Indoor Space Calibration Using Crowdsourced Sensor Trajectories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional indoor positioning systems face inefficiencies and labor-intensive calibration processes, limiting their deployment and scalability due to the need for manual data collection and reliance on prior knowledge of space geometry, which affects localization accuracy and speed.
Innovation Solution
A method for automated calibration of indoor spaces using crowdsourced sensor data from mobile devices, employing machine learning to identify anchor points and refine trajectories, creating structural and fingerprint maps without prior knowledge of the space's layout, utilizing inertial and signal sensors to track device movement and generate maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual calibration processes are used to collect sensor data for indoor positioning, then localization accuracy can be improved, but the calibration time and labor effort increase significantly
Solution Approach 1:
The system enables self-service calibration by automatically utilizing sensor data from mobile devices that naturally move through the indoor space during normal operations. The calibration process serves itself by harvesting data from routine device usage without requiring dedicated manual calibration efforts, thereby maintaining localization accuracy while eliminating time loss.
Solution Approach 2:
The patent replaces the mechanical manual calibration process with an automated computational system. Instead of physically moving devices to predetermined calibration points, the system uses machine learning algorithms to process sensor data from naturally occurring device movements, substituting mechanical calibration actions with automated data processing and computational analysis.
2Speed
If prior knowledge of space geometry is required for calibration, then localization speed can be improved, but the system complexity and deployment difficulty increase
Solution Approach 1:
The system performs self-mapping by automatically constructing spatial representations of the indoor environment through analysis of sensor data from mobile devices. The calibration process serves itself by deriving space geometry information from natural device trajectories and sensor measurements, eliminating the need for external prior knowledge while maintaining localization speed.
Solution Approach 2:
The system performs preliminary mapping actions automatically during the calibration phase, creating spatial representations and anchor point configurations before actual localization operations begin. This preliminary automated mapping enables fast localization speed without requiring manual preparation of space geometry information.
3Measurement precision
If extensive manual data collection is performed during calibration, then the quality of localization data can be improved, but the productivity and scalability of the system deteriorate
Solution Approach 1:
The system harvests calibration data from natural mobile device usage patterns during normal operations. Instead of requiring dedicated data collection efforts, the calibration process serves itself by accumulating sensor data from routine device movements, maintaining high data quality while maximizing calibration efficiency and system productivity.
Solution Approach 2:
The system continuously collects sensor data during normal device operations rather than performing discrete manual data collection sessions. This continuous data accumulation from ongoing device usage maintains high data quality for localization while maximizing calibration efficiency, as the useful action of data collection occurs continuously without interrupting normal system productivity.
4Measurement precision
If traditional calibration methods are used that require coordinated data collection, then measurement accuracy can be maintained, but the ease of operation and deployment simplicity are reduced
Solution Approach 1:
The system performs self-calibration by automatically processing sensor data from mobile devices without requiring coordinated human operations. The calibration process serves itself by autonomously identifying anchor points, constructing spatial maps, and optimizing localization parameters, thereby maintaining measurement accuracy while dramatically simplifying deployment procedures.
Solution Approach 2:
Instead of requiring devices to be positioned at predetermined locations for calibration, the system inverts the approach by allowing devices to move freely and deriving spatial information from their natural trajectories. This inverted calibration methodology maintains localization accuracy while eliminating the operational complexity of coordinated data collection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and a device for performing automated calibration are described. In an example, sensorial data gathered for a space to be mapped in an indoor area is obtained from one or more data recording devices. From the sensorial data, a plurality of trajectories are derived which indicate movement of the one or more data recording devices in the space. Based on the plurality of trajectories, a structural map representing a plurality of map constraints of the space is created. Further, from the sensorial data, calibration information associated with a series of locations in the space is extracted and used along with the structural map to generate a fingerprint map of the space.