Body-Worn Pedestrian Localization Using ANN Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing localization systems, such as GPS, visual odometry, and beacon-based systems, fail to provide accurate and efficient pedestrian localization in GPS-denied environments like buildings and forests, often requiring expensive equipment, infrastructure, or time-consuming installations.
Innovation Solution
A body-worn localization device using a neural processing unit (NPU) with a pre-trained artificial neural network (ANN) and sensors like IMU and pressure/temperature sensors to determine real-world location, stance, and activity, without relying on pre-existing infrastructure, and a network of nodes for real-time map updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS systems are used for localization, then position determination is achieved, but precision is lost or system fails in GPS-denied environments like buildings and forests
Solution Approach 1:
The patent introduces an intermediary localization system that operates between GPS and the physical environment. The system uses a network of nodes deployed in the environment that mediate localization by providing reference points for trilateration, enabling position determination in GPS-denied areas while maintaining precision through geometric calculation methods.
Solution Approach 2:
The patent replaces the satellite-based electromagnetic signal system (GPS) with a ground-based geometric positioning system. Instead of relying on satellite signals that penetrate the atmosphere, the system uses line-of-sight radio frequency signals between nodes and user devices, substituting the mechanical/satellite infrastructure with a distributed ground network.
2Measurement precision
If visual odometry is used for localization, then position and orientation can be determined, but expensive image capturing equipment is required
Solution Approach 1:
The patent extracts the essential localization function from complex visual processing systems. Instead of using full image capture and processing pipelines, the system extracts positioning capability through simple signal transmission and geometric calculation, removing the need for expensive cameras and image processing hardware while retaining localization accuracy.
Solution Approach 2:
The patent creates a virtual copy of the physical environment through a digital map constructed from node positions. This digital representation allows the system to determine location through geometric relationships rather than visual analysis, copying spatial information in a computationally efficient format that eliminates the need for complex imaging equipment.
3Reliability
If traditional beacon-based localization systems are used, then localization can be achieved, but installation is time-consuming and requires existing infrastructure
Solution Approach 1:
The patent performs preliminary actions by pre-deploying a network of localized nodes throughout the environment before user arrival. These nodes are positioned in advance and form a ready-made localization infrastructure, eliminating the need for time-consuming installation during actual use and enabling immediate localization capability.
Solution Approach 2:
The system enables self-service localization where user devices autonomously determine their position by communicating with the pre-deployed node network. The infrastructure serves itself by using the geometric relationships between nodes to automatically establish coordinate systems and provide positioning services without requiring manual configuration or existing building infrastructure.
4Reliability
If beacon-based systems are used, then localization is possible, but power and radio infrastructure must be present at installation location
Solution Approach 1:
The patent creates a universal localization system where nodes can be deployed in diverse environments without requiring specific infrastructure. The nodes perform multiple functions including position reference, signal transmission, and environmental mapping, making the system adaptable to buildings, forests, and other GPS-denied areas regardless of existing power or radio infrastructure.
Solution Approach 2:
The system dynamically adapts to different environments by allowing flexible node deployment configurations. Nodes can be positioned according to environmental features rather than fixed infrastructure points, and the localization algorithm dynamically adjusts to the geometric arrangement of nodes, enabling versatility across diverse settings from indoor spaces to outdoor forests.
Data Source
AI summary
A localisation device having a first sensor that is configured to provide first measurement data and a neural processing unit (NPU) that includes a pre-trained artificial neural network (ANN) and a processor that is in communication with the first sensor and the NPU. The processor is configured to collect the first measurement data from the first sensor over a time period and determine a real-world location using the ANN.