VII WORKSHOP IN AUTOMATION AND ROBOTICS
September 07, 2023
09.00-13.00 hrs.[Chile Continental Time]
Research line in Automation and Robotics, Departamento de Ingeniería de Sistemas y Computación, Universidad Católica del Norte, Antofagasta, Chile
Date and Time:
Thursday, september 07th, 2023. 09 - 13 hrs. [Chile Continental Time]
The Workshop in Automation and Robotics is aimed at building a space in which students, academics, professionals, and researchers disseminate the up-today progress of the scientific and professional activity in disciplines of Automation, Robotics, and other related- areas applied to the Industry and Service fields.
Advances in Self-Tuning Optimal Control of Self-Balancing Robotic Systems
Dr. Omer Saleem
Assessment of Multispectral Vegetation Features for Digital Terrain Modeling in Forested Regions
Dr. Tito Arévalo
11.00 - 11.30 hrs. COFFEE BREAK
Agricultural and Forestry Robotics: AI, Applications and Challenges
Dr. Francisco Javier Yandún Narvaez
Advances and challenges in developing robotic solutions for soft-fruit harvestings
Dr. César Leonardo Guevara Gordillo
Invited Talk #1
9.00-10.00am (Chile Continental Time).
Speech Title: Advances in Self-Tuning Optimal Control of Self-Balancing Robotic Systems
Speaker: Omer Saleem
Ph.D. in Electrical Engineering of the University of Engineering and Technology Lahore (UET Lahore), Pakistan.
Dr. Omer Saleem received his B.S. degree in Electrical Engineering and M.S. degree in Electrical Engineering (with a specialization in control systems) from the University of Engineering and Technology (UET), Lahore, Pakistan, in 2010 and 2014, respectively. He completed my Ph.D. in Electrical Engineering in Sep. 2021 from UET, Lahore, with a 4.00/4.00 CGPA. His area of specialization in Ph.D. was adaptive control systems with the main focus on the formulation of hierarchical self-tuning controllers for under-actuated electromechanical systems. He formulated an innovative indirect hierarchical self-adaptive control procedure that methodically transforms a generic LQR into an online self-tuning state-feedback controller by augmenting it with pre-configured adaptation laws. This adaptive control was applied to robustify the stabilization control of self-balancing robotic systems. Presently, he is serving as an Assistant Professor and MS Program Coordinator at the Department of Electrical Engineering, FAST - National University of Computer and Emerging Sciences (NUCES), Lahore, Pakistan. At FAST-NUCES. He instructs undergraduate and graduate courses - namely, feedback control systems, power electronics, industrial process control, robot control systems, intelligent systems, and electronic circuit design. He has supervised numerous senior-year-design projects and MS Theses in the field of robotics and control. He is currently also co-supervising a Ph.D. student in the area of fractional control. He has published more than 30 research papers in Thomson JCR-indexed impact-factor journals. He is a Senior Member of IEEE. His current research interests include the design of adaptive and self-tuning state-feedback control systems for mechatronic systems and power electronic converters using nonlinear sliding surfaces and expert systems. At FAST-NUCES, he also serves as the Head of the "Engineering Cybernetics Research Group" (ECRG). The group undertakes research in the area of self-tuning and adaptive control law formulation for under-actuated self-balancing robotic mechanisms using online expert intelligent systems.
Invited Talk #2
10.00-11.00 am (Chile Continental Time).
Speech Title: Assessment of Multispectral Vegetation Features for Digital Terrain Modeling in Forested Regions
Speaker: Tito Arévalo
Dr. en Ingeniería Electrónica de la Universidad Técnica Federico Santa María, Valparaíso, Chile. Académico Pontificia Universidad Católica de Chile, Santiago, Chile.
Abstract. Bare-earth extraction in forested regions has been considered challenging because of the lack of ground point information. In these regions, vision systems cannot capture any information about ground points under the canopy. Thus, the challenge of generating a digital terrain model by cameras increases. Nevertheless, one might alleviate ground filtering using vegetation’s features (e.g., chlorophyll). In this regard, this article evaluated two machine-learning approaches [i.e., conditional random field (CRF), artificial neural network (ANN)] for generating digital terrain models when biophysical or biochemical features of vegetation are given. Terrain models were generated from multispectral image-based point clouds. A fivefold cross-validation methodology evaluated the CRF and ANN. The point clouds were retrieved from two study areas at different illumination and flight altitudes. Vegetation features were computed as vegetation indices from the multispectral point clouds. Results suggested that by using these indices, the classification of ground points could be enhanced. In particular, the vegetation indices that yielded the best outcomes were normalized difference vegetation index, green NDVI, and modified chlorophyll absorption reflectance index. Moreover, it was shown that CRF generates elevation models more smoothly than a triangular irregular network method. Thus, a CRF could be promising for classifying ground points in forested regions using geometric and vegetation features from photogrammetric point cloud.
