Indoor Positioning With Beacon-Based Bayesian Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing positioning technologies struggle to accurately determine the location of vehicles in indoor environments due to interference from structures, leading to low accuracy and difficulty in integrating with other sensor information, which is crucial for safe autonomous driving.
Innovation Solution
An indoor positioning method using a mobile terminal and multiple positioning sensors that apply a probability inference method, specifically Bayesian inference, to calculate the location based on signal strengths and unique identifications, forming a probability distribution for precise positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS signal is used for positioning in outdoor environment, then positioning accuracy is maintained, but positioning fails or error increases in indoor spaces due to structural interference
Solution Approach 1:
The patent introduces positioning sensors (beacons) as intermediary devices installed throughout the indoor space. These beacons act as mediators between the mobile terminal and the positioning system, transmitting signals that enable accurate indoor positioning without relying on GPS. The beacons create a localized positioning infrastructure that overcomes the blockage caused by building structures.
Solution Approach 2:
The patent replaces the GPS satellite-based electromagnetic positioning system with a local indoor positioning system using beacons and mobile terminals. This substitution involves using a different physical approach - measuring signal strengths from multiple nearby beacons and applying geometric algorithms (Apollonius spheres, trilateration) to determine position, rather than relying on satellite signals that cannot penetrate buildings.
2Measurement precision
If traditional indoor positioning methods are used, then positioning can be achieved, but accuracy is low and integration with other sensor information is difficult
Solution Approach 1:
The patent changes the parameter representation from deterministic point coordinates to probability distributions. Instead of outputting a single position point, the system outputs a probability distribution that represents the uncertainty and reliability of the position estimate. This parameter change enables easier integration with other sensor information through probabilistic fusion methods, as the distribution can be combined with other sensor data using standard probabilistic techniques.
Data Source
AI summary
Embodiments of the present disclosure provide an indoor positioning method for measuring a location of a mobile terminal using the mobile terminal and a plurality of positioning sensors communicating therewith. The indoor positioning method includes selecting a period for acquiring sensor information related to a relationship between the mobile terminal and the positioning sensors; acquiring a variable group corresponding to a positioning sensor selected from among the plurality of positioning sensors for each of unit periods of the selected period, the sensor information being acquired from the plurality of positioning sensors; applying a probability inference method to the variable group; and acquiring a final location of the mobile terminal based on the probability inference method, wherein the sensor information comprises signal strengths received by the mobile terminal from the positioning sensors, and a unique identification of the positioning sensors.