Antti Hietanen: Safe and autonomous robots using computer vision
There is a long tradition in using robots in heavy industry, but the future challenges are more in little workshops with small batch sizes and tailored products. To make this possible, the robots must be affordable, easy to reprogram, autonomous and essentially safe to use. Msc (Tech) Antti Hietanen introduces novel technologies for future robotics in his PhD thesis. His robots use depth sensors as eyes to pick and manipulate objects and to safely work in the same workspace with humans.
In his PhD thesis Antti Hietanen introduces novel methods for robots to detect objects and their pose for robotic grasping and to avoid hitting humans in the same shared workspace. The methods have been published in some of the top journals and conferences of robotics and automation and they already have received notable attention. For communicating with humans the robots use augmented reality technologies. The methods are now under rigorous testing in the heavy industrial automation lab lead by Professor Minna Lanz.
“We test these methods using wrestling dummies – that’s because those heavy robots can accidentally crush a human without even noticing it,” Professor Lanz explains.
“During my PhD studies I learned a lot about robotics and, in particular, I learned to combine robotics with my background in information technology. Today computer vision and machine learning (artificial intelligence) play an essential role robotics,” tells Antti Hietanen from Sensible 4 company in Espoo, Finland, where he develops similar technologies for autonomous vehicles. He recommends to combine information technology, automation technology and robotics in engineering studies.
“Yeah, that must be the magic combination for the forthcoming years and needs of industry and society!” he continues.
MSc (Technology) Antti Hietanen works in Sensible 4 company which is one of the leading robotics developers in Finland and where Hietanen continues his work on robotics.
The doctoral dissertation of MSc (Tech) Antti Hietanen in the field of Information Technology titled Computer Vision for Robotics: Feature Matching, Pose Estimation and Safe Human-Robot Collaboration will be publicly examined in the Faculty of Information Technology and Computing Sciences at Tampere University at 15 January 2021 noon in public online defense. The Opponents will be Professor Patric Jensfelt from KTH Royal Institute of Technology and Professor Juha Röning from University of Oulu. The Custos will be Professor Joni Kämäräinen.
The dissertation is available online at the http://urn.fi/URN:ISBN:978-952-03-1840-6
Photo: Antti Hietanen’s personal album