The burgeoning demand for sustainable agricultural practices has driven the exploration of innovative technologies to optimize crop health and yield. Among these, optical sensors such as RGB, LiDAR, and depth cameras stand out as a cornerstone for the next revolution in precision agriculture, offering novel pathways for crop reconstruction and weed management through 3D reconstruction and 2D advanced classification techniques. This seminar delves into cutting-edge advancements in optical sensing technologies and their transformative potential in agriculture, focusing on the dual capabilities of 3D spatial reconstruction and precise weed detection and classification. The presentation will commence with an overview of optical sensing principles and their application in capturing high-resolution, three-dimensional representations of agricultural fields. It will highlight how these 3D models provide a detailed understanding of crop and weed distribution, facilitating targeted interventions. The seminar will further explore the integration of optical sensors with machine learning algorithms, illustrating how this synergy enhances the accuracy of weed identification, differentiating between crop plants and a variety of weed species with remarkable precision. Case studies demonstrating the successful deployment of optical sensors in diverse agricultural settings will be discussed, showcasing significant reductions in herbicide usage, improved crop yields, and enhanced environmental sustainability. The seminar will further explore the integration of optical sensors with machine learning algorithms, illustrating how this synergy enhances the accuracy of weed identification, differentiating between crop plants and a variety of weed species with remarkable precision. Biography: Dr. Hugo Moreno Párrizas is a postdoc researcher in the Centre for Automation and Robotics (CAR) belonging to the CSIC (Spanish Research Council). He undergraduates in his first Diploma at the School of Surveying, Geodesy, and Cartography Engineering (UPM). He worked in large civil engineering projects as a chief surveyor. Acquiring deep expertise in Geodesy, GPS Technologies, Gyro–theodolite, Guidance systems for tunneling machines, and heavy machinery on roads), tunnels, roads, and housing projects. Later on, he postgraduate with a Master of Science in Geological and Geotechnical Engineering, acquiring deep insight into soul behavior and knowledge of finite elements in relation to soil modeling and predicting behavior. He took a second postgraduate completing a Master of Science in Agro-Engineering at the Upper Technical School of Agricultural Engineers. Later on, he received an international Ph.D. at the CAR-CSIC. His expertise focuses on integrating sensor technologies (such as LiDAR, depth cameras, and RGB cameras) and geospatial technologies with machine learning algorithms (including the subfield of deep learning) for weed management. He completed his doctoral thesis at CAR-CSIC and defended it at ETSIAAB (UPM), continuing his professional career at CAR-CSIC. The results of his research have also been presented at conferences from 2013 to the present at SEMh (Weed Science Society of Spain), ECPA, ISPA, and EWRS. His research line began in 2012 with developing detection and classification systems for weeds using LiDAR systems. In the last five years, he has participated in the European project DACWEED (Detection and ACtuation system for WEED management) funded by EIT Food (grant agreement nº 20140), where he developed a prototype for weed classification using artificial intelligence techniques (deep learning) connected to a commercial sprayer. Currently, his research line follows the same working scheme by implementing deep learning techniques (neural networks) through SWEET (Sustainable weed management by agroecological and technical approaches/grant agreement nº TED2021-130031B-I00). In this latest project, new developments and improvements are being implemented for mechanical weed control both retrospectively with a prescription map and in real-time in cereal crops (in row crops such as corn and tomatoes) to regulate mechanical control and herbicide applications specifically.  These non-destructive techniques based on deep learning aim to ultimately reduce the dosage of pesticides, leading to savings for farmers and a more environmentally sustainable model by applying chemical control only where necessary and specific to weed species. He has also been a guest lecturer at universities such as the Polytechnic University of Madrid (ETSIAAB) in the Expert Course in Precision Agriculture Tools and the Master in Precision Agriculture and the Southwest University of China (Summer Module in Precision Agriculture). He has also supervised professional development as the former CEO and co-founder of Kids Engineers Ltd., a start-up dedicated to STEAM education, through educational cooperation agreements between Spanish universities (UPM, UCM, ESNE, UVA, and ESNE) and the University of Texas at Dallas, overseeing more than 40 undergraduate and master's students.