Jump to content

Robotics and Automation

Robotics & Embodied AI

Robotics and Automation

Robotics & Embodied AI

The Robotics & Embodied AI research group focuses on the intersection of robotics, learning and embodied artificial intelligence. We explore how physical embodiment influences AI decision-making, learning, and interaction in real-world environments. Our interdisciplinary team combines expertise in machine learning, computer vision, natural language processing, and robotics to develop intelligent systems that can learn, adapt, and collaborate effectively with humans. Our strong focus on practical applications, particularly in manufacturing and human-robot interaction, facilitates the creation of robotic systems that are more intuitive, adaptable, and capable of complex reasoning and task execution.

Key focus areas

  1. 1

    Learning for Embodied Robotic Intelligence

    Our research in this area focuses on developing advanced learning algorithms that enable robots to acquire, refine, and adapt their skills in real-world environments. We explore how physical embodiment influences AI decision-making and learning processes, integrating multi-modal perception to enhance robots' understanding of their surroundings. Our work extends to applying cutting-edge reinforcement learning techniques to complex robotic systems, including dual arm manipulators and humanoid robots, pushing the boundaries of what autonomous systems can achieve.

  2. 2

    Advanced AI for Multimodal Human-Robot Interaction

    This focus area centers on creating more intuitive and effective ways for humans and robots to communicate and collaborate. We utilize state-of-the-art Multi Modal Large Language Models (MLLMs) and generative AI to implement sophisticated dialogue systems and virtual assistants. Our research also encompasses the design of multi-modal interfaces that combine language, vision, and other sensory inputs, enabling more natural and context-aware interactions. We study the social dynamics of human-robot teams, aiming to create robots that can understand and respond to human needs and intentions more effectively.

  3. 3

    Robotic Systems for Complex Tasks

    In this focus area, we concentrate on developing robotic systems capable of performing complex tasks in challenging, real-world environments. Our research aims to enhance robot adaptability and performance across various domains, with a particular emphasis on applications in manufacturing and service robotics. We design intuitive interfaces for human-robot collaboration, implementing safety protocols and addressing ethical considerations in the deployment of autonomous systems. Our goal is to create robust, versatile robotic solutions that can seamlessly integrate into diverse operational settings, improving efficiency and capabilities in industry, healthcare and every day life.