Agricultural robots have been increasingly recognized as the key intelligent equipment for high productivity and quality in sustainable farming. It is often required to integrate environmental perception, autonomous decision-making, and precise execution in a complex field. However, spatiotemporal variations in soil conditions, crop morphology, terrain landscapes, loading, and meteorological factors have introduced substantial uncertainty into robotic perception, decision-making, and control systems, particularly for the open, dynamic, and unstructured scenarios. Agricultural robots can be extended into the crop-robot-environment coupling system. As a result, their intelligent operation is still restricted by incomplete sensing information, insufficient model accuracy, limited adaptability of control strategies, low repeatability of field validation, and weak coordination among robots, tasks, and environments. Digital twin can be expected to provide a promising potential approach for the dynamic connection between physical entities and virtual models, according to real-time data interaction, model updating, and predictive simulation. Virtual representations can be constructed to integrate agricultural robots, operating environments, and task processes. Therefore, digital twins can support state monitoring, process prediction, strategy evaluation, and control optimization, thereby reducing the dependence on costly and time-consuming field trials. In this study, a systematic review was presented on the technical pathways and prospects of digital twins in agricultural robots. Agricultural machinery evolved from conventional equipment to intelligent robotic systems. Agricultural robots were summarized from the perspectives of perception, decision-making, and control systems. Furthermore, the concept, system elements, and functional value of digital twins were analyzed to clarify their applicability to agricultural robotic systems under complex field conditions. Empowerment pathways of digital twins were summarized within a perception-decision-control framework. In perception enhancement, dynamic virtual mapping was realized to integrate digital twins with the robot state data, environmental information, historical data, and mechanisms. Multi-source data fusion, virtual sensing, state reconstruction, and anomaly correction were used to improve the reliability and interpretability of sensing information. In decision optimization, controllable and evaluable virtual environments were provided for process simulation, multi-strategy comparison, intelligent algorithm training, and policy iteration. Path planning, task allocation, parameter matching, and risk avoidance were shifted from experience-based decision-making into prediction and optimization. In control improvement, digital twins were used to connect decision outputs, execution, external disturbances, and control constraints using online updating, disturbance prediction, and adaptive parameter tuning, thereby supporting control compensation, parameter adaptation, and closed-loop optimization. Representative scenarios were then analyzed, including fault and health prediction, autonomous operation, as well as scheduling and coordination. In fault and health prediction, real sensing was integrated with virtual simulation data for the key component and anomaly detection, degradation prediction, and maintenance. In autonomous operation, virtual testing and optimization of motion and operation-control algorithms were adjusted to trajectories, parameters, and execution strategies in agricultural robots, according to field conditions. In scheduling and coordination, field environments, robot fleets, tasks, and environmental constraints were simulated to evaluate scheduling and then optimize task allocation, path coordination, and resource deployment. Digital twins enhanced the role of agricultural robots as information hubs in the production chain. Transition conductions were accelerated from mechanical into data-driven intelligent systems. Finally, the challenges were summarized from the current application of the digital twin. Their future directions were also proposed in agricultural robotics. This finding can contribute to the theoretical and practical guidance of digital twins in agricultural robots.

Behavioral responses of group-housed pigs are strongly influenced by pigsty environmental conditions, yet their dynamic coupling is difficult to quantify under commercial farming scenarios. This difficulty arises from high inter-pig similarity, complex interactions, and rapidly changing environmental conditions, which pose significant challenges for existing vision-based multi-pig behavior detection and tracking methods. To address these challenges, this study proposes a PIG-Net–based dynamic coupling analysis framework that integrates behavior detection, multi-pig tracking, and behavior-environment interaction analysis. The model uses an EfficientRepBiFusion backbone with bidirectional feature fusion and a lightweight LSDGCD detection head, achieving mean Average Precision (mAP) of 93.5% for PIG YOLO on four pig behaviors—standing, dog-sitting, lateral lying, and prone lying. The integrated PIG-Net system achieves stable tracking performance with identification average rate (IDF1) of 90.7%, multiple object tracking accuracy (MOTA) of 88.6%, and a real-time processing speed of 26 FPS, while environmental sensors continuously record temperature, humidity, and CO₂ levels for long-term correlation analysis. Based on long-term monitoring, Pearson correlation analysis was applied to quantify the associations between pig behaviors and environmental factors, highlighting significant correlations with coefficients |r| ranging from 0.65 to 0.76. By combining these quantitative results with temporal and dimensionality reduction analyses, temperature, humidity, and CO₂ were identified as the primary environmental drivers. Active behaviors decreased under elevated temperature and humidity and increased during cooler and drier periods, whereas prone lying and lateral lying increased under thermal and moisture stress. Elevated CO₂ concentrations further suppressed activity, reflecting inhibitory effects of degraded air quality. These findings provide a quantitative basis for behavior-environment coupling assessment and early health warning in group-housed pigs.

Large high clearance self-propelled sprayers were widely used in field plant protection due to their high-efficiency operation capabilities. Influenced by the characteristics of field operations such as high power, heavy weight, high ground clearance, and fast operation speed, the comprehensive requirements for the ride comfort, handling stability and road friendliness of the sprayer were increasingly strong. At the present stage, the chassis structure of the high clearance self-propelled sprayer that attaches great importance to the improvement of comprehensive performance still has the problems of severe bumps, weak handling performance and serious road damage in complex field environments. Therefore, this paper proposes an optimization design method for hydro-pneumatic suspension system of a high clearance self-propelled sprayer based on the improved MOPSO (Multi-Objective Particle Swarm Optimization) algorithm, covering the entire process of configuration design, parameter intelligent optimization, and system verification of the high clearance self-propelled sprayer chassis. Specifically, chassis structure of the hydro-pneumatic suspension suitable for the high clearance self-propelled sprayer was designed, and a design method combining the improved MOPSO algorithm based on time-varying fusion strategy and adaptive update with the parameter optimization of hydro-pneumatic suspension based on this algorithm was proposed, and finally the software simulation and bench performance verification were carried out. The results show that the optimized hydro-pneumatic suspension has excellent vibration reduction effect, and the body acceleration, suspension dynamic deflection and tire deflection were increased by 16.5%, 9.9% and 0.9% respectively, compared with those before optimization. The comprehensive performance of the hydro-pneumatic suspension designed in this study is better than that of the traditional suspension.