Dr. Tito Arévalo Ramirez was born in Loja Ecuador in 1992. He is currently assistant professor at Pontificia Universidad Católica de Chile. He received the Ph.D. degree in Electronic Engineering at Universidad Técnica Federico Santa María. His research interests include machine learning, remote sensing, field robotics, agriculture, and forestry.
Invited Talk #3
11.30-12.15 am (Chilean Continental Time).
Speech Title: Agricultural and Forestry Robotics: AI, Applications and Challenges
Speaker: Dr. Francisco Javier Yandún Narvaez
Dr., en Ingeniería Electrónica de la Universidad Técnica Federico Santa María, Valparaíso, Chile. Project Scientist at the Kantor Laboratory, Carnegie Mellon University, Robotics Institute
Abstract. Innovations in Robotics have the potential to play a pivotal role in agriculture and forestry by assisting or even replacing humans in challenging, repetitive or dangerous tasks. We will discuss an overview of research in robotics with application to agriculture and forestry, spanning topics related to manipulation, object detection, mapping and localization, and semantic segmentation. In addition, we will discuss the details of implementing a simultaneous, localization and mapping (SLAM) and semantic segmentation methodology to aid in forest fire mitigation efforts.
Dr. Francisco Yandun earned his PhD in Electronic Engineering from the Universidad Técnica Federico Santa María in Valparaíso, Chile, before taking up a Postdoctoral Fellowship at the Robotics Institute at Carnegie Mellon University. He now holds the position of Project Scientist at the Kantor Lab, where he focuses on automating crop pruning with robotics technologies and advancing algorithms for modelling, localization and mapping. In addition to publishing his research in numerous scientific journals, Dr Yandun has also reviewed and edited the work of other scientists for publication.
Invited Talk #4
12.15-13.00 pm (Chilean Continental Time).
Speech Title: Advances and challenges in developing robotic solutions for soft-fruit harvestings
Speaker: Dr. César Leonardo Guevara Gordillo
Dr., in Electronics Engineering at the Universidad Técnica Federico Santa María, Valparaíso, Chile. Lecturer in Agri-Robotics at the Lincoln Institute for Agri-food Technology (LIAT), and member of the Lincoln Centre for Autonomous Systems (L-CAS).
Abstract. The introduction of mobile robots for the automation of soft-fruit harvesting has great potential to solve current problems related to labor shortages in countries like the UK. Although there are some companies that already offer solutions for the collection and transport of fruits using autonomous robotic platforms. There are still several technological and societal challenges that need to be resolved so that the agricultural robotics industry can fully develop and be accepted as a profitable solution. This talk will focus on robotic solutions that are currently being tested in both experimental agricultural settings, as well as various solutions that are being implemented commercially. We will focus on the current advances and challenges in the use of mobile robots for the harvest and transport of fruits, as well as the key factors that will define the future of horticulture.
Dr. Leonardo Guevara did his undergraduate studies at "Escuela Politécnica Nacional" (Ecuador), obtaining his Engineering degree in Electronics and Control in 2016. He continued with his postgraduate studies at "Universidad Técnica Federico Santa María" (Chile), obtaining his Ph.D. in Electronic Engineering in 2021. During the Ph.D. Leonardo had the opportunity to do research visits at the University of Coimbra (Portugal) and the Poznan University of Technology (Poland) where he worked on research topics related to the autonomous navigation of articulated vehicles that are intended to be used for transporting fruit in robot-assisted harvesting operations. Maintaining this research line, he joined the University of Lincoln (UK) as a Postdoctoral Research Associate in 2021. During his postdoc, Leonardo worked on the MeSAPro project whose aim was to ensure the autonomy of robots that interact with human coworkers in agricultural operations. His is currently working as a Lecturer in Agri-robotics at the Lincoln Institute for Agri-food Technology (LIAT) and also as a member of the Lincoln Centre for Autonomous Systems (L-CAS). His current research interests focus on robotics applied to agriculture, robot motion control/planning, and human-robot interaction.