This paper proposes an improved initial alignment method for a strap-down inertial navigation system/global navigation satellite system (SINS/GNSS) integrated navigation system with large misalignment angles. Its methodology is based on the three-dimensional special Euclidean group and extended Kalman filter (SE2(3)/EKF) and aims to overcome the challenges of achieving fast alignment under large misalignment angles using traditional methods. To accurately characterize the state errors of attitude, velocity, and position, these elements are constructed as elements of a Lie group. The nonlinear error on the Lie group can then be well quantified. Additionally, a group vector mixed error model is developed, taking into account the zero bias errors of gyroscopes and accelerometers. Using this new error definition, a GNSS-assisted SINS dynamic initial alignment algorithm is derived, which is based on the invariance of velocity and position measurements. Simulation experiments demonstrate that the alignment method based on SE2(3)/EKF can achieve a higher accuracy in various scenarios with large misalignment angles, while the attitude error can be rapidly reduced to a lower level.
Introduction: Oil-based emulsion solution is a common pesticide formulation in agricultural spraying, and its spray characteristics are different from that of water spraying. The well understanding of its spray characteristics is the theoretical basis to improve the pesticide spraying technology. The objective of the present study is to deepen the understanding of the spray characteristics of oil-based emulsion. Method: In this paper, the spatial distribution characteristics of spray droplets of oil-based emulsion were captured visually using the high-speed photomicrography. On the basis of image processing method, the droplet size and distribution density of spray droplets at different spatial locations were analyzed quantitatively. The effects of nozzle configuration and emulsion concentration on spray structures and droplet spatial distribution were discussed. Results: Oil-based emulsion produced a special perforation atomization mechanism compared with water spray, which led to the increase of spray droplet size and distribution density. Nozzle configuration had a significant effect on oil-based emulsion spray, with the nozzle changed from ST110-01 to ST110-03 and ST110-05; the sheet lengths increased to 18 and 28 mm, respectively, whereas the volumetric median diameters increased to 51.19% and 76.00%, respectively. With emulsion concentration increased from 0.02% to 0.1% and 0.5%, the volumetric median diameters increased to 5.17% and 14.56%, respectively. Discussion: The spray droplet size of oil-based emulsion spray can be scaled by the equivalent diameter of discharge orifice of nozzles. The products of volumetric median diameters and corresponding surface tensions were nearly constant for the oil-based emulsion spray of different emulsion concentrations. It is expected that this research could provide theoretical support for improving the spraying technology of oil-based emulsion and increasing the utilization of pesticide.
Introduction: Development of weed and crop detection algorithms provides theoretical support for weed control and becomes an effective tool for the site-specific weed management. For weed and crop object detection tasks in the field, there is often a large difference between the number of weed and crop, resulting in an unbalanced distribution of samples and further posing difficulties for the detection task. In addition, most developed models tend to miss the small weed objects, leading to unsatisfied detection results. To overcome these issues, we proposed a pixel-level synthesization data augmentation method and a TIA-YOLOv5 network for weed and crop detection in the complex field environment. Methods: The pixel-level synthesization data augmentation method generated synthetic images by pasting weed pixels into original images. In the TIA-YOLOv5, a transformer encoder block was added to the backbone to improve the sensitivity of the model to weeds, a channel feature fusion with involution (CFFI) strategy was proposed for channel feature fusion while reducing information loss, and adaptive spatial feature fusion (ASFF) was introduced for feature fusion of different scales in the prediction head. Results: Test results with a publicly available sugarbeet dataset showed that the proposed TIA-YOLOv5 network yielded an F1-scoreweed, APweed and mAP@0.5 of 70.0%, 80.8% and 90.0%, respectively, which was 11.8%, 11.3% and 5.9% higher than the baseline YOLOv5 model. And the detection speed reached 20.8 FPS. Discussion: In this paper, a fast and accurate workflow including a pixel-level synthesization data augmentation method and a TIA-YOLOv5 network was proposed for real-time weed and crop detection in the field. The proposed method improved the detection accuracy and speed, providing very promising detection results.
An improved robust cubature Kalman filter (RCKF) based on variational Bayesian (VB) and transformed posterior sigma points error is proposed in this paper, which not only retains the robustness of RCKF, but also exhibits adaptivity in the presence of time-varying noise. First, a novel sigma-point update framework with uncertainties reduction is developed by employing the transformed posterior sigma points error. Then the VB is used to estimate the time-varying measurement noise, where the state-dependent noise is addressed in the iteratively parameter estimation. The new filter not only reduces the uncertainty on sigma points generation but also accelerates the convergence of VB-based noise estimation. The effectiveness of the proposed filter is verified on integrated navigation, and numerical simulations demonstrate that VB-RCKF outperforms VB-CKF and RCKF.
In order to improve the accuracy of GNSS/INS working in GNSS-denied environment, a robust cubature Kalman filter (RCKF) is developed by considering colored measurement noise and missing observations. First, an improved cubature Kalman filter (CKF) is derived by considering colored measurement noise, where the time-differencing approach is applied to yield new observations. Then, after analyzing the disadvantages of existing methods, the measurement augment in processing colored noise is translated into processing the uncertainties of CKF, and new sigma point update framework is utilized to account for the bounded model uncertainties. By reusing the diffused sigma points and approximation residual in the prediction stage of CKF, the RCKF is developed and its error performance is analyzed theoretically. Results of numerical experiment and field test reveal that RCKF is more robust than CKF and extended Kalman filter (EKF), and compared with EKF, the heading error of land vehicle is reduced by about 72.4%.
In order to improve the accuracy and robustness of GNSS/INS navigation system, an improved iterated cubature Kalman filter (IICKF) is proposed by considering the state-dependent noise and system uncertainty. First, a simplified framework of iterated Gaussian filter is derived by using damped Newton-Raphson algorithm and online noise estimator. Then the effect of state-dependent noise coming from iterated update is analyzed theoretically, and an augmented form of CKF algorithm is applied to improve the estimation accuracy. The performance of IICKF is verified by field test and numerical simulation, and results reveal that, compared with non-iterated filter, iterated filter is less sensitive to the system uncertainty, and IICKF improves the accuracy of yaw, roll and pitch by 48.9%, 73.1% and 83.3%, respectively, compared with traditional iterated KF.
