Smart agriculture development mainly depends on the intelligence and reliability of autonomous agricultural machinery. Automatic navigation systems (ANSs) play a key role in intelligent agricultural machinery design, as they not only reduce farmers’ workloads but also improve their land utilization rates. In this paper, a tractor ANS based on dynamic path search and a fuzzy Stanley model (FSM) was designed, and its capability for whole-field path tracking was tested. First, the tracking performance of the steering control module was validated after the automatic reconstruction of the tractor platform. Then, a navigation decision system was established based on a unified reference waypoint search framework, where the path generation for whole-field coverage was presented. Finally, the gain coefficient of the Stanley model (SM) was adjusted adaptively according to the tracking error by utilizing the fuzzy logic controller. Subsequently, the developed tractor ANS was tested in the field. The experiment’s results indicate that the FSM outperformed the SM in straight path tracking and whole-field path tracking. When the tractor traveled at a speed of 1 m/s, the maximum lateral tracking error for the straight path was 10 cm, and the average lateral tracking error was 5.2 cm, showing improvements of 16.7% and 10.3% compared to the SM. Whole-field autonomous navigation showed that the maximum lateral tracking error was improved from 34 cm for the SM to 27 cm for the FSM, a reduction of approximately 20.6%, illustrating the superiority of the FSM in the application of whole-field path tracking. As the maximum tracking error of whole-field autonomous navigation appears in the turning stage, where tractors often stop working, the designed ANS satisfies the requirements of a self-driving system for unmanned tractors.

Integrating global navigation satellite systems (GNSSs) with inertial navigation systems (INSs) has been widely recognized as an ideal solution for autonomous vehicle navigation. However, GNSSs suffer from disturbances and signal blocking inevitably, making the performance of GNSS/INSs degraded in the occurrence of measurement outliers. It has been proven that the sigma points-based Kalman filter (KF) performs better than an extended KF in cases where large prior uncertainty is present in the state estimation of a GNSS/INS. By modifying the sigma points directly, the resampling-free sigma point update framework (SUF) propagates more information excepting Gaussian moments of prescribed accuracy, based on which the resampling-free cubature Kalman filter (RCKF) was developed in our previous work. In order to enhance the adaptivity and robustness of the RCKF, the resampling-free SUF depending on dynamic prediction residue should be improved by suppressing the time-varying measurement outlier. In this paper, a robust observability-constrained RCKF (ROCRCKF) is proposed based on adaptive measurement noise covariance estimation and outlier detection, where the occurrence of measurement outliers is modelled by the Bernoulli variable and estimated with the state simultaneously. Experiments based on car-mounted GNSS/INSs are performed to verify the effectiveness of the ROCRCKF, and result indicates that the proposed algorithm outperforms the RCKF in the presence of measurement outliers, where the heading error and average root mean square error of the position are reduced from 1.96° and 6.38 m to 0.27° and 5.95 m, respectively. The ROCRCKF is robust against the measurement outliers and time-varying model uncertainty, making it suitable for the real-time implementation of GNSS/INSs in GNSS-challenged environments.

Automatic navigation based on dual-antenna real-time kinematic (RTK) positioning has been widely employed for unmanned agricultural machinery, whereas GNSS inevitably suffers from signal blocking and electromagnetic interference. In order to improve the reliability of an RTK-based navigation system in a GNSS-challenged environment, an integrated navigation system is preferred for autonomous navigation, which increases the complexity and cost of the navigation system. The information fusion of integrated navigation has been dominated by Kalman filter (KF) for several decades, but the KF cannot assimilate the known knowledge of the navigation context efficiently. In this paper, the geometric characteristics of the straight path and path-tracking error were employed to formulate the constraint measurement model, which suppresses the position error in the case of RTK-degraded scenarios. The pseudo-measurements were then imported into the KF framework, and the smoothed navigation state was generated as a byproduct, which improves the reliability of the RTK positioning without external sensors. The experiment result of the mobile vehicle automatic navigation indicates that the tracking error-constrained KF (EC-KF) outperforms the trajectory-constrained KF (TC-KF) and KF when the RTK system outputs a float or single-point position (SPP) solution. In the case where the duration of the SPP solution was 20 s, the positioning errors of the EC-KF and TC-KF were reduced by 38.50% and 24.04%, respectively, compared with those of the KF.

