Recent research published in the journal Nature Machine Intelligence by researchers at the University of Edinburgh suggests that an AI-powered robot capable of preparing coffee in a busy kitchen could pave the way for the next generation of intelligent machines.
Coffee-making robot pours water from a kettle into a cup. Image Credit: Ruaridh Mon-Williams
Researchers state that by integrating advanced AI, sensitive sensors, and precise motor skills, the robot can interact with its environment in a more human-like manner than ever before.
Robots may be able to perform duties that were previously limited to humans according to new technology created by a team from the University of Edinburgh.
According to experts, robots perform best in environments with strict controls, like factories and production lines, but they have trouble in dynamic, unpredictable environments, like kitchens.
This is because robots have historically relied on pre-programmed actions and responses and are unable to adjust in real-time to unforeseen challenges.
The Edinburgh team has now developed a robot that can interact with items and humans in difficult situations by combining advancements in AI and delicate motor abilities. According to the team, prior advancements in these fields have mostly occurred independently of one another.
The innovative device, a robotic arm with seven movable joints, first processes verbal instructions before analyzing its surroundings.
It then looks around the kitchen for a mug, figuring out how to unlock cabinets with opening mechanics it has never seen before. The robot then measures and blends water from a kettle with ground coffee from a jar in a predetermined ratio.
According to the scientists, the robot's technology allows it to easily adjust to unanticipated situations, including when someone bumps into or moves the cup while it is operating.
The Engineering and Physical Sciences Research Council (EPSRC) provided funding for the study. Ruaridh Mon-Williams, a Ph.D. candidate enrolled at Princeton University, the Massachusetts Institute of Technology, and the University of Edinburgh, spearheaded the project.
We are glimpsing a future where robots with increasingly advanced intelligence become commonplace. Human intelligence stems from the integration of reasoning, movement, and perception, yet AI and robotics have often advanced separately. Our work demonstrates the power of combining these approaches and underscores the growing need to discuss their societal implications.
Ruaridh Mon-Williams, School of Informatics, University of Edinburgh
Journal Reference:
Mon-Williams, R., et al. (2025) Embodied large language models enable robots to complete complex tasks in unpredictable environments. Nature Machine Intelligence. doi.org/10.1038/s42256-025-01005-x