from the accelerometer measurements, the system obtains accurate horizontal acceleration information for motion detection without GPS, thereby improving accuracy while maintaining low power consumption
Solution Approach 2:
The patent uses a feedback mechanism where the estimated gravity component is continuously updated and subtracted from accelerometer measurements. The system monitors the stability of gravity estimates and adjusts the estimation process accordingly, ensuring accurate motion detection while maintaining energy efficiency
3Device complexity
If simple mean of accelerometer measurement is used, then computation is simplified, but accuracy deteriorates with continuous acceleration
Solution Approach 1:
The patent implements a dynamic gravity estimation process that adapts to changing motion conditions. The system monitors the stability of gravity estimates and switches between different estimation strategies based on detected motion patterns, ensuring accurate gravity removal during both stationary and continuously accelerating phases while maintaining computational efficiency
4Device complexity
If L2 norm of accelerometer measurement is used, then computation is simplified, but horizontal acceleration information is masked by gravity
Solution Approach 1:
The patent extracts and removes the gravity component from accelerometer measurements before computing motion parameters. By subtracting the estimated gravity vector from the total acceleration, the system preserves horizontal acceleration information that would otherwise be masked, enabling accurate detection of transportation modes while maintaining computational simplicity
Data Source
Figure 1~3
Figure 2
AI summary
The invention relates to a method for generating an estimation of earth's gravity. The method comprises: obtaining (210) one or more acceleration data values and one or more orientation data values over a period of time; generating (220) magnitude of orientation change from the orientation data values; determining (230) a stability value based on the acceleration data values and the magnitude of orientation change; comparing (240) the determined stability value to a threshold value; and generating (250) an estimation of earth's gravity over the period of time on the basis of the acceleration data values if the comparison indicates that the determined stability value is below the threshold value. The invention also relates to an apparatus implementing the method.