Soil hydraulic properties are key parameters for characterizing water infiltration, storage, and transport in soils, as well as their interactions with surface runoff and groundwater recharge. They play an essential role in hydrological simulation, soil and water conservation assessment, and agricultural water management. However, conventional methods for estimating soil hydraulic properties are often labor-intensive, time-consuming, and limited in their applicability. To address these limitations, this study proposes a method for estimating soil hydraulic properties based on physics-informed neural networks (PINN) and the Richardson–Richards equation (RRE). By incorporating the governing equation of soil water movement as the physical constraint, the proposed method is trained only with soil moisture observations and does not require initial or boundary conditions as training constraints, thereby reducing modeling complexity while maintaining physical consistency.Three representative soil types, namely cork oak forest soil, Pinus tabuliformis forest soil, and farmland soil, were selected as the study objects. Time series of volumetric water content measured at four soil depths during infiltration experiments were used as the input data for the PINN model. Meanwhile, the inverse modeling module of HYDRUS was employed to estimate soil hydraulic parameters and construct reference soil water characteristic curves (SWCC) and unsaturated hydraulic conductivity (K_u) data for validating the PINN estimation results. The root mean square error (RMSE), relative root mean square error (RRMSE), and relative error were used to evaluate the estimation accuracy of the proposed model.The results showed that the RMSEs between the volumetric water contents estimated from the PINN-derived SWCCs and the measured volumetric water contents were 0.144, 0.040, and 0.015 cm³/cm³ for the cork oak forest soil, Pinus tabuliformis forest soil, and farmland soil, respectively, with corresponding RRMSEs of 39.26%, 9.45%, and 4.85%. In the unsaturated range where the volumetric water content was lower than 0.4 cm³/cm³, the RMSEs for the two forest soils further decreased to 0.072 and 0.008 cm³/cm³, respectively, while the corresponding RRMSEs decreased to 37.16% and 2.37%, indicating that the proposed model achieved higher estimation accuracy in the unsaturated range. The relative errors between the PINN-estimated saturated hydraulic conductivity (K_s) and the measured values were 2.56%, 2.50%, and 7.89%, respectively. Owing to the lack of measured unsaturated hydraulic conductivity data, the PINN-estimated K_u values were further compared with the reference hydraulic conductivity curves constructed from independent observations. The corresponding RMSEs were 3.0×10⁻⁴, 7.3×10⁻⁵, and 7.1×10⁻³ cm/min, with corresponding RRMSEs of 8.39%, 1.19%, and 39.96%, respectively. Overall, the proposed PINN model effectively captured the hydraulic behavior of different soil types and provided stable estimates of soil hydraulic properties. Moreover, the method does not require the explicit specification of initial or boundary conditions, thereby reducing modeling complexity and demonstrating strong potential under limited data availability. This study provides a reliable technical approach for soil hydrological analysis and offers promising applications in soil moisture dynamics research, agricultural water management, and ecohydrological modeling.

Stem expansion and contraction are closely related to the plant water status. It is of great significance to real-time monitor the stem expansion and contraction, in order to improve the utilization rate of agricultural water resources in irrigation systems. Therefore, the sensors have been mainly used to monitor the expansion and contraction of plant stems at present. The linear micro-displacement can be measured to assess the plant water status. However, it is still limited to the high cost and inconvenient installation. In this study, a flexible and wearable sensor was designed and developed using the piezoresistive effect. A flexible pressure electrode was used as the sensing element to monitor the pressure variation, which was attached to the surface of plant stems. Different from the linear micro-displacement sensor, the developed sensor was used to assess the pressure variation caused by the expansion and contraction of the stem. The pressure signal detection and data collection were also designed to convert the pressure into electrical signals for storage. A series of experiments were conducted to evaluate the performance of the pressure sensor. Firstly, the performance test and sensor calibration were conducted under laboratory conditions. Then, the pressure sensor was installed on a tomato stem in a greenhouse, in order to observe the pressure variation. A comparison was made with the linear micro-displacement sensor. Finally, the variations in the expansion and contraction of tomato stems were observed under sufficient irrigation and water deficit using the two types of sensors. The results show that the average relative change rate of the stability test for flexible pressure sensors was 0.109%; There was negligible output variation caused by bending; The determination coefficient of calibration was greater than 0.99, and the most suitable working range was 2-100 kPa; The determination coefficient of the measurements between the two types of sensors was 0.955 1, compared with the linear micro displacement sensors. The greenhouse experiments show that the determination coefficients of the measurements using the two types of sensors were 0.767 2 and 0.851 9, respectively, for full irrigation and deficient irrigation. The flexible and wearable pressure sensor can be expected to monitor the variations in the expansion and contraction of tomato stems. The water deficit stress of the tomato can also be diagnosed as well. The pressure sensor achieved better performance during the experiment, such as the low cost and easy installation, compared with the linear micro-displacement sensors.

A root is one of the most important vegetative organs of plants. It is of great significance to explore the growth status of root under different water stress in modern agriculture. However, the existing observation of plant root are cumbersome, laborious and time-consuming. It is the high demand to meet the requirements of precision irrigation and water-saving under field conditions. In this study, a pipeline robot system was developed for the synchronous, in-situ and the high-throughput monitoring of plant root and root soil water using STM32. The system consisted of pipeline robot, data base station and PVC transparent pipeline. The pipeline robot system was embedded in the soil, where the macro camera and soil moisture sensor were carried by the robot. The root images were captured to obtain the soil moisture data while the robot cruising. Meanwhile, the robot shared the functions of autonomous timing cruise, wireless communication using command data and active return after encountering obstacles. The distorted correction, registered on plane, identified and segmented images were obtained for the parameters of the root area, length and density of the plants in the direction of the pipeline. The results of laboratory test show that: 1) The pipeline robot was clearly captured the root images, indicating the excellent performance on distortion correction and plane registration. The true length and area corresponding to a single pixel were 44 μm and 0.002 mm², respectively, whereas, the shooting range of a single root image was 14.17×10.60 mm. The characteristic information of root was obtained using MATLAB, where the images taken by the automatic cruise of pipeline robot. The image processing operations included the image distortion correction, image preprocessing, root region recognition and segmentation, and root feature extraction. Compared with the root characteristic parameters measured by the excavation , the relative errors of the root image processing program were 12.29%, 3.40% and 12.50%, respectively; 2) There was an excellent linear relationship between the output voltage of the soil moisture sensor that carried by the pipeline robot and the soil volumetric moisture content, where the coefficient of determination was 0.990; 3) The pipeline robot presented a high accuracy of autonomous cruise positioning, with a mean relative error of 1.47%. The field experiment show that: 1) The pipeline robot system was captured the plant root images with high-throughput in the field environment. Root growth dynamics was obtained to further extract the parameters of root length, area, average diameter and density from the images; 2) The pipeline robot was accurately monitor the soil moisture in the root zone, where the mean relative error of the measured was 2.23%, compared with the drying measurement; 3) Once the system was initially fully charged, the pipeline robot system operated independently for no less than 7 days, where the maximum cruise monitoring distance was about 48m. The pipeline robot system can be expected to realize in-situ and high-throughput measurement of plant root and soil moisture in the root zone under field environment. The growth of root can be extracted after image recognition and segmentation. The finding can also provide the technical support to monitor the growth status of root in water-saving irrigation.