Abstract: The society as whole has to operate following all three pillars of sustainability; environmental protection, social equity, and economic viability, for a better future. For industry, this means use of lower emission technologies, improved safety for employers, and financial savings via scalability and low-cost equipment based operations. Automation of machinery is a critical step towards sustainable operations. Autonomous machines must be able to locate themselves at the industrial sites, which is not a simple task indoors. They also need to be able to take a hold of objects with precision and lift them up safely. At the same time, the machines have to observe their surroundings to avoid accidents. Computer vision, especially via using low-cost monocular cameras, could enable systems fulfilling these requirements sustainably.

In this presentation, we will discuss our research developing self-supervised deep learning based computer vision methods for localizing the crane in challenging industrial environments and our plans to extend the research for semantic 3D detection of the objects in the premises.  

Speakers:  L. Ruotsalainen is an associate professor at the Department of Computer Science, University of Helsinki. She leads a research group named Spatiotemporal data analysis for sustainability science (SDA) which performs research on estimation and ML methods using spatiotemporal data. She has a long research career in the navigation field. She is a member of the steering group of the Finnish Center for AI (FCAI) and leads a FCAI Highlight area called Sustainable AI. She is also a professor of the Helsinki Institute of sustainability Science (HELSUS), which is a cross-faculty research unit in sustainability science within the University of Helsinki.

Affiliation: University of Helsinki

Place of Seminar:  Zoom