Underground Transit Navigation Using Sequential Semantic Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional GPS and GNSS technologies face challenges in providing accurate location and navigation data in areas with signal interference or blockage, such as underground transportation systems, leading to drift errors and limited accuracy in dead reckoning navigation.
Innovation Solution
A method that processes sensor data from mobile devices, including IMU and non-GNSS sensors, to determine time-sequenced semantic events like turn angles, which are compared to known events to infer direction and location within a transportation system, even without satellite-based data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If satellite-based positioning (GPS/GNSS) is used for navigation, then location accuracy is improved, but it becomes unavailable in underground or indoor environments due to signal blockage
Solution Approach 1:
The patent introduces intermediary sensors (accelerometer, gyroscope, magnetometer, barometer) that mediate between the mobile device and the environment to determine location and direction when satellite signals are unavailable. These sensors serve as intermediaries that capture motion and environmental data to infer position without direct satellite contact.
Solution Approach 2:
The system performs preliminary actions by collecting and processing sensor data continuously to establish a baseline of motion patterns and semantic events before navigation is needed. This preliminary data collection enables the system to function independently of satellite signals by having already processed motion characteristics and environmental context.
2Reliability
If dead reckoning navigation using device sensors is used, then navigation becomes available without satellite signals, but drift errors accumulate reducing accuracy
Solution Approach 1:
The patent implements feedback mechanisms by continuously comparing observed semantic events against expected patterns and using this feedback to correct drift errors. The system monitors sensor data over time, identifies deviations from expected motion patterns, and adjusts calculations to maintain accuracy despite the absence of satellite correction signals.
Solution Approach 2:
The system changes parameters by transitioning from raw sensor data to processed semantic events, and further to contextualized navigation information. By transforming acceleration, angular velocity, and pressure data into meaningful events (turns, stops, elevations), the system maintains precision through parameter transformation rather than direct integration of raw sensor readings.
3Ease of operation
If traditional sensor-based navigation is used in multi-one-dimensional transportation systems, then basic direction can be determined, but accurate location and direction determination fails due to signal interference and drift
Solution Approach 1:
The patent adds another dimension to navigation by incorporating semantic event sequencing and contextual analysis beyond traditional 2D spatial coordinates. The system processes motion data, environmental data, and temporal patterns as additional dimensions, enabling accurate location determination in complex multi-one-dimensional transportation systems where traditional GPS fails.
Solution Approach 2:
The system segments the continuous sensor data stream into discrete semantic events (turns, stops, elevations, transitions). By dividing the navigation problem into identifiable event segments that can be matched against known transportation system patterns, the system achieves accurate location determination through event sequence recognition rather than continuous coordinate tracking.
Data Source
AI summary
An approach is provided for determining travel direction and/or location data based on sequential semantic events. The approach, for example, involves processing sensor data collected from at least one sensor of a mobile device to determine a set of observed time-sequenced semantic events. The semantic events are associated with traveling within a transportation system, and at least one semantic event of the set of observed time-sequenced semantic events is based on a turn angle value determined from the sensor data. The approach also involves initiating a comparison of the set of observed time-sequenced semantic events against a set of known time-sequenced semantic events and/or a known piece of information associated with the transportation system. The approach further involves determining a direction of travel and/or a location of the mobile device within the transportation system based on the comparison. The approach further involves providing the direction of travel and/or the location as an output.