Train Satellite Selection Using Dynamic Model Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the instantaneous kinematic state of guided vehicles, such as trains, face challenges in accurately selecting usable satellite signals due to signal disturbances like the 'Alternative Path Phenomenon, which can lead to erroneous speed calculations.
Innovation Solution
A method that calculates and compares instantaneous Doppler coefficients and pseudoranges from visible satellites, using a dynamic model of the guided vehicle based on past kinematic states and satellite ephemeris to filter out disturbed signals by verifying differences and covariances against thresholds, ensuring accurate selection of usable satellites for determining the kinematic state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a preliminary selection method using multiple remote receivers is used to detect disturbed signals, then the reliability of satellite signal selection is improved, but the device complexity and cost increase
Solution Approach 1:
The system uses its own on-board kinematic sensors (odometer, inertial unit, GPS) to generate expected signal values and compare them with received satellite signals, eliminating the need for external reference receivers. The vehicle itself serves as the reference frame for detecting signal disturbances.
Solution Approach 2:
A dynamic model of the guided vehicle acts as an intermediary, translating physical kinematic state (position, speed, acceleration) into expected satellite signal characteristics. This model mediates between the vehicle's motion and the satellite signals, enabling disturbance detection without additional receivers.
2Reliability
If random selection of four satellites is used for speed calculation, then the measurement confidence is improved, but the time required for reliable kinematic state determination increases
Solution Approach 1:
The system performs preliminary validation of each satellite signal against the dynamic model before inclusion in the calculation algorithm. This pre-screening ensures that only reliable satellites are selected, making the subsequent random selection of four satellites faster and more reliable.
Solution Approach 2:
The system continuously compares expected signal values (from the dynamic model) with actual received satellite signals, creating a feedback loop that rapidly identifies and excludes disturbed signals. This feedback mechanism accelerates the satellite selection process while maintaining high measurement confidence.
3Measurement precision
If signal disturbance detection is implemented, then the accuracy of instantaneous speed measurement is improved, but the computational complexity increases
Solution Approach 1:
The system changes the parameter being monitored from raw satellite signal strength to dynamic consistency between expected and received signal values. By comparing kinematic parameters (position, speed, acceleration) derived from different sources, the system detects disturbances without complex signal processing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively filters out disturbed signals, providing accurate and reliable instantaneous kinematic state measurements by quickly identifying and excluding affected satellites, thereby improving the precision of speed and position calculations.
Implementation Method 1
calculating an instantaneous measured value of one of a Doppler coefficient and a pseudorange, from the signal received from said visible satellite
Implementation Method 2
calculating an instantaneous measured value of one of a Doppler coefficient and a pseudorange, from the signal received from said visible satellite
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method for selecting usable satellites (LSVU) from among the satellites (LSV) of a localization constellation, to determine an instantaneous kinematic state of a train, comprising the steps of determining a measured value (Di, PDi) and an estimated value (D*i, PD*i) of a Doppler coefficient and/or a pseudodistance, then comparing the measured and estimated values and, in case of discrepancy, removing the satellite from the list of usable satellites. The estimated value results from a dynamic model (M) of the train, which uses only a kinematic state at a past time (E(t-1)) to calculate an estimated instantaneous kinematic state (E*(t)) and which uses a map of the track (4) on which the guided vehicle is moving.