ICRA 2024 Workshop on “Robot Embodiment through Visuo-Tactile Perception”

This is the official website for “ViTac 2024: Robot Embodiment through Visuo-Tactile Perception”.

Abstract: This is the 5th time for us to organise the ViTac workshop in the ICRA conferences, after ICRA 2019, 2020, 2021 and 2023 ViTac Workshops. It will also be timely for us to report the progress of the Special Issue (now called Special Collection) we are organising on Tactile Robotics at the IEEE Transactions on Robotics. The recent trend in embodied intelligence, i.e., intelligence that requires and leverages a physical body of a robot, demands rich physical interactions with the surrounding environment, which cannot be achieved without perception of the environment with vision and tactile sensing. Recent years have seen rapid advancements in tactile sensor development, integration of information from visual and tactile modalities, leveraging vision and tactile sensing in agile grasping and manipulation, and simulations that involve both vision and tactile sensing. It will be timely to bring together the experts and young researchers in the field to discuss these important topics in robotics. This proposed full-day workshop will cover the recent advancements in the area of visuo-tactile sensing and perception, with the aim to empower robots with visual and tactile intelligence. It will further enhance active collaboration and address challenges for this important topic and applications.

In our one-day workshop, we will discuss topics relating to visuo-tactile perception, from the development of tactile sensors, to integration of vision and touch for robot physical interaction tasks, to the creation of simulations of tactile sensors, and Sim2Real learning for visuo-tactile perception. The workshop will bring together experts from a diverse range of disciplines and encompass engineers, computer scientists, cognitive scientists and sensor developers to discuss the important topic. We will report the progress of the Special Issue (now called Special Collection) we are organising on Tactile Robotics at the IEEE Transactions on Robotics and invite the authors of the accepted papers in our special issue to present their works in our workshop. As part of the workshop, we will also organise a Challenge named ManiSkill-ViTac to benchmark the use of vision and tactile sensing in robot manipulation tasks.

For past ViTac workshops, please check ViTac2023, ViTac2021, ViTac2020, ViTac2019.

Time: 17th May 2024
Location: Yokohama, Japan.

(Tentative) Program
09:00-09:10 Organisers - Shan Luo Welcome & updates on TRO Special Collection
Session 1: Development of touch sensors - Hardware intelligence
09:10-09:20 Nathan Lepora & Shan Luo Session Intro
09:20-09:40 Ravinder Dahiya Movement Detection using Self-Powered e-Skin
09:40-10:00 Antonio Bicchi From optic flow to tactile flow. Background and new results
10:00-10:30 Coffee break, posters and demos
Session 2: Robot embodiment with vision and tactile sensing
10:30-10:40 Wenzhen Yuan & Shan Luo Session Intro
10:40-11:00 Perla Maiolino Proxy-tactile technology and tactile data generation for robot self-awareness and control
11:00-11:20 Van Anh Ho How to make robot “feel” safe?: Embodiment in robot safety with vision-based multimodal sensing
11:20-12:00 Poster presenters Lightning poster presentations (2mins each)
12:00-14:00 Lunch break, posters and demos
Session 3: Challenges in visual-tactile perception
14:00-14:10 Rui Chen & Hao Su Session intro
14:10-14:30 Mark Cutkosky Invited talk
14:30-14:50 Roberto Calandra Towards General In-hand Manipulation with Touch
14:50-15:10 ManiSkill-ViTac Challenge Conclusion of ManiSkill-ViTac Challenge, Award Ceremony
15:10-15:30 ManiSkill-ViTac Challenge Invited talk from challenge winners
15:30-16:00 Coffee break, posters and demos
Session 4 The future of robot embodiment with visual-tactile perception
16:00-16:10 Kaspar Althoefer & Gordon Cheng Session intro
16:10-16:30 Huaping Liu Large-scale tactile perception for interaction and manipulation
16:30-16:50 Maria Bauza Villalonga Visuo-tactile sensing for precise robotic manipulation with multiple embodiments
16:50-17:30 All speakers Group discussion
17:30 Finish

Organisers:
Shan Luo, King’s College London
Nathan Lepora, University of Bristol
Wenzhen Yuan, University of Illinois Urbana-Champaign
Rui Chen, Tsinghua University
Hao Su, University of California San Diego
Kaspar Althoefer, Queen Mary University of London
Gordon Cheng, Technische Universität München

Invited speakers:
Antonio Bicchi, Italian Institute of Technology and the University of Pisa, Italy
Mark Cutkosky, Stanford University, USA
Roberto Calandra, Technische Universität Dresden, Germany
Huaping Liu, Tsinghua University, China
Perla Maiolino, University of Oxford, UK
Ravinder Dahiya, Northeast University, USA
Van Anh Ho, JAIST, Japan
Maria Bauza Villalonga, DeepMind

Key dates
Posters and live demonstrations will be selected from a call for extended abstracts, reviewed by the organisers. The best posters will be invited to talk at the workshop. All submissions will be reviewed using a single-blind review process. Accepted contributions will be presented during the workshop as posters.

Expected contributions should be submitted in the form of extended abstracts (max 2 pages + references) in IEEE Conference paper format. Submissions must be sent in pdf (<5MB), to: shan.luo@kcl.ac.uk, indicating [ICRA 2024 Workshop] in the email subject.
Submission Deadline: 15th March 30th March, 2024
Notification of acceptance: 30th March 15th April, 2024
Camera-ready deadline: 15th April 30th April, 2024
Workshop day: 13th May or 17th May 2024

The accepted papers will be invited to publish their accepted extended abstracts in Springer Nature conference proceedings series Lecture Notes in Computer Science (LNCS).

Topics of Interest:
● Development of tactile sensors for robot tasks
● Simulation of tactile sensors and Sim2Real learning
● Visual-tactile sensing and perception
● The control and design of robotic visuo-tactile systems for human-robot interaction
● Interactive visuo-tactile perception in robot grasping and manipulation
● Cognitive control of movement and sensory-motor representations with vision and touch
● Bio-inspired approaches and cognitive architectures for the visuo-tactile perception
● Computational methods for processing vision and touch data in robot learning

We thank the support from the following IEEE RAS Technical Committees: