An Acquisition Method of Agricultural Equipment Roll Angle Based on Multi-Source Information Fusion

Yang Li, Honglei Jia, Jiangtao Qi, Huibin Sun, Xinliang Tian, Huili Liu, Xuhui FanSensors 2020 · DOI

TL;DR

This paper presents a novel method for accurately measuring the roll angle of agricultural machinery by fusing data from an Inertial Navigation System (INS) and a BeiDou Navigation Satellite System (BDS). This integrated approach significantly improves roll angle estimation accuracy compared to using either sensor alone, paving the way for more precise autonomous agricultural operations.

What Problem Does This Paper Address?

Modern agriculture is increasingly relying on intelligent machinery for tasks like plowing and sowing. The accurate measurement of an agricultural machine's roll angle (its tilt side-to-side) is crucial for several reasons: it directly impacts the quality of operations like tillage depth and soil compaction, it's essential for autonomous navigation systems to maintain stability and follow planned paths, and it can provide early warnings for potential rollovers, preventing damage and ensuring safety. While inertial sensors (like accelerometers and gyroscopes) can measure roll angle, their readings tend to drift over time. Satellite navigation systems (like BDS) offer stable, drift-free position and attitude information but can be susceptible to signal blockage. This paper addresses the challenge of achieving high-accuracy, reliable roll angle measurements in dynamic agricultural environments by combining the strengths of both sensor types.

Methods

The core of this paper is a multi-source information fusion approach, specifically employing a Kalman filter to integrate data from an MTi-300 AHRS (INS) and an XW-GI 5630 (BDS).

Here's a breakdown of their methodology:

  • Sensor Setup: An MTi-300 AHRS (containing accelerometers, gyroscopes, and magnetometers) and an XW-GI 5630 BDS receiver were installed on agricultural equipment.
  • Data Fusion Strategy: The system utilizes a centralized Kalman filter. This means that raw data from both the INS and BDS are fed into a single Kalman filter algorithm.
  • Kalman Filter Implementation:
    • State Equation: The paper defines a comprehensive state vector that includes errors in latitude, longitude, velocity (eastward, northward), attitude (eastward, northward, upward), gyro random drift, and accelerometer zero bias. A state transition matrix (F) and a process noise matrix (G) are defined to model how these states evolve over time.
  • Measurement Equation: This equation relates the observed measurements (from INS and BDS) to the system's state. The paper formulates measurement equations for both position and velocity, accounting for the errors from each sensor. The observation matrix (H) is crucial here, mapping the state vector to the expected measurements.
  • System Switching: A key feature is the system's ability to switch modes. When BDS signals are valid, the system uses fused data for correction. If BDS signals become invalid (e.g., due to signal blockage), the system seamlessly switches to an INS-only mode, leveraging the Kalman filter's previous corrections to maintain reasonable accuracy for a short period until BDS signals are reacquired.
  • Software Interface: A man-machine interactive interface was developed using MATLAB GUI for data acquisition, display, and storage, providing a user-friendly way to monitor the roll angle in real-time.

What's novel here is the specific application of this Kalman filtering fusion to agricultural equipment roll angle estimation, combining the distinct characteristics of INS and BDS for this particular task, and the implementation of a robust switching mechanism.

Datasets

The paper does not introduce new benchmark datasets. Instead, it relies on field experimental data collected during plowing and sowing operations at Jilin University. The ground truth for roll angle measurements was obtained using a high-precision electronic level bar.

Key Insights and Ideas

  • Synergy of INS and BDS: The most significant insight is the demonstration that fusing INS and BDS data provides a substantial improvement in roll angle accuracy over using either sensor in isolation. This highlights the power of sensor fusion in overcoming individual sensor limitations.
  • Kalman Filter's Effectiveness: The Kalman filter proves to be a robust tool for state estimation and error correction in this dynamic agricultural environment. Its ability to model system dynamics and measurement uncertainties is key to achieving the observed accuracy gains.
  • Resilience to Signal Loss: The implemented switching mechanism, allowing the system to operate on INS data alone when BDS is unavailable, is critical for practical applications. This ensures continuous monitoring even in challenging environments where satellite signals might be intermittent.
  • Practical Applicability: The successful application of the integrated system in real-world field experiments validates its feasibility and effectiveness for agricultural machinery.

Key Results

The paper quantifies the performance improvement through mean absolute errors (MAE):

  • INS only: MAE of 0.78°
  • BDS only: MAE of 0.75°
  • Integrated INS/BDS System: MAE of 0.72° (as stated in the abstract, though Table 3 shows 0.73° for the integrated system, which is a minor discrepancy but the trend is clear).

While the absolute error reduction might seem small, it represents a significant improvement in precision for critical agricultural operations. The integrated system also demonstrated the lowest maximum absolute error (0.93°) compared to INS (0.98°) and BDS (0.95°). The system also performed well even during periods of satellite signal loss (50-70s).

Limitations and Future Work

  • Environmental Factors: The experiments were conducted under sunny conditions. The paper acknowledges that factors like tractor vibration can introduce errors, and the performance under more adverse weather or extreme terrain conditions might need further investigation.
  • Sensor Cost: While the paper mentions using "low-cost" sensors, the overall cost-effectiveness for widespread adoption in all agricultural machinery might be a consideration.
  • Generalizability: The study focused on roll angle. Future work could extend this fusion approach to accurately estimate pitch and yaw angles for a more complete 3D attitude estimation.
  • Algorithm Optimization: Further optimization of the Kalman filter parameters and potentially exploring more advanced fusion algorithms (e.g., Extended Kalman Filter, Unscented Kalman Filter) could lead to even greater precision.
  • Autonomous Operation Integration: The paper lays the groundwork for navigation and independent operation. Future research could focus on directly integrating this roll angle estimation into control systems for autonomous agricultural vehicles.

Summary updated

Abstract

Accurately obtaining roll angles is one of the key technologies to improve the positioning accuracy and operation quality of agricultural equipment. Given the demand for the acquisition of agricultural equipment roll angles, a roll angle monitoring model based on Kalman filtering and multi-source information fusion was established by using the MTi-300 AHRS inertial sensor (INS) and XW-GI 5630 BeiDou Navigation Satellite System (BDS), which were installed on agricultural equipment. Data of the INS and BDS were fused by MATLAB; then, Kalman filter was used to optimize the data, and the state equation and measurement equation of the integrated system were established. Then, an integrated monitoring terminal man-machine interactive interface was designed on MATLAB GUI, and a roll angle monitoring system based on the INS and BDS was designed and applied into field experiments. The mean absolute error of the integrated monitoring system based on multi-source information fusion during field experiments was 0.72°, which was smaller compared with the mean absolute errors of roll angle monitored by the INS and BDS independently (0.78° and 0.75°, respectively). Thus, the roll angle integrated model improves monitoring precision and underlies future research on navigation and independent operation of agricultural equipment.
BibTeX
@inproceedings{li2020acquisition,
  author = {Yang Li and Honglei Jia and Jiangtao Qi and Huibin Sun and Xinliang Tian and Huili Liu and Xuhui Fan},
  title = {{An Acquisition Method of Agricultural Equipment Roll Angle Based on Multi-Source Information Fusion}},
  booktitle = {Sensors},
  year = {2020},
  doi = {10.3390/s20072082},
}
Figure