Non-destructive in-season grain yield (GY) prediction would strongly facilitate the selection process in plant breeding but remains challenging for phenologically and morphologically diverse germplasm, notably under high-yielding conditions. In recent years, the application of drones (UAV) for spectral sensing has been established, but data acquisition and data processing have to be further improved with respect to efficiency and reliability. Therefore, this study evaluates the selection of measurement dates, sensors, and spectral parameters, as well as machine learning algorithms. Multispectral and RGB data were collected during all major growth stages in winter wheat trials and tested for GY prediction using six machine-learning algorithms. Trials were conducted in 2020 and 2021 in two locations in the southeast and eastern areas of Germany. In most cases, the milk ripeness stage was the most reliable growth stage for GY prediction from individual measurement dates, but the maximum prediction accuracies differed substantially between drought-affected trials in 2020 (R² = 0.81 and R² = 0.68 in both locations, respectively), and the wetter, pathogen-affected conditions in 2021 (R² = 0.30 and R² = 0.29). The combination of data from multiple dates improved the prediction (maximum R² = 0.85, 0.81, 0.61, and 0.44 in the four-year*location combinations, respectively). Among the spectral parameters under investigation, the best RGB-based indices achieved similar predictions as the best multispectral indices, while the differences between algorithms were comparably small. However, support vector machine, together with random forest and gradient boosting machine, performed better than partial least squares, ridge, and multiple linear regression. The results indicate useful GY predictions in sparser canopies, whereas further improvements are required in dense canopies with counteracting effects of pathogens. Efforts for multiple measurements were more rewarding than enhanced spectral information (multispectral versus RGB).

Remote-sensing data has become essential for site-specific farming methods. It is also a powerful tool for monitoring the agroecosystem services offered by integrating cover crops (CC) into crop rotations. This study presents a method to determine the canopy height (CH), defined as the average height of the crop stand surface, including tops and gaps, of heterogeneous and multi-species CC using commercial unmanned aerial vehicles (UAVs). Images captured with red–green–blue cameras mounted on UAVs in two missions varying in ground sample distances were used as input for generating three-dimensional point clouds using the structure-from-motion approach. These point clouds were then compared to manual ground measurements. The results showed that the agreement between the methods was closest when CC presented dense and smooth canopies. However, stands with rough canopies or gaps showed substantial differences between the UAV method and ground measurements. We conclude that the UAV method is substantially more precise and accurate in determining CH than measurements taken with a ruler since the UAV introduces additional dimensions with greatly increased resolution. CH can be a reliable indicator of biomass yield, but no differences between the investigated methods were found, probably due to allometric variations of different CC species. We propose the presented UAV method as a promising tool to include site-specific information on CC in crop production strategies.

This study assessed the potential of multispectral (MS) and RGB imagery acquired by unmanned aerial vehicles (UAVs) for predicting wheat grain yield (GY) in field trials. We investigated the influence of single-date measurements and optimal multi-date strategies across diverse locations and years in southeast and eastern Germany. UAV-based MS and RGB data were collected throughout the growing season, and machine learning models were developed using vegetation indices to predict GY. Results revealed that the accuracy of single-date predictions varied, with later stages, especially grain-filling, demonstrating higher performance. Combining data from multiple dates improved GY prediction, and especially the inclusion of June acquisitions (likewise around grain-filling phase) consistently enhances the prediction accuracy of multi-date models. However, the optimal combination of dates was found to be year- and location-specific. Overall, MS data slightly outperformed RGB data, although RGB showed advantages in specific instances, particularly during later growth stages. This study provides valuable insights for optimizing UAV-based phenotyping in wheat breeding trials, highlighting the importance of strategic measurement timing and sensor selection for accurate GY prediction.

This paper presents a method to develop an intelligent master-slave system between agricultural vehicles, which will enable a semi-autonomous agricultural vehicle (slave) to follow a leading tractor (master) with a given lateral and longitudinal offset. In our study not only the follow-up motions but also the site-specific control of the apparatus such as rear and front power lift was considered. In the first part of this paper the recent research works in the area autonomous farming were discussed and the restrictions of these research works were illustrated. In the second part an approach to construct a master-slave system between two agricultural vehicles was demonstrated. In the next part the mathematic modeling of this master-slave system and the simulation results about the control algorithm were demonstrated. Afterwards the result of a real field test was presented and the safety considerations about such an intelligent vehicle system were made.

Grain yield (GY) prediction based on non-destructive UAV-based spectral sensing could make screening of large field trials more efficient and objective. However, the transfer of models remains challenging, and is affected by location, year-dependent weather conditions and measurement dates. Therefore, this study evaluates GY modelling across years and locations, considering the effect of measurement dates within years. Based on a previous study, we used a normalized difference red edge (NDRE1) index with PLS (partial least squares) regression, trained and tested with the data of individual dates and date combinations, respectively. While strong differences in model performance were observed between test datasets, i.e., different trials, as well as between measurement dates, the effect of the train datasets was comparably small. Generally, within-trials models achieved better predictions (max. R² = 0.27–0.81), but R²-values for the best across-trials models were lower only by 0.03–0.13. Within train and test datasets, measurement dates had a strong influence on model performance. While measurements during flowering and early milk ripeness were confirmed for within- and across-trials models, later dates were less useful for across-trials models. For most test sets, multi-date models revealed to improve predictions compared to individual-date models.

In this paper, we demonstrate the capabilities of Pleiades-1a imagery for very high resolution (VHR) crop yield estimation by utilizing the predictor variables from the horizontal-spectral information, through Normalized Difference Vegetation Indices (NDVI), and the vertical-volumetric crop characteristics, through the derivation of Crop Canopy Models (CCMs), from the stereo imaging capacity of the satellite. CCMs captured by Unmanned Aerial Vehicles are widely used in precision farming applications, but they are not suitable for the mapping of large or inaccessible areas. We further explore the spatiotemporal relationship of the CCMs and the NDVI for five observation dates during the growing season for eight selected crop fields in Germany with harvester-measured ground truth crop yield. Moreover, we explore different CCM normalization methods, as well as linear and non-linear regression algorithms, for the crop yield estimation. Overall, using the Extremely Randomized Trees regression, the combination of CCMs and NDVI achieves an R² coefficient of determination of 0.92.