\n \n \n
\n
\n\n \n \n \n \n \n \n Accelerating Robot Trajectory Learning for Stochastic Tasks.\n \n \n \n \n\n\n \n Vidaković, J.; Jerbić, B.; Sekoranja, B.; Švaco, M.; and Suligoj, F.\n\n\n \n\n\n\n
IEEE Access, 8: 71993–72006. 2020.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{vidakovic_accelerating_2020,\n\ttitle = {Accelerating {Robot} {Trajectory} {Learning} for {Stochastic} {Tasks}},\n\tvolume = {8},\n\tissn = {2169-3536},\n\turl = {https://ieeexplore.ieee.org/document/9062516/},\n\tdoi = {10.1109/ACCESS.2020.2986999},\n\tabstract = {Learning from demonstration provides ways to transfer knowledge and skills from humans to robots. Models based solely on learning from demonstration often have very good generalization capabilities but are not completely accurate when adapting to new scenarios. This happens especially when learning stochastic tasks because of the correspondence problem and unmodeled physical properties of tasks. On the other hand, reinforcement learning (RL) methods such as policy search have the capability to refine an initial skill through exploration, where the learning process is often very dependent on the initialization strategy and is efficient in finding only local solutions. These two approaches are, therefore, frequently combined. In this paper, we present how the iterative learning of tasks can be accelerated by a learning from demonstration (LfD) method based on the extraction of via-points. The paper provides an evaluation of the approach on two different primitive motion tasks.},\n\tlanguage = {en},\n\turldate = {2020-06-12},\n\tjournal = {IEEE Access},\n\tauthor = {Vidaković, Josip and Jerbić, Bojan and Sekoranja, Bojan and Švaco, Marko and Suligoj, Filip},\n\tyear = {2020},\n\tpages = {71993--72006},\n}\n\n
\n
\n\n\n
\n Learning from demonstration provides ways to transfer knowledge and skills from humans to robots. Models based solely on learning from demonstration often have very good generalization capabilities but are not completely accurate when adapting to new scenarios. This happens especially when learning stochastic tasks because of the correspondence problem and unmodeled physical properties of tasks. On the other hand, reinforcement learning (RL) methods such as policy search have the capability to refine an initial skill through exploration, where the learning process is often very dependent on the initialization strategy and is efficient in finding only local solutions. These two approaches are, therefore, frequently combined. In this paper, we present how the iterative learning of tasks can be accelerated by a learning from demonstration (LfD) method based on the extraction of via-points. The paper provides an evaluation of the approach on two different primitive motion tasks.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n RONNA G4—Robotic Neuronavigation: A Novel Robotic Navigation Device for Stereotactic Neurosurgery.\n \n \n \n \n\n\n \n Jerbić, B.; Švaco, M.; Chudy, D.; Šekoranja, B.; Šuligoj, F.; Vidaković, J.; Dlaka, D.; Vitez, N.; Župančić, I.; Drobilo, L.; Turković, M.; Žgaljić, A.; Kajtazi, M.; and Stiperski, I.\n\n\n \n\n\n\n In
Handbook of Robotic and Image-Guided Surgery, pages 599–625. Elsevier, 2020.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@incollection{jerbic_ronna_2020,\n\ttitle = {{RONNA} {G4}—{Robotic} {Neuronavigation}: {A} {Novel} {Robotic} {Navigation} {Device} for {Stereotactic} {Neurosurgery}},\n\tisbn = {978-0-12-814245-5},\n\tshorttitle = {{RONNA} {G4}—{Robotic} {Neuronavigation}},\n\turl = {https://linkinghub.elsevier.com/retrieve/pii/B9780128142455000359},\n\tlanguage = {en},\n\turldate = {2020-01-03},\n\tbooktitle = {Handbook of {Robotic} and {Image}-{Guided} {Surgery}},\n\tpublisher = {Elsevier},\n\tauthor = {Jerbić, Bojan and Švaco, Marko and Chudy, Darko and Šekoranja, Bojan and Šuligoj, Filip and Vidaković, Josip and Dlaka, Domagoj and Vitez, Nikola and Župančić, Ivan and Drobilo, Luka and Turković, Marija and Žgaljić, Adrian and Kajtazi, Marin and Stiperski, Ivan},\n\tyear = {2020},\n\tdoi = {10.1016/B978-0-12-814245-5.00035-9},\n\tkeywords = {CV, RM, h2020, prijava},\n\tpages = {599--625},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Stereotactic Neuro-Navigation Phantom Designs: A Systematic Review.\n \n \n \n \n\n\n \n Švaco, M.; Stiperski, I.; Dlaka, D.; Šuligoj, F.; Jerbić, B.; Chudy, D.; and Raguž, M.\n\n\n \n\n\n\n
