In automated orchard operations, the straight-line locomotion stability of ground-based weeding robots is critical for ensuring path coverage efficiency and operational reliability. To address the response lag and high-frequency oscillations often observed in conventional PID and fixed-lookahead Pure Pursuit controllers, this study proposes an adaptive lookahead Pure Pursuit method incorporating angular velocity feedback. By dynamically adjusting the lookahead distance according to real-time attitude changes, the method enhances coordination between path curvature and robot stability. To enable systematic evaluation, three time-series-based metrics are introduced: mean absolute yaw error (MAYE), peak-to-peak fluctuation amplitude, and the standard deviation of angular velocity, with overshoot occurrences included as an additional indicator. Field experiments demonstrate that the proposed method outperforms baseline algorithms, achieving lower yaw errors (0.61–0.66°), reduced maximum deviation (≤3.7°), and smaller steady-state variance (<0.44°²), thereby suppressing high-frequency jitter and improving turning convergence. Under typical working conditions, the method achieved a mean yaw deviation of 0.6602°, a fluctuation of 5.59°, an angular velocity standard deviation of 10.79°/s, and 155 overshoot instances. The yaw angle remained concentrated around the target orientation, while angular velocity responses stayed stable without loss-of-control events, indicating a favorable balance between responsiveness and smoothness. Overall, the study validates the robustness and adaptability of the proposed strategy in complex orchard scenarios and establishes a reusable evaluation framework, offering theoretical insights and practical guidance for intelligent agricultural machinery optimization.

The multiscale interactive system composed of wind, leaves, and droplets serves as a critical dynamic unit in precision orchard spraying. Its coupling mechanisms fundamentally influence pesticide transport pathways, deposition patterns, and drift behavior within crop canopies, forming the foundational basis for achieving intelligent and site-specific spraying operations. This review systematically examines the synergistic dynamics across three hierarchical scales: Droplet–leaf surface wetting and adhesion at the microscale; leaf cluster motion responses at the mesoscale; and the modulation of airflow and spray plume diffusion by canopy architecture at the macroscale. Key variables affecting spray performance—such as wind speed and turbulence structure, leaf biomechanical properties, droplet size and electrostatic characteristics, and spatial canopy heterogeneity—are identified and analyzed. Furthermore, current advances in multiscale modeling approaches and their corresponding experimental validation techniques are critically evaluated, along with their practical boundaries of applicability. Results indicate that while substantial progress has been made at individual scales, significant bottlenecks remain in the integration of cross-scale models, real-time acquisition of critical parameters, and the establishment of high-fidelity experimental platforms. Future research should prioritize the development of unified coupling frameworks, the integration of physics-based and data-driven modeling strategies, and the deployment of multimodal sensing technologies for real-time intelligent spray decision-making. These efforts are expected to provide both theoretical foundations and technological support for advancing precision and intelligent orchard spraying systems.

Hilly orchards are characterized by undulating terrain, canopy occlusion, irregular tree-row structures, and marked variations in ground adhesion conditions, which challenge the autonomous navigation of agricultural machinery by degrading perception and localization, increasing reference path uncertainty, and reducing closed-loop robustness. Perception and localization errors can propagate progressively along the chain of “sensor observation–vehicle pose estimation/environmental-structure perception–reference path generation–path-tracking controller inputs–vehicle closed-loop response,” ultimately affecting path-tracking accuracy, control smoothness, and operational safety. This review examines perception and localization error propagation and robust path-tracking control for agricultural machinery in hilly orchards. The review first summarizes the navigation roles and error characteristics of key observation sources and then analyzes how perception, localization, and reference path generation provide state variables, reference variables, and safety constraints to controllers. It subsequently compares typical path-tracking methods under multi-source disturbances and actuator constraints and discusses key challenges and an integrated robust design direction coupling perception and localization, reference path generation, and path-tracking control. The proposed integrated framework represents a synthesis-derived design direction rather than a complete architecture that has already been experimentally validated in hilly orchard field environments.

