Research Progress on Agricultural Equipments for Precision Planting and Harvesting

As the global population continues to grow, arable land resources are becoming increasingly scarce, climate change is intensifying, and the number of people employed in agriculture continues to decline; agricultural production is facing the dual pressures of yield enhancement and sustainable resource utilization. Conventional crude agriculture is predicated on the experience of planting and harvesting, and large-scale mechanical operations are generally employed. However, such practices are not without their drawbacks, as they tend to result in seed waste, harvesting loss, low yield, and poor quality of operation. This phenomenon consequently gives rise to diminished resource utilization and an augmented environmental burden.

In this context, the employment of precision agriculture technologies is regarded as a pivotal factor driving the process of agricultural modernization. The development of precision planting and harvesting equipment represents a critical technological advancement, with the potential to transform the conventional mechanized production model. The evolution of agricultural equipment towards intelligence, data, and adaptive direction [1,2,3] is driven by the rapid development of the Internet of Things (IoT) [4], satellite navigation (GNSS) [5], artificial intelligence (AI) [6], big data analysis [7], and the web-based spatial decision support system [8]. Precision planting equipment has been demonstrated to optimize the spatial distribution of crops and the growth environment through the implementation of variable planting techniques [9,10], depth control mechanisms [11,12], and real-time monitoring systems [13,14]. In addition, the utilization of precision harvesting equipment has been demonstrated to facilitate the efficient and low-loss harvesting [15,16] and quality grading of produce [17,18]. This is achieved through the integration of multi-sensor fusion technology [19,20], yield prediction algorithms [20,21,22], and adaptive regulation mechanisms [3,23,24]. However, there is an imperative for the enhancement of the stability of the existing equipment. In complex farmland scenarios, there is a requirement for the capacity for collaborative decision-making with multiple sources of data, as well as for the incorporation of intelligence throughout the entire chain. There is an urgent need for interdisciplinary technology integration and systematic innovation.

The employment of precision planting and harvesting technologies improves agricultural production efficiency and resource sustainability. The optimization of the spatial distribution of seeds and soil suitability through the utilization of precision planting equipment has been demonstrated to reduce seed wastage [25]. The integration of precision harvesting equipment with real-time monitoring capabilities has been demonstrated to result in a reduction in field loss and an enhancement in crop yield [26]. In consideration of the challenges posed by climate change and labor, the employment of an autonomous navigation system [27] in conjunction with a climate prediction model-driven dynamic decision-making system [28] has been demonstrated to alleviate the pressure on the agricultural workforce due to its advanced age, whilst concomitantly enhancing disaster resilience. This technology system has the capacity to promote the digital transformation of agriculture, whilst simultaneously constructing a digital twin farm based on soil–crop–weather multi-dimensional data [29]. The system has the capacity to facilitate data support for the optimization of the entire agricultural chain and the construction of precision agriculture standards. Furthermore, it has been demonstrated that a reduction in the ineffective energy consumption of agricultural machinery can be achieved through the implementation of precision variable operations [30]. This development has been identified as a significant technical approach that contributes to the realization of the objective of carbon neutrality in the agricultural sector. Moreover, it is anticipated that this initiative will facilitate the advancement of green agriculture in a comprehensive manner, thereby promoting the sustainable development of agriculture.

This Special Issue is devoted to an examination of technological innovations in the domains of precision planting and harvesting. The value of these innovations is manifested in two principal ways. Firstly, they contribute to the assurance of food security and the enhancement of farmers’ income. Secondly, they serve as a pivotal medium for the implementation of the United Nations’ Sustainable Development Goals (SDGs). Among these goals, “End hunger, achieve food security and improved nutrition and promote sustainable agriculture” and “Ensure sustainable consumption and production patterns” are notable. As the fundamental support system propelling the transformation of global agriculture 4.0, technological advancements in this domain are poised to directly enhance agricultural production efficiency, optimize resource utilization, and minimize carbon emissions. It is anticipated that these developments will provide comprehensive solutions for the cultivation of smart agriculture.

Thus far, this Special Issue has amassed a total of five papers, which primarily concentrate on three significant technical impediments—the inadequate adaptability of equipment mechanisms, the suboptimal dynamic matching accuracy, and the inadequate complex environment sensing ability.

For the first technical impediment, characterized by the inadequate adaptability of equipment mechanisms, the crux of the issue pertains to the inherent limitations exhibited by conventional agricultural machinery in complex operating scenarios. These limitations encompass such characteristics as irrational mechanical structure design, a high propensity for secondary damage, and suboptimal energy consumption efficiency.

