Guillermo Trinidad Barnech

Agricultural production is a cornerstone of Uruguay's economy, contributing significantly to its GDP with an impact of approximately 11 points. This sector is characterized by its high technological demand and a declining labor force. Traditionally, the agricultural workforce has not required extensive technical skills, but the rapid advancement of technology necessitates a shift towards more technologically adept labor. Thus, the adoption of advanced technologies and the retraining of the workforce are imperative. Labor retraining within the agricultural domain is essential to fortify the sector's competitiveness, sustainability, and resilience amidst contemporary challenges. Precision agriculture advocates integrating cutting-edge technologies to optimize crop management and agricultural resource utilization. Implementing autonomous robotics has the potential to mitigate labor requirements, foster workforce technological education, and propel advancements toward precision agriculture. Founded in 2013, the MINA group at the Faculty of Engineering of the University of the Republic (Uruguay) has been actively engaged in robotics projects tailored for agricultural applications. Initially focusing on orchards of pome fruits from 2013 to 2023, the group undertook tasks such as harvest support and estimation of harvest quantity and quality. Subsequently, efforts have been directed towards pest control measures targeting pests such as birds and ants, and weed management. This paper delineates the intricacies of these distinct projects, elucidating the technologies employed and developed, outlining achieved results to date, and envisaging the potential for widespread adoption of this technology at a feasible cost.

We introduce both conceptual and empirical findings arising from the amalgamation of a robotics cognitive architecture with an embedded physics simulator, aligning with the principles outlined in the intuitive physics literature. The employed robotic cognitive architecture, named CORTEX, leverages a highly efficient distributed working memory known as deep state representation. This working memory inherently encompasses a fundamental ontology, state persistency, geometric and logical relationships among elements, and tools for reading, updating, and reasoning about its contents. Our primary objective is to investigate the hypothesis that the integration of a physics simulator into the architecture streamlines the implementation of various functionalities that would otherwise necessitate extensive coding and debugging efforts. Furthermore, we categorize these enhanced functionalities into broad types based on the nature of the problems they address. These include addressing challenges related to occlusion, model-based perception, self-calibration, scene structural stability, and human activity interpretation. To demonstrate the outcomes of our experiments, we employ CoppeliaSim as the embedded simulator and both a Kinova Gen3 robotic arm and the Open-Manipulator-P as the real-world scenarios. Synchronization is maintained between the simulator and the stream of real events. Depending on the ongoing task, numerous queries are computed, and the results are projected into the working memory. Participating agents can then leverage this information to enhance overall performance.

The agricultural industry is being transformed, thanks to recent innovations in computer vision and deep learning. However, the lack of specific datasets collected in natural agricultural environments is, arguably, the main bottleneck for novel discoveries and benchmarking. The present work provides a novel dataset, Magro, and a framework to expand data collection. We present the first version of the Magro Dataset V1.0, consisting of nine ROS bags (and the corresponding raw data) containing data collected in apple and pear crops. Data were gathered, repeating a fixed trajectory on different days under different illumination and weather conditions. To support the evaluation of loop closure algorithms, the trajectories are designed to have loop closures, revisiting some places from different viewpoints. We use a Clearpath’s Jackal robot equipped with stereo cameras pointing to the front and left side, a 3D LIDAR, three inertial measurement units (IMU), and wheel encoders. Additionally, we provide calibrated RTK GPS data that can be used as ground truth. Our dataset is openly available, and it will be updated to have more data and variability. Finally, we tested two existing state-of-the-art algorithms for vision and point cloud-based localization and mapping on our novel dataset to validate the dataset’s usability.

MicroRNAs are small RNAs that regulate gene expression through complementary base pairing with their target mRNAs. A substantial understanding of microRNA target recognition and repression mechanisms has been reached using diverse empirical and bioinformatic approaches, primarily in vitro biochemical or cell culture perturbation settings. We sought to determine if rules of microRNA target efficacy could be inferred from extensive gene expression data of human tissues. A transcriptome-wide assessment of all the microRNA–mRNA canonical interactions’ efficacy was performed using a normalized Spearman correlation (Z-score) between the abundance of the transcripts in the PRAD-TCGA dataset tissues (RNA-seq mRNAs and small RNA-seq for microRNAs, 546 samples). Using the Z-score of correlation as a surrogate marker of microRNA target efficacy, we confirmed hallmarks of microRNAs, such as repression of their targets, the hierarchy of preference for gene regions (3′UTR > CDS > 5′UTR), and seed length (6 mer < 7 mer < 8 mer), as well as the contribution of the 3′-supplementary pairing at nucleotides 13–16 of the microRNA. Interactions mediated by 6 mer + supplementary showed similar inferred repression as 7 mer sites, suggesting that the 6 mer + supplementary sites may be relevant in vivo. However, aggregated 7 mer-A1 seeds appear more repressive than 7 mer-m8 seeds, while similar when pairing possibilities at the 3′-supplementary sites. We then examined the 3′-supplementary pairing using 39 microRNAs with Z-score-inferred repressive 3′-supplementary interactions. The approach was sensitive to the offset of the bridge between seed and 3′-supplementary pairing sites, and the pattern of offset-associated repression found supports previous findings. The 39 microRNAs with effective repressive 3′supplementary sites show low GC content at positions 13–16. Our study suggests that the transcriptome-wide analysis of microRNA–mRNA correlations may uncover hints of microRNA targeting determinants. Finally, we provide a bioinformatic tool to identify microRNA–mRNA candidate interactions based on the sequence complementarity of the seed and 3′-supplementary regions.