Indoor Positioning via Sensor Fusion and Trajectory Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Global Positioning System (GPS) technology is ineffective in indoor environments due to signal blocking by building structures, leading to inaccuracies in navigation and location determination within indoor spaces.
Innovation Solution
A method utilizing a combination of GPS data, radio frequency (RF) data, sensor data, and sequential trajectory data to determine the location of areas of interest, such as building entrances and indoor rooms, through clustering and filtering techniques, with crowdsourced data analysis for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS system is used for location determination, then outdoor positioning accuracy is improved, but indoor positioning effectiveness deteriorates due to signal blocking by building structures
Solution Approach 1:
The system segments the positioning problem into two distinct environments: outdoor GPS-based positioning and indoor sensor-based positioning. The mobile device automatically switches between these segmentation-based approaches depending on the detected environment, using GPS data when outdoors and sensor data (accelerometer, magnetometer, barometer) when indoors, thereby resolving the contradiction between outdoor accuracy and indoor effectiveness
Solution Approach 2:
The system introduces an intermediary environment detection mechanism that mediates between GPS and sensor data based on signal availability. When GPS signals are blocked by building structures, the intermediary detection system activates alternative sensor fusion algorithms, allowing seamless transition between positioning modes and maintaining reliability across both indoor and outdoor environments
2Reliability
If multiple data sources (GPS, RF, sensor data) are combined for location determination, then positioning reliability in mixed environments is improved, but system complexity increases
Solution Approach 1:
The system implements a universal positioning framework that can handle multiple data sources (GPS, RF, accelerometer, magnetometer, barometer) through a single unified algorithmic approach. The sensor fusion module universally processes all available data types using the same mathematical framework, eliminating the need for separate processing pipelines for each sensor type and thereby managing complexity while maintaining multi-environment reliability
Solution Approach 2:
The system dynamically changes processing parameters based on data quality and availability. When GPS signals are weak or unavailable, the system automatically adjusts sensor fusion weights and thresholds. This parameter adaptation allows the system to maintain reliability across varying environmental conditions without requiring complex manual configuration or multiple fixed algorithms
Data Source
AI summary
In one aspect, GPS data, ambient signal data, radio frequency data and/or other types of data are used to determine the location of an entrance to an area or a building. In another aspect, sequential trajectory data is collected and the data is analyzed and/or processed to determine the location of an area of interest.