Invited Speakers

Prof. Miguel Torres-Torriti received the B.Sc. and M.Sc. degrees in electrical engineering from the Pontificia Universidad Católica de Chile (PUC Chile), in 1996 and 1998, respectively, and the Ph.D. degree from McGill University, Montreal, Canada, in 2003. He is a member of IEEE since 2001.

Between 2004 and 2005 he was Senior Applications Engineer with General Electric Chile in the implementation of the multivariable process controllers for the ENAP Bio Bio and Magallanes refineries. In 2005 he joined the Department of Electrical Engineering of the School of Engineering of PUC Chile, where he is currently Associate Professor, Director of the undergraduate program in Robotics Engineering, and Associate Dean of Technology. Between 2014 and 2016 he served as Associate Dean Social Responsibility. His fields of experience comprise systems modeling and control, estimation, robot dynamics, sensors and perception. His current research projects include the development of sensing, motion planning and control strategies for mobile manipulators in agriculture and mining.

Past projects have included the development of a sensor subsystems for astronomical instrumentation, data fusion of satellite imagery with wireless sensor network data for early warning of flashfloods, and the development of assistive devices for persons with motor disabilities.

Digital Sense is a Computer Vision and Machine Learning R&D studio, that helps innovative companies face highly-complex technological challenges with sophisticated AI and Image Processing solutions. In this presentation we show our work for Montes Del Plata, an important Eucalyptus cellulose pulp producer committed to sustainability. Our team developed a tractor-mountable computer vision module that detects weeds in Eucalyptus plantations in real time as herbicide is sprayed. To this end, we selected components, designed and built the image acquisition system and embedded processing module. As the machine operates, a video of the planted line is fed to the embedded computer, where image processing algorithms and a deep-learning model run in real time to detect and locate weeds in order to enable targeted spraying.