Page Not Found
Page not found. Your pixels are in another canvas.
Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
About me
About me
Page not in menu
This is a page not in th emain menu
Posts
Future Blog Post
Published:
This post will show up by default. To disable scheduling of future posts, edit config.yml and set future: false.
Blog Post number 4
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 3
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 2
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 1
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
portfolio
Portfolio item number 1
Short description of portfolio item number 1
Portfolio item number 2
Short description of portfolio item number 2
publications
A point and command interface for grasping unknown objects with robotic manipulators
Published in Politecnico di Milano, 2017
This is my masters thesis! It uses a graph-based RGBD image segmenter to implement a point-and-grasp interface for a ABB Yumi robot.
Recommended citation: Correia Marques, Joao Marcos. (2017). "A point and command interface for grasping unknown objects with robotic manipulators." Politecnico di Milano. https://www.politesi.polimi.it/bitstream/10589/137453/5/2017_12_Correia_Marques.pdf
Heuristic active learning for the prediction of epileptic seizures using single EEG channel
Published in 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2018
This paper investigates heuristics to enable active learning on EEG data to predict epileptic seizures
Recommended citation: Joao, M. C. Marques, et al. "Heuristic active learning for the prediction of epileptic seizures using single EEG channel." 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). IEEE, 2018. https://ieeexplore.ieee.org/abstract/document/8621506
Automatic summarization of technical documents in the oil and gas industry
Published in 8th Brazilian Conference on Intelligent Systems (BRACIS), 2019
This paper is about evaluating extractive summarization techniques to generate abstracts for papers in the field of oil-and-gas.
Recommended citation: Correia Marques, Joao Marcos, Cozman, Fabio G., Ferreira Dos Santos, Ismael H.. (2019). "Automatic summarization of technical documents in the oil and gas industry." 8th Brazilian Conference on Intelligent Systems (BRACIS). http://sites.poli.usp.br/p/fabio.cozman/Publications/Article/marques-cozman-santos-bracis2019.pdf
Optimized Coverage Planning for UV Surface Disinfection
Published in ICRA, 2021
This paper introduces an algorithm to compute the optimal (shortest) path for a UV light across an environment while guaranteeing that all surfaces get the necessary dose of UV radiation for disinfection - and evaluates different robot designs for this task.
Recommended citation: Marques, Joao Marcos Correia, et al. "Optimized coverage planning for UV surface disinfection." 2021 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2021. https://arxiv.org/abs/2103.14137
Commodity Telepresence with the AvaTRINA Nursebot in the ANA Avatar XPRIZE Semifinals
Published in RSS 2022 Workshop on” Towards Robot Avatars: Perspectives on the ANA Avatar XPRIZE Competition, 2022
In this short paper, we discuss the main features of our semifinals entry to the ANA Avatar XPrize challenge in Miami, Florida, which placed 4th in the semifinals. A more detailed paper with our finals submission will follow soon.
Recommended citation: Marques, Joao MC, et al. "Commodity Telepresence with the AvaTRINA Nursebot in the ANA Avatar XPRIZE Semifinals." RSS 2022 Workshop on” Towards Robot Avatars: Perspectives on the ANA Avatar XPRIZE Competition. 2022. https://motion.cs.illinois.edu/papers/RSS2022Workshop-Marques-TelepresenceXPRIZE.pdf
Real-time Semantic 3D Reconstruction for High-Touch Surface Recognition for Robotic Disinfection
Published in IROS, 2022
This paper presentes a real-time semantic 3D reconstruction system for reconstructing and classifying environments in high-touch or low touch - and extends my previous disinfection planner to include this semantic information, showing that incorporating this information during planning improves disinfection performance by up to 20%
Recommended citation: Qiu, Ri-Zhao, Sun Yixiao, Correia Marques, Joao Marcos, Hauser, Kris. (2022). "Real-time Semantic 3D Reconstruction for High-Touch Surface Recognition for Robotic Disinfection" IROS 2022. https://ieeexplore.ieee.org/abstract/document/9981300
Pan-Tilt-Roll Televisualization with Adjustable Baseline Stereo
Published in 2nd Workshop Toward Robot Avatars, IEEE International Conference on Robotics and Automation (ICRA), 2023
Depth perception plays an important role in teleoperated task completion. Most previous works on teleoperated depth perception with Head Mounted Displays had focused their attention on matching the user’s Intra-Pupillary Distance (IPD) on the HMD’s displays, but used fixed stereo cameras on the robot side to capture the environment. In this workahop paper, we developed a custom adjustable baseline stereo rig for use in teleoperated robotics and show, through a small user study (n=16), that matching the robot’s stereo baseline with the user’s IPD has a small, but significant effect in task completion time for depth-critical operations. This suggests that future Avatar design could benefit from robot-side IPD adjustment devices, especially for fine manipulation tasks. We also observed that scaling the number of Degrees of Freedom on an avatar’s neck showed diminishing returns after 2 DoFs, hinting that commercial pan-tilt units might be sufficient for most avatar applications, though more research is needed with a more diverse task set.
