Wireless Tracking Map Reconstruction via Trajectory Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for indoor map reconstruction are either costly due to the use of hardware like lidar or inaccurate due to reliance on inertial sensing, and lack digital indoor maps, making it difficult to achieve high-accuracy, low-cost trajectory tracking in various environments.
Innovation Solution
A system that collects sensing data to generate maps by segmenting and bundling trajectories based on similarity measures, fusing them to compute shapes, and generating maps from these shapes, utilizing wireless tracking technology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If costly hardware like lidar is used for indoor map reconstruction, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical/optical sensing systems (lidar, cameras) with wireless signal-based tracking systems. The system uses wireless signals to collect spatial coordinate data and sensing data from multiple devices, processes this data through algorithms to reconstruct indoor maps, thereby eliminating the need for costly hardware while maintaining or improving measurement precision.
Solution Approach 2:
The patent creates a virtual copy of the physical measurement process. Instead of using physical sensors to directly measure the environment, the system collects wireless signal data (which can be received by multiple devices simultaneously) and generates a digital representation of the indoor map through computational algorithms, achieving accurate mapping without physical measurement hardware.
2Device complexity
If inertial sensing is used for trajectory tracking, then device complexity is reduced, but measurement precision deteriorates due to accumulative errors
Solution Approach 1:
The patent introduces wireless signals as an intermediary between the tracking system and the environment. Instead of relying on inertial sensors that directly measure motion (prone to accumulative errors), the system uses wireless signals to mediate the tracking process, collecting spatial and sensing data that can be processed to achieve higher precision without significantly increasing device complexity.
Solution Approach 2:
The system implements feedback mechanisms by continuously collecting wireless signal data and comparing it with expected patterns. The processing system can detect and correct deviations from expected trajectories using feedback from multiple data sources, thereby maintaining high precision while keeping the system relatively simple.
3Device complexity
If crowdsourcing methods are used for map reconstruction, then device complexity is reduced, but measurement precision deteriorates due to inaccurate data from multiple users
Solution Approach 1:
The patent segments the data collection and processing into distinct phases. It collects raw wireless signal data from multiple devices, then processes this data through specialized algorithms to segment and refine individual trajectories. This segmentation allows the system to maintain simplicity in data collection while achieving high precision through sophisticated processing of segmented data segments.
Solution Approach 2:
The system changes the parameters used for data collection and processing. Instead of relying on traditional inertial sensor parameters that accumulate error, the system uses wireless signal parameters (signal strength, phase, frequency) that can be processed to eliminate accumulative errors. This parameter transformation enables high precision map reconstruction from multi-user data without sacrificing system simplicity.
Data Source
AI summary
Methods, apparatus and systems for map reconstruction based on wireless tracking are described. In one example, a described system comprises: a sensor configured to collect sensing data in a venue and obtain a plurality of trajectories, and a processor. Each trajectory is a time series of spatial coordinates (TSSC) representing a path traversed by a respective object in the venue. Each TSSC is accompanied by at least one respective time series of sensing data (TSSD) collected while the respective object traverses the path in the venue. The processor is configured for: segmenting each TSSC and its accompanying at least one TSSD into segments, bundling the plurality of trajectories based on similarity measures between pairs of the segments, fusing the bundled trajectories to generate fused trajectories, computing a shape of the fused trajectories, and generating a map of the venue based on the computed shape.