Sensor fusion for improving the estimation of roll and pitch for an agricultural sprayer

Sensor fusion using a Discrete Kalman Filter (DKF) was applied to integrate the attitude angle estimates obtained from a Digital Elevation Model (DEM) and a Terrain Compensation Module (TCM) sensor to improve the roll and pitch angle estimates of a self-propelled sprayer. Vehicle attitude and field elevation were measured at two speeds (5.6 and 9.6 km h−1), using a self-propelled agricultural sprayer equipped with Real-Time Kinematic-Differential Global Positioning System (RTK-DGPS) receiver, a TCM sensor and an Inertial Measurement Unit (IMU). The DKF-, DEM-based roll and pitch estimates, the TCM sensor roll and the GPS-based pitch estimates were compared with the reference IMU measurements to validate the performance of the fusion algorithm. A second order autoregressive (AR) model was developed to model the irregular spiked noise in TCM roll and high-frequency noise in GPS-based pitch angle estimates.The AR modelled error states were incorporated into the DKF algorithm and the measurement noise covariance was estimated from the AR model, which limited the fine tuning of noise covariance to the process noise covariance only. The DKF was able to overcome the out-of-bound situation (data outage) in DEM while it estimated the attitude of the self-propelled sprayer. Additionally, the fusion algorithm was proven to be effective in improving attitude estimate of the self-propelled agricultural sprayer, which can be extended to facilitate the automatic control of the implements that interact with the soil surface on an undulating topographic surface.