To address the problem of sudden yield reduction in the second season due to stubble compression after regeneration rice harvesting, Xing et al. [31] proposed an eccentric parallelogram-type ratoon rice stub straightening device. In this study, a coupled DEM–MBD simulation was employed to simulate the contact mechanical properties of the stubble and the correcting teeth. The three-factor and three-level response surface method was utilized to optimize the key parameters (a rotational speed of 75 rpm, a forward speed of 1.4 m/s, and a stub entry angle of 39°) with the objective of achieving a 90.35% straightening rate in the mechanized harvesting crush zone. The designed ratoon rice stub straightening device can meet the requirements of field operations and effectively improve the yield of the second season of ratoon rice, which has a positive significance for promoting the cultivation of ratoon rice.

In addressing the challenges posed by inadequate power matching and suboptimal operating efficiency in root herb transplanter systems, Yu et al. [32] devised a fuzzy PID electronic control system underpinned by STM32 technology. The construction of a speed closed-loop control model is achieved by integrating the real-time acquisition of track traveling speed using a Hall sensor with roller motor encoder feedback. The simulation results demonstrate that the steady-state error is reduced by 36.41% and the steady-state time is shortened by 47.26% in comparison with the traditional PID. Furthermore, a transplant qualification rate of more than 95.6% is achieved in the transplanting experiment of Codonopsis pilosula and Astragalus membranaceus, thus resolving the issue of unstable mechanical power output in a mountain environment.

The second technical impediment, characterized by suboptimal dynamic matching accuracy, can be attributed to the disparity between the rapid movement of materials and the slower pace of mechanical operations during the planting and harvesting processes. This discrepancy has been shown to result in leakage, reseeding, and elevated damage rates.

Wang et al. [33] proposed an air–mass matching strategy for multi-winged curved combination centrifugal fans with the aim of improving the low cleaning efficiency of the longitudinal axial flow threshing device of a combine harvester. The design of the three-sub fan with transverse zoning was established based on a non-uniform airflow field model, as simulated using CFD–DEM coupled simulation. Experiments have demonstrated that the optimized cleaning loss rate is reduced by up to 25%, and the impurity content rate is reduced by 32.2%. This solves the local overloading or underloading problems caused by traditional homogeneous airflow.

In addressing the challenge of inadequate positioning accuracy in the transportation of seedling trays, Yao et al. [34] devised a dual-axis positioning tray conveying device. The implementation of a dual-sensor positioning algorithm, in conjunction with a variable displacement positioning method, has been demonstrated to achieve the maximum deviation of 1.34 mm and 0.99 mm in the initial X/Y positioning of the seedling tray, with an intermission conveying value of 0.85 mm and 0.98 mm. The device has been demonstrated to facilitate the adaptive adjustment of 21- to 288-hole seedling trays. The designed device provides essential technological foundations for seedling tray transport and retrieval steps in fully automated transplanting machines.

The third technical impediment, characterized by the inadequate complex environment sensing ability, is fundamentally due to discrepancies in natural lighting, target occlusion, and the presence of small-scale features. These factors result in elevated rates of false and missed detections within the framework of conventional detection models.

In addressing the challenge of the inadequate detection accuracy of tea buds amidst the intricate environment of tea gardens, Chen et al. [35] proposed an improved object detection model—RT-DETR-Tea. The model incorporates a GD-Tea feature fusion module, which serves to enhance the expression of large and small tea bud features in natural environments. The experimental results demonstrate that the precision and mean average precision of the proposed RT-DETR-Tea model are 96.1% and 79.7%, respectively, which are increased by 5.2% and 2.4% compared to those of the original model, indicating the model’s effectiveness.

These five papers collected in this Special Issue concentrate on the three core bottlenecks of agricultural mechanization and propose innovative solutions from three perspectives—institutional design, dynamic control, and intelligent perception. It is evident that the adaptability of equipment to complex operating objects can be significantly improved through DEM–MBD simulation, the parametric response surface method, and non-uniform dynamics design. Furthermore, the integration of fuzzy PID, multi-sensor closed-loop feedback, and the gas-quality matching strategy allows for the precise matching of speed, position, and load. Finally, the transformer framework and multi-scale feature fusion overcome the precision limitations of small-target detection in the natural environment.

It is imperative that future research endeavors delve deeper into multi-machine cooperative operations (e.g., planting–transplanting–harvesting linkage), the lightweight deployment of edge computing (reducing model arithmetic requirements), and the virtual verification of digital twins (reducing the cost of physical trial and error). These areas of research are crucial in propelling the development of agricultural equipment towards full-process intelligence and networking.