Robust Dead Reckoning: Calibration, Covariance Estimation, Fusion and Integrity Monitoring

Abstract: To measure system states and local environment directly with high precision, expensive sensors are required. However, highly accurate system states and environmental perception can also be achieved using data fusion techniques and digital maps. One crucial task of multi-sensor state estimation is to project different sensor measurements into the same temporal, spatial and physical domain, estimate their covariance matrices as well as the exclusion of erroneous measurements. This paper presents a generic approach for robust estimation of vehicle movement (odometry). We will shortly present our calibration procedure, including the estimation of sensor alignments, offset / scaling errors, covariances / correlations and time delays. An improved algorithm for wheel diameter estimation is presented. Additionally an approach for robust odometry will be shown as odometry estimations are fused under known covariances, while outliers are detected using a chi-squared test. Utilizing our robust odometry, local environmental views can be associated and fused. Furthermore our robust odometry can be used to detect and exclude erroneous position estimates.

• The paper introduces a detailed calibration procedure that corrects sensor alignments and offsets for precise measurements.
• It develops covariance estimation and fusion methods using chi-squared tests to reliably detect and exclude outlier data.
• The paper proposes a wheel diameter estimation algorithm that significantly enhances odometry accuracy and overall vehicle state estimation.

The paper "Robust Dead Reckoning: Calibration, Covariance Estimation, Fusion and Integrity Monitoring" focuses on enhancing the precision of vehicle movement estimations through a series of sophisticated techniques. The authors aim to achieve robust dead reckoning without the necessity for expensive sensors, by leveraging data fusion methodologies and digital maps.

The paper begins by discussing the calibration procedure essential for accurate sensor data integration. This includes estimating sensor alignments, identifying and correcting offset and scaling errors, and calculating covariances and correlations. Additionally, the paper addresses the issue of time delays between sensor measurements.

Key contributions of the paper include:

1. Calibration Procedure: A detailed approach to calibrating sensors is provided. This encompasses techniques for determining sensor alignments and pinpointing offset and scaling errors, thus ensuring that sensor readings are precise and reliable.
2. Covariance Estimation: The authors elaborate on methods to estimate covariance matrices for various sensor measurements. This step is crucial for accurately fusing data from multiple sensors.
3. Wheel Diameter Estimation Algorithm: The paper introduces an improved algorithm specifically designed for accurate wheel diameter estimation. This is a vital parameter for odometry calculations and impacts the overall accuracy of vehicle movement estimation.
4. Robust Odometry: Innovative methods for robust odometry are presented, where odometry estimations are fused taking into account known covariances. A notable feature is the use of a chi-squared test to detect outliers in the data, ensuring that erroneous measurements do not affect the overall estimation.
5. Error Detection and Exclusion: Utilizing the robust odometry framework, the authors propose techniques to detect and exclude erroneous position estimates. This can significantly enhance the reliability of vehicle navigation systems.
6. Environmental View Association and Fusion: The robust odometry method allows for the association and fusion of local environmental views. This capability is critical for creating a coherent perception of the vehicle's surroundings, which can be particularly beneficial in complex navigation tasks.

Overall, the paper contributes valuable methodologies for achieving high-precision vehicle state estimation and environmental perception without relying on prohibitively expensive sensors. The techniques discussed have significant implications for advancing autonomous vehicle technologies, particularly in terms of cost-effectiveness and reliability.