Performance Criteria for the Identification of Inertial Sensor Error Models

Area of Science:

• Engineering
• Measurement Science
• Signal Processing

Background:

• Accurate sensor error modeling is crucial for reliable system performance.
• Traditional models for inertial sensor errors require rigorous evaluation.
• Identifying and quantifying sensor errors impacts system precision.

Purpose of the Study:

• To define and calculate performance criteria for sensor error model identification.
• To assess the efficiency of identification solutions based on initial data.
• To compare the performance of various suboptimal identification algorithms.

Main Methods:

• Development of a calculation procedure based on a joint hypothesis recognition and parameter estimation algorithm.
• Application of the Bayesian approach for integrated problem-solving.
• Performance analysis of established inertial sensor error models.

Main Results:

• Established performance criteria for evaluating sensor error model identification.
• Demonstrated the influence of initial data on identification efficiency.
• Provided a comparative analysis of suboptimal algorithm performance.

Conclusions:

• The proposed performance criteria offer a robust method for assessing sensor error identification.
• The Bayesian-based algorithm effectively integrates hypothesis recognition and parameter estimation.
• The study validates the utility of the performance criteria through analysis of inertial sensor models.