Unmanned aerial vehicle (UAV) imagery provides an efficient approach for monitoring cotton defoliation and boll-opening rates. Deep learning, particularly convolutional neural networks (CNNs), has been widely applied in image processing and agricultural monitoring, achieving strong performance in tasks such as disease detection, weed recognition, and yield prediction. However, existing models often suffer from heavy computational costs and slow inference speed, limiting their real-time deployment in agricultural fields. To address this challenge, we propose a lightweight cotton maturity recognition model, RTCMNet (Real-time Cotton Monitoring Network). By incorporating a multi-scale convolutional attention (MSCA) module and an efficient feature fusion strategy, RTCMNet achieves high accuracy with substantially reduced computational complexity. A UAV dataset was constructed using images collected in Xinjiang, and the proposed model was benchmarked against several state-of-the-art networks. Experimental results demonstrate that RTCMNet achieves 0.96 and 0.92 accuracy on defoliation rate and boll-opening rate classification tasks, respectively. Meanwhile, it contains only 0.35 M parameters—94% fewer than DenseNet121—and only requires an inference time of 33 ms, representing a 97% reduction compared to DenseNet121. Field tests further confirm its real-time performance and robustness on UAV platforms. Overall, RTCMNet provides an efficient and low-cost solution for UAV-based cotton maturity monitoring, supporting the advancement of precision agriculture.

In order to address the current deficit in research on discrete elements within the context of the cotton seed pressing process, a combination of physical tests and simulation experiments were conducted on Xinluzao 84 cottonseed. The Plackett-Burman experimental design was employed to identify the parameters that had a significant impact on the accumulation angle. Subsequently, the Box-Behnken design was employed to establish a second-order regression model for the accumulation angle and the significant parameters. This resulted in the optimal parameter combination for the maximum friction coefficient, minimum sliding friction coefficient and critical shear stress being 0.207, 0.388, and 5×10⁶ Pa, respectively. In order to calibrate the cottonseed bonding parameters, a Box-Behnken test was designed based on the results of the cottonseed particle crushing test. The optimal parameter combinations of normal stiffness, tangential stiffness, and bond radius were obtained as 1.25×10⁹ N/m², 7.57×10⁸ N/m², and 0.727 mm, respectively. The obtained parameters were verified by the simulation of stacking angle and the crushing test simulation. The relative error was 0.84% and 0.46%, respectively. This serves to validate the cottonseed bonding model and simulation parameters, thereby providing a foundation for optimising the core structural parameters of the cottonseed shelling machine.

Aiming at the problems of low operating efficiency and the unclear mechanisms in the bundling process of existing residual film recycling machines, this paper designs a chain network-type residual film bundling device and analyzes the motion characteristics of the film bundling process using high-speed camera technology. A mechanical analysis of the bundling process was conducted, and a test rig for the chain network residual film bundling device was built. The bundling process was studied via a high-speed camera. Field tests were carried out with the density of the film bale as the evaluation indicator and the forward speed of the machine, the rotational speed of the active film-removing roller, and the rotational speed of the film-rolling support roller as influencing factors. A Box–Behnken experimental design was used to optimize the working parameters of the device. The results show that when the machine’s forward speed is 5.8 km/h, the active stripping roller rotates at 170 rpm, the roll support roller operates at 210 rpm, and the film bale density reaches 124.44 kg/m³, with a relative error of only 1.34 kg/m³ compared to the predicted value. This verifies the effectiveness of the device and demonstrates that it can meet the requirements of mechanized residual film recycling.