Morning workshop: Robotics for InterAction Technology

Giovanni Berselli (University of Genoa), Antonio Bicchi (University of Pisa), Cristina Piazza (Technical University of Munich), Bruno Siciliano (University of Naples Federico II)

Overview

A recent breakthrough in robotics marks a significant milestone as Figure, a leading robotics company, has unveiled Figure 01, a humanoid robot capable of engaging in full-fledged conversations with humans. This remarkable feat is made possible by integrating OpenAI’s cutting-edge technology in Figure 01, showcasing the potential of robotics to bridge the gap between humans and machines. Figure 01’s integration of OpenAI’s sophisticated visual and language models empowers it to comprehend language and interpret visual information seamlessly. The collaboration between Figure and OpenAI represents a convergence of Robotics and AI, promising to transform industries across the board. By enhancing robots’ ability to interact naturally with humans and perform complex tasks, this interdisciplinary approach opens doors to diverse applications in manufacturing, healthcare, and personal assistance sectors, among others. As humanoid robots become more advanced and autonomous, they hold the potential to redefine societal norms and expectations. This workshop is aimed at addressing the challenges of novel paradigms for human‒machine interaction in the context of InterAction Technology (IAT).

The expected outcomes of the workshop are:

• To understand the importance of InterAction Technology for the development of future AI-empowered robots
• To study the relationship between human and machine towards a resilient and sustainable ecosystem
• To identify new emerging areas for scientific collaboration with new communities of users and developers of robotics technology

Please see this PDF for more information on the workshop, or see “Robotics for InterAction Technology: Italy’s key role in the next revolution” for more on the topic.

Intended Audience

The workshop is addressed to:

• Graduate students new to the fields of robotics, AI and shared control
• Experienced researchers in aforementioned fields looking to expand the complexity and relevance of case studies used to demonstrate their techniques
• Industry practitioners interested in and working on problems related to Industry 5.0 and autonomous systems may also benefit from the workshop and be able to contribute valuable insights to academic participants

Schedule

The plan is to have a brief introduction by the organizers, to be followed by four technical presentations of the invited speakers who will address different aspects, application scenarios and use cases of IAT. The detailed agenda will be finalized depending on the participation of additional speakers following the call for contributions we are about to launch.

Afternoon workshop: Applied Machine Vision for Robotic Applications

Callen Fisher (Stellenbosch University)

Overview

“Applied Machine Vision for Robotic Applications” is a dynamic half-day workshop designed to equip participants with practical knowledge and skills in utilizing machine vision technologies within robotic systems. Participants will delve into the fundamental principles of machine vision, exploring its role in enhancing robotic perception and decision-making processes. Through hands-on demonstrations and interactive sessions, attendees will learn how to integrate machine vision algorithms into various robotic applications, such as object recognition, navigation, and inspection. By the end of the workshop, participants will have gained a comprehensive understanding of how to leverage machine vision effectively to optimize robotic performance and efficiency in real-world scenarios.

Intended audience

This workshop is intended for researchers (both academic and industrial) and practitioners interested in investigating robotic vision problems. The workshop will be accessible for those with limited experience in machine vision and will feature a collection of interactive Jupyter worksheets to actively engage those attending.

Schedule

Duration: Half Day (4 hours)

1. Welcome and Introduction (15 minutes)
   1. Introduction of the facilitator and participants.
   2. Overview of the workshop objectives and agenda.
2. Introduction to Machine Vision (30 minutes)
   1. Basic concepts of machine vision.
   2. Importance and applications in robotics.
   3. Key components of a machine vision system.
3. Hardware Components for Machine Vision (30 minutes)
   1. Cameras: Types (e.g., RGB, depth cameras) and specifications.
   2. Lighting: Importance and techniques.
   3. Lenses: Types and selection criteria.
   4. Discussion on hardware setup for different applications.
4. Software and Algorithms (45 minutes)
   1. Image acquisition and preprocessing techniques.
   2. Introduction to computer vision libraries (e.g., OpenCV).
   3. Key algorithms: edge detection, contour detection, and feature extraction.
   4. Machine learning in vision: Overview of convolutional neural networks (CNNs).
5. Break (15 minutes)
6. Practical Session: Building a Simple Vision System (1 hour)
   1. Hands-on exercise, either with web camera or images/video provided
   2. Basic image processing tasks (e.g., filtering, thresholding).
   3. Real-time object detection using pre-trained models.
7. Integration with Robotics (30 minutes)
   1. Overview of robotics frameworks (e.g., ROS).
   2. Techniques for integrating vision systems with robotic systems.
8. Advanced Topics and Future Trends (20 minutes)
   1. Deep learning in machine vision: Current trends and future prospects.
   2. Discussion on emerging technologies (e.g., 3D vision, AI-driven vision systems).
   3. Mapping and localisation.
9. Q&A and Wrap-up (10 minutes)