Pedestrian Dead Reckoning with Body-Fixed Motion Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In challenging signal environments, such as urban canyons and dense foliage, radio navigation-based position solutions are inaccurate due to signal attenuation, and existing dead reckoning (DR) methods accumulate errors over time, especially when pedestrian motion involves changes in direction, as they use inadequate motion models that do not distinguish between user and device motion.
Innovation Solution
A system and method that determine rotational velocity and user acceleration based on sensor signals, using a body-fixed assumption to update a user state estimator, incorporating periodic motion constraints to improve the accuracy of device state estimation by differentiating between user and device motion, thereby reducing errors in pedestrian motion prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a typical motion model assigns constant velocity to the user regardless of motion activity class, then the device complexity is reduced, but the measurement precision of user motion deteriorates
Solution Approach 1:
The motion model transitions from a static constant velocity assumption to a dynamic model that adapts based on detected motion activity classes. The system continuously monitors accelerometer and gyroscope data to identify changes in motion patterns (walking, running, driving, stationary) and adjusts the velocity prediction accordingly, allowing the model to respond to changing user behavior while maintaining manageable complexity through algorithmic adaptation.
Solution Approach 2:
The system changes the parameters of the motion model based on detected motion activity. When a specific motion class is identified (e.g., walking vs. driving), the model adjusts its parameters such as expected velocity ranges, acceleration patterns, and directional constraints to match the characteristics of that activity class, thereby improving measurement precision without requiring a completely different model for each scenario.
2Device complexity
If the DR component uses a motion model that does not account for relative motion between user and device, then the device complexity is reduced, but the reliability of position solution deteriorates
Solution Approach 1:
The motion estimation is segmented into two distinct components: user motion (body movement) and device motion (device orientation changes relative to the user). The system separately estimates user translation and rotation from inertial sensors, then independently tracks device orientation changes using sensor fusion. This segmentation allows the DR component to accurately subtract only the device's rotational contribution from the total measured motion, isolating the user's true positional changes and improving position solution reliability.
Solution Approach 2:
The system introduces an intermediary computational layer that processes sensor data to separate user motion from device motion. By using a motion model that explicitly accounts for relative device orientation and applying sensor fusion algorithms, the system creates an intermediate representation of pure user motion that can be used for accurate dead reckoning, without requiring direct measurement of user position.
3Measurement precision
If the DR component accumulates motion estimates over multiple epochs, then the position solution quality is improved through statistical combination, but the loss of time increases due to error accumulation
Solution Approach 1:
The system implements feedback mechanisms where the DR component continuously monitors the quality and consistency of its position estimates. When error accumulation reaches certain thresholds or when external signals (such as WiFi sightings or barometric pressure changes) become available, the system uses these feedback signals to detect and correct drift. This allows the system to maintain high position quality through multi-epoch integration while periodically resetting accumulated errors, effectively managing the trade-off between measurement precision and time loss.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the accuracy of device state estimation by using a more precise motion model that accounts for user and device motion, leading to improved position and velocity estimates with reduced error accumulation, especially in scenarios like walking or device orientation changes.
Implementation Method 1
determining, by the device, user acceleration based on the user speed and the rotational velocity of the user
Implementation Method 2
determining, by a device, a rotational velocity of a user of the device based on a sensor signal
Data Source
AI summary
Systems, methods, devices and computer-readable storage mediums are disclosed for device state estimation with body-fixed assumption. In some implementations, a method comprises: determining, by a device, a rotational velocity of a user of the device based on a sensor signal; determining, by the device, user speed; determining, by the device, user acceleration based on the user speed and the rotational velocity of the user; and updating a user state estimator based on the user acceleration.