In wind-assisted orchard spraying operations, the dynamic response of leaves—manifested through changes in their posture—critically influences droplet deposition on both sides of the leaf surface and the penetration depth into the canopy. These factors are pivotal in determining spray coverage and the spatial distribution of pesticide efficacy. However, current research lacks comprehensive quantification and correlation analysis of the temporal response characteristics of leaves under wind disturbances. To address this gap, a systematic analytical framework was proposed, integrating real-time leaf segmentation and tracking, geometric feature quantification, and statistical correlation modeling. High-frame-rate videos of fluttering leaves were acquired under controlled wind conditions, and background segmentation was performed using principal component analysis (PCA) followed by clustering in the reduced feature space. A fine-tuned Segment Anything Model 2 (SAM2-FT) was employed to extract dynamic leaf masks and enable frame-by-frame tracking. Based on the extracted masks, time series of leaf area and inclination angle were constructed. Subsequently, regression analysis, cross-correlation functions, and Granger causality tests were applied to investigate cooperative responses and potential driving relationships among leaves. Results showed that the SAM2-FT model significantly outperformed the YOLO series in segmentation accuracy, achieving a precision of 98.7% and recall of 97.48%. Leaf area exhibited strong linear coupling and directional causality, while angular responses showed weaker correlations but demonstrated localized synchronization. This study offers a methodological foundation for quantifying temporal dynamics in wind–leaf systems and provides theoretical insights for the adaptive control and optimization of intelligent spraying strategies.

For the problem that, under the soybean-maize strip intercropping pattern, the installation of anti-drift shields changes the mass and inertia characteristics of the boom of a shielded sprayer, making it prone to severe vibration under complex field excitations and thereby affecting application uniformity and operational safety, a semi-active boom vibration damping system based on a magnetorheological damper was designed and tested. First, an equivalent mechanical model of the boom considering added mass and stiffness variation was established, and the characteristics of excitation sources such as road spectra and start-stop impacts were clarified. Second, based on the Bouc–Wen model, the magnetorheological damper was selected and its output force boundaries were designed. The stroke was determined to be 80–110 mm, and the maximum damping forces in the compression and rebound strokes were 1.15 kN and 3.44 kN, respectively. Furthermore, a semi-active vibration damping system with an LK3U-14MT PLC as the core controller was developed. Finally, simulation analysis, free-vibration decay tests, obstacle-crossing tests, and field operation tests were conducted to verify the vibration damping performance of the developed semi-active system. The field test results showed that, at typical operating speeds of 3–5 km/h, the semi-active vibration damping system reduced the maximum vertical amplitude at the boom tip by 28.6–43.1% and the maximum inclination angle by 12.2–55.6%, effectively improving the attitude stability and roll resistance of the boom of the shielded sprayer.

Real-time kinematic (RTK) navigation, which enables centimeter-level positioning accuracy through carrier-phase differential correction, provides high-accuracy positioning for orchard robots, but short-term outages caused by canopy occlusion and signal interference may interrupt path guidance and increase lateral drift. To address this issue, this study proposes a vision-based takeover navigation method for orchard robots under short-term RTK failure conditions. First, an improved YOLOv11-based road segmentation and completion model, termed YOLOv11-VF, was developed. By introducing a Squeeze-and-Excitation (SE) channel attention mechanism, the model jointly perceives visible road regions and occluded road completion regions, thereby producing continuous and complete road semantic representations. Second, a structured geometric road representation was constructed from the segmentation results to extract the navigation reference line, and a joint direction-position constraint mechanism was established by integrating the reference line with the robot reference view axis. A hierarchical constraint strategy based on a travel corridor and a deadband region was further designed to jointly determine heading deviation and lateral drift. Finally, road segmentation, navigation-line extraction, parameter analysis, and vision-based takeover experiments were conducted in a standardized orchard environment. The results showed that YOLOv11-VF achieved Precision, Recall, AP₅₀, mAP_@0.5:0.95, and F1 values of 92.31%, 88.56%, 94.40%, 67.41%, and 90.40, respectively, showing the best overall segmentation performance among all compared models while maintaining good real-time performance. The proposed method also demonstrated high consistency in navigation-line extraction and maintained mean absolute deviations of 0.0176 ± 0.0041 m to 0.0718 ± 0.0138 m during RTK outage intervals over 10 repeated trials, indicating good path-following capability and operational stability.