diff --git a/config/_default/params.yaml b/config/_default/params.yaml index 64da188..cfeff68 100644 --- a/config/_default/params.yaml +++ b/config/_default/params.yaml @@ -41,7 +41,7 @@ header: show_search: true highlight_active_link: false logo: - text: 'IDRA Labs' + text: 'IDRA Lab' filename: 'icon.png' # Site footer diff --git a/content/authors/gustavofuentevilla/_index.md b/content/authors/gustavofuentevilla/_index.md new file mode 100644 index 0000000..e5bcef7 --- /dev/null +++ b/content/authors/gustavofuentevilla/_index.md @@ -0,0 +1,76 @@ +--- +# Display name +title: Gustavo Fuentevilla + +# Full name (for SEO) +first_name: Jorge Gustavo +last_name: Pérez Fuentevilla + +# Username (this should match the folder name) +authors: + - gustavofuentevilla + +# Is this the primary user of the site? +superuser: false + +# Role/position +role: PhD Student + +# Organizations/Affiliations +organizations: + - name: University of Trento + url: 'https://webapps.unitn.it/du/en/Persona/PER0277918' + +# Short bio (displayed in user profile at end of posts) +bio: PhD student interested in mobile manipulation and active sensing algorithms for collaborative robots. + +interests: + - Mobile manipulators + - Motion planning + - Active sensing + - Ergodic control + - Robot manipulation + - Estimation and control for collaborative robots + +education: + courses: + - course: MSc in Electrical Engineering + institution: Center for research and advanced studies of the national polytechnic institute (CINVESTAV-IPN) + year: 2022 + - course: BSc in Mechatronics Engineering + institution: Benemérita Universidad Autónoma de Puebla (BUAP) + year: 2019 + +# Social/Academic Networking +# For available icons, see: https://docs.hugoblox.com/getting-started/page-builder/#icons +# For an email link, use "fas" icon pack, "envelope" icon, and a link in the +# form "mailto:your-email@example.com" or "#contact" for contact widget. +social: + - icon: envelope + icon_pack: fas + link: 'mailto:jg.perezfuentevilla@unitn.it' + - icon: linkedin + icon_pack: fab + link: https://www.linkedin.com/in/gustavo-fuentevilla/ + - icon: google-scholar + icon_pack: ai + link: https://scholar.google.it/citations?hl=en&user=TAOg_wwAAAAJ + - icon: github + icon_pack: fab + link: https://github.com/gustavofuentevilla +# Link to a PDF of your resume/CV from the About widget. +# To enable, copy your resume/CV to `static/files/cv.pdf` and uncomment the lines below. +# - icon: cv +# icon_pack: ai +# link: files/cv.pdf + +# Enter email to display Gravatar (if Gravatar enabled in Config) +email: '' + +# Organizational groups that you belong to (for People widget) +# Set this to `[]` or comment out if you are not using People widget. +user_groups: + - PhD Students +--- + +Gustavo is a current PhD student in the Department of Industrial Engineering. His research work and experience includes mobile manipulation control design, analysis, and coordination. Currently working in active sensing algorithms for collaborative robots, safe motion planning and ergodic control. One of his main interests and motivation is combining classical design with novel AI approaches for robotic issues. \ No newline at end of file diff --git a/content/authors/gustavofuentevilla/avatar.jpg b/content/authors/gustavofuentevilla/avatar.jpg new file mode 100644 index 0000000..64f3a61 Binary files /dev/null and b/content/authors/gustavofuentevilla/avatar.jpg differ diff --git a/content/authors/veronicacampana/_index.md b/content/authors/veronicacampana/_index.md index 14fce14..47169aa 100644 --- a/content/authors/veronicacampana/_index.md +++ b/content/authors/veronicacampana/_index.md @@ -14,7 +14,7 @@ authors: superuser: false # Role/position -role: PhD student +role: PhD Student # Organizations/Affiliations organizations: @@ -22,7 +22,7 @@ organizations: url: 'https://webapps.unitn.it/du/en/Persona/PER0281361/' # Short bio (displayed in user profile at end of posts) -bio: +bio: My research interests include motion control of robots in shared environments. interests: - Motion control diff --git a/content/event/irim2024/featured.png b/content/event/2024-10-25_irim/featured.png similarity index 100% rename from content/event/irim2024/featured.png rename to content/event/2024-10-25_irim/featured.png diff --git a/content/event/irim2024/index.md b/content/event/2024-10-25_irim/index.md similarity index 100% rename from content/event/irim2024/index.md rename to content/event/2024-10-25_irim/index.md diff --git a/content/event/2024-12-05_delprete_seminar/index.md b/content/event/2024-12-05_delprete_seminar/index.md new file mode 100644 index 0000000..dc6460c --- /dev/null +++ b/content/event/2024-12-05_delprete_seminar/index.md @@ -0,0 +1,74 @@ +--- +title: Virtual Seminar Prof. Andrea Del Prete + +event: Seminar "Globally Optimal and Safe Robot Control" +event_url: https://docs.google.com/document/d/1ZHZ0zE2q8QiXXyLHZETxacUszOmrpWBfUZXTWs3IAHs/edit?tab=t.0 + +location: Virtual +# address: +# street: 450 Serra Mall +# city: Stanford +# region: CA +# postcode: '94305' +# country: United States + +summary: First seminar of the 2024/2025 virtual seminar series of the IEEE RAS TC on Model-based Optimization for Robotics. +# abstract: We are thrilled to announce our participation in the **I-RIM 3D 2024** conference. Our team has several exciting activities planned! + +# Talk start and end times. +# End time can optionally be hidden by prefixing the line with `#`. +date: '2024-12-05' +# date_end: '2024-12-05' +all_day: false + +# Schedule page publish date (NOT talk date). +publishDate: '2024-11-19T00:00:00Z' + +authors: [andreadelprete] +tags: [] +profile: false + +# Is this a featured talk? (true/false) +featured: true + +# image: +# caption: 'Image credit: [**Unsplash**](https://unsplash.com/photos/bzdhc5b3Bxs)' +# focal_point: Right + +url_code: '' +url_pdf: '' +url_slides: '' +url_video: '' + +# Markdown Slides (optional). +# Associate this talk with Markdown slides. +# Simply enter your slide deck's filename without extension. +# E.g. `slides = "example-slides"` references `content/slides/example-slides.md`. +# Otherwise, set `slides = ""`. +slides: + +# Projects (optional). +# Associate this post with one or more of your projects. +# Simply enter your project's folder or file name without extension. +# E.g. `projects = ["internal-project"]` references `content/project/deep-learning/index.md`. +# Otherwise, set `projects = []`. +research: [] +--- + +## Upcoming Seminar + +**Organizers**: [IEEE RAS TC on Model-based Optimization for Robotics](https://www.tcoptrob.org/) + +**Speaker**: [Andrea Del Prete](/author/andrea-del-prete/) + +**Title**: Globally Optimal and Safe Robot Control + +**Abstract**: In recent years, advanced data-driven control methods are unlocking the potential of complex robotics systems, and we can expect this trend to continue at an exponential rate in the near future. However, these methods present two major shortcomings. First, their training process is excessively data hungry and strongly dependent on the exploration strategy. Second, the resulting policies cannot ensure constraint satisfaction (i.e. safety). In this talk I will discuss how we are using tools from optimal control to address these two issues. + +First, I will present "Continuous Actor Critic with Trajectory Optimization" (CACTO), a novel algorithm that combines Trajectory Optimization (TO) and Reinforcement Learning (RL). CACTO learns a control policy via TO-guided RL policy search. Our method is validated on several non-convex problems with different robotic systems. Our results show the great capabilities of CACTO in escaping local minima, while being more sample-efficient than DDPG and PPO. To address the issue of safety, a well-known tool is the control-invariant set (a.k.a. safe set). Unfortunately, for nonlinear systems, such sets can only be approximated. I will present some novel MPC schemes that guarantee safety under weaker assumptions than classic methods. Our key idea is to make the safe-set constraint move backward (i.e. recede) over the horizon. We evaluated our approaches on simulated robot manipulators, empirically demonstrating that they lead to less constraint violations, while retaining good tracking cost and computation times. + +**Date**: Thursday, December 5th 2024 + +**Time**: 09:00-10:00 AM EST + +**Link**: [here](https://columbiauniversity.zoom.us/j/91247893326?pwd=L2JWU21aQzc4cU1ZQklEb0QrWGQvdz09) \ No newline at end of file diff --git a/content/post/iros2024/index.md b/content/post/iros2024/index.md index f19b6e4..957b347 100644 --- a/content/post/iros2024/index.md +++ b/content/post/iros2024/index.md @@ -11,6 +11,6 @@ Workshop and Paper presentation at IROS2024 -During the IROS2024 conference some members of the IDRA-Lab had the opportunity to hold a workshop and present a paper. On the second day of conference [Matteo Saveriano](/author/matteo-saveriano/) and [Edoardo Lamon](/author/edoardo-lamon/) thogehter with other researchers held a workshop on **Variable Impedance Learning and Control**. During the workshop, several research avenues were explored, from exploratory behaviors to learning from human demonstration. The potential of recent AI methods, like large language models (LLMs), to emulate human behavior suggests that combining model-based and data-driven approaches could revolutionize robot interaction. Researchers also discussed the importance of enabling robots to ensure safety and reliability while maintaining explainability in their decisions. +During the IROS2024 conference some members of the IDRA-Lab had the opportunity to hold a workshop and present a paper. On the second day of conference [Matteo Saveriano](/author/matteo-saveriano/) and [Edoardo Lamon](/author/edoardo-lamon/) thogehter with other researchers held a workshop on **Variable Impedance Learning and Control** (link available [here](https://sites.google.com/view/vilc-workshop-iros24/)). During the workshop, several research avenues were explored, from exploratory behaviors to learning from human demonstration. The potential of recent AI methods, like large language models (LLMs), to emulate human behavior suggests that combining model-based and data-driven approaches could revolutionize robot interaction. Researchers also discussed the importance of enabling robots to ensure safety and reliability while maintaining explainability in their decisions. On Wednesday our PhD student [Luca Beber](/author/luca-beber/) presented his paper titled **Towards Robotised Palpation for Cancer Detection through Online Tissue Viscoelastic Characterisation with a Collaborative Robotic Arm** where a new method to estimate the Viscoelastic properties of tissues was proposed. During the poster session, Luca had the opportunity to discuss his work with other researchers and receive feedback on his work. \ No newline at end of file