Frontiers in Neurorobotics, 14: 549603. October 2020.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n\n\n\n
\n
@article{svaco_stereotactic_2020,\n\ttitle = {Stereotactic {Neuro}-{Navigation} {Phantom} {Designs}: {A} {Systematic} {Review}},\n\tvolume = {14},\n\tissn = {1662-5218},\n\tshorttitle = {Stereotactic {Neuro}-{Navigation} {Phantom} {Designs}},\n\turl = {https://www.frontiersin.org/articles/10.3389/fnbot.2020.549603/full},\n\tdoi = {10.3389/fnbot.2020.549603},\n\turldate = {2020-10-31},\n\tjournal = {Frontiers in Neurorobotics},\n\tauthor = {Švaco, Marko and Stiperski, Ivan and Dlaka, Domagoj and Šuligoj, Filip and Jerbić, Bojan and Chudy, Darko and Raguž, Marina},\n\tmonth = oct,\n\tyear = {2020},\n\tkeywords = {RM},\n\tpages = {549603},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Intelligent Algorithms for Non-parametric Robot Calibration:.\n \n \n \n \n\n\n \n Turković, M.; Švaco, M.; and Jerbić, B.\n\n\n \n\n\n\n In Galambos, P.; and Madani, K., editor(s),
Proceedings of the International Conference on Robotics, Computer Vision and Intelligent Systems, pages 51–58, Budapest, Hungary, 2020. SCITEPRESS - Science and Technology Publications\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{turkovic_intelligent_2020,\n\taddress = {Budapest, Hungary},\n\ttitle = {Intelligent {Algorithms} for {Non}-parametric {Robot} {Calibration}:},\n\tisbn = {978-989-758-479-4},\n\tshorttitle = {Intelligent {Algorithms} for {Non}-parametric {Robot} {Calibration}},\n\turl = {https://www.scitepress.org/DigitalLibrary/Link.aspx?doi=10.5220/0010176900510058},\n\tdoi = {10.5220/0010176900510058},\n\turldate = {2020-11-17},\n\tbooktitle = {Proceedings of the {International} {Conference} on {Robotics}, {Computer} {Vision} and {Intelligent} {Systems}},\n\tpublisher = {SCITEPRESS - Science and Technology Publications},\n\tauthor = {Turković, Marija and Švaco, Marko and Jerbić, Bojan},\n\teditor = {Galambos, Péter and Madani, Kurosh},\n\tyear = {2020},\n\tpages = {51--58},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Increasing the Accuracy of Robotic Neurosurgical Procedures Through Robot Calibration.\n \n \n \n \n\n\n \n Drobilo, L.; Švaco, M.; and Jerbić, B.\n\n\n \n\n\n\n In
2020 43rd International Convention on Information, Communication and Electronic Technology (MIPRO), pages 1180–1188, Opatija, Croatia, September 2020. IEEE\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{drobilo_increasing_2020,\n\taddress = {Opatija, Croatia},\n\ttitle = {Increasing the {Accuracy} of {Robotic} {Neurosurgical} {Procedures} {Through} {Robot} {Calibration}},\n\tisbn = {978-953-233-099-1},\n\turl = {https://ieeexplore.ieee.org/document/9245233/},\n\tdoi = {10.23919/MIPRO48935.2020.9245233},\n\turldate = {2020-11-10},\n\tbooktitle = {2020 43rd {International} {Convention} on {Information}, {Communication} and {Electronic} {Technology} ({MIPRO})},\n\tpublisher = {IEEE},\n\tauthor = {Drobilo, L. and Švaco, Marko and Jerbić, Bojan},\n\tmonth = sep,\n\tyear = {2020},\n\tpages = {1180--1188},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Task Dependent Trajectory Learning from Multiple Demonstrations Using Movement Primitives.\n \n \n \n \n\n\n \n Vidaković, J.; Jerbić, B.; Šekoranja, B.; Švaco, M.; and Šuligoj, F.\n\n\n \n\n\n\n In Berns, K.; and Görges, D., editor(s),
Advances in Service and Industrial Robotics, volume 980, pages 275–282. Springer International Publishing, Cham, 2020.\n
\n\n
\n\n
\n\n
\n\n \n \n Paper\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@incollection{berns_task_2020,\n\taddress = {Cham},\n\ttitle = {Task {Dependent} {Trajectory} {Learning} from {Multiple} {Demonstrations} {Using} {Movement} {Primitives}},\n\tvolume = {980},\n\tisbn = {978-3-030-19647-9 978-3-030-19648-6},\n\turl = {http://link.springer.com/10.1007/978-3-030-19648-6_32},\n\tlanguage = {en},\n\turldate = {2019-11-04},\n\tbooktitle = {Advances in {Service} and {Industrial} {Robotics}},\n\tpublisher = {Springer International Publishing},\n\tauthor = {Vidaković, Josip and Jerbić, Bojan and Šekoranja, Bojan and Švaco, Marko and Šuligoj, Filip},\n\teditor = {Berns, Karsten and Görges, Daniel},\n\tyear = {2020},\n\tdoi = {10.1007/978-3-030-19648-6_32},\n\tpages = {275--282},\n}\n\n
\n
\n\n\n\n
\n\n\n\n\n\n