Recommended citation: Naughton, P., Nam, J. S., Marques, J. M. C., Peng, J.C., Zhu, Y., Kong, Q., Hauser, K. (2023). "Pan-Tilt-Roll Televisualization with Adjustable Baseline Stereo". In 2nd Workshop Toward Robot Avatars, IEEE International Conference on Robotics and Automation ICRA https://www.ais.uni-bonn.de/ICRA2023AvatarWS/contributions/ICRA_2023_Avatar_WS_Naughton.pdf
Commodity Telepresence with Team AVATRINA’s Nursebot in the ANA Avatar XPRIZE Finals
Published in 2nd Workshop Toward Robot Avatars, IEEE International Conference on Robotics and Automation (ICRA), 2023
This workshop paper presents the techniques and technologies we employed in our successful entry on the ANA Avatar XPrize finals, in 2022 in Los Angeles. Our team, AvaTRINA, placed 4th overall, being the only team to complete the full obstacle course without the use of any bulky Load Force Feedback devices on the operator`s arms. We believe augmented reality and assistive teleoperation modes are one viable path towards scalable and equitable adoption of avatar technologies.
Recommended citation: Marques, J.M.C., Peng, J.C., Naughton, P., Zhu, Y., Nam, J. S., Hauser, K. (2023) "Commodity Telepresence with Team AVATRINA’s Nursebot in the ANA Avatar XPRIZE Finals". In 2nd Workshop Toward Robot Avatars, IEEE International Conference on Robotics and Automation ICRA https://www.ais.uni-bonn.de/ICRA2023AvatarWS/contributions/ICRA_2023_Avatar_WS_Marques.pdf
Immersive Commodity Telepresence with the AVATRINA Robot Avatar
Published in International Journal of Social Robotics, 2024
Immersive robotic avatars have the potential to aid and replace humans in a variety of applications such as telemedicine and search-and-rescue operations, reducing the need for travel and the risk to people working in dangerous environments. Many challenges, such as kinematic differences between people and robots, reduced perceptual feedback, and communication latency, currently limit how well robot avatars can achieve full immersion. This paper presents AVATRINA, a teleoperated robot designed to address some of these concerns and maximize the operator’s capabilities while using a commodity light-weight human–machine interface. Team AVATRINA took 4th place at the recent $10 million ANA Avatar XPRIZE competition, which required contestants to design avatar systems that could be controlled by novice operators to complete various manipulation, navigation, and social interaction tasks. This paper details the components of AVATRINA and the design process that contributed to our success at the competition. We highlight a novel study on one of these components, namely the effects of baseline-interpupillary distance matching and head mobility for immersive stereo vision and hand-eye coordination.
Recommended citation: Marques, J.M.C.*, Naughton, P.*, Peng, J.-C.*, Zhu, Y.*, Nam, J. S., Kong, Q., Zhang, X., Penmetcha, A., Ji, R., Fu, N., Ravibaskar, V., Yan, R., Malhotra, N., and Hauser, K., (2024). "Immersive Commodity Telepresence with the TRINA Robot Avatar ". International Journal of Social Robotics https://link.springer.com/article/10.1007/s12369-023-01090-1
On the Overconfidence Problem in Semantic 3D Mapping
Published in International Conference on Robotics and Automation (ICRA) 2024, 2024
Semantic 3D mapping, the process of fusing depth and image segmentation information between multiple views to build 3D maps annotated with object classes in real-time, is a recent topic of interest. This paper highlights the fusion overconfidence problem, in which conventional mapping methods assign high confidence to the entire map even when they are incorrect, leading to miscalibrated outputs. Several methods to improve uncertainty calibration at different stages in the fusion pipeline are presented and compared on the ScanNet dataset. We show that the most widely used Bayesian fusion strategy is among the worst calibrated, and propose a learned pipeline that combines fusion and calibration, GLFS, which achieves simultaneously higher accuracy and 3D map calibration while retaining real-time capability and adding only 525 learned parameters to the pipeline. We further illustrate the importance of map calibration on a downstream task by showing that incorporating proper semantic fusion to an indoor object search agent improves its success rates.
Recommended citation: Marques, J.M.C., Zhai, A., Wang, S. and Hauser, K., (2024). On the Overconfidence Problem in Semantic 3D Mapping”, 2024 IEEE International Conference on Robotics and Automation (ICRA) https://arxiv.org/abs/2311.10018
talks
Talk 1 on Relevant Topic in Your Field
Published:
This is a description of your talk, which is a markdown files that can be all markdown-ified like any other post. Yay markdown!
Conference Proceeding talk 3 on Relevant Topic in Your Field
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.