Madrid – Researchers at Universidad Carlos III de Madrid (UC3M) are making strides in assistive robotics, developing a new methodology that allows robots to learn complex, two-armed tasks by observing humans. This breakthrough, presented at the IROS 2025 robotics conference, focuses on improving how robots coordinate their movements, bringing closer the day when they can reliably assist with everyday chores like setting and clearing tables, and providing support for aging populations. The core of this advancement lies in combining observational learning with communication between the robot’s limbs, enabling a more natural and intuitive approach to robotic task acquisition.
The challenge in robotics has long been coordinating two arms working in tandem. Traditionally, programming such coordination required painstakingly writing thousands of lines of code to define every movement. The UC3M team’s approach, however, utilizes a robot named ADAM (Autonomous Domestic Ambidextrous Manipulator) to learn by watching people perform the tasks. This “imitation learning” allows ADAM to replicate human actions, but crucially, the system doesn’t simply copy; it adapts. As Alicia Mora, a researcher from the Mobile Robots Group at the UC3M Robotics Lab, explained, ADAM is already capable of performing assistive tasks in home environments, including setting the table, tidying the kitchen, and even retrieving items like medication or a coat when requested.
“It can, for example, set the table and clear it afterwards, tidy the kitchen, or bring a user a glass of water or medication at the indicated time,” Mora said. “It can also help them when they are going out by bringing a coat or an article of clothing.”
The ‘Rubber Band’ Effect: Adapting to Real-World Variability
The team’s innovation doesn’t stop at simple imitation. A key element of their work addresses the inherent unpredictability of the real world. If a robot learns to grasp an object in a specific location, a slight shift in that object’s position can throw off a system that relies on exact replication. To overcome this, the researchers developed techniques that allow the learned movements to behave like a “rubber band.” If the target changes position, the robot’s trajectory smoothly adjusts to reach it, maintaining the core essence of the action – like keeping a bottle upright to prevent spills. This adaptability is crucial for robots operating in dynamic environments.
ADAM: An Assistive Robot Focused on Aging Populations
ADAM is specifically designed to assist elderly individuals with daily tasks, both in their homes and in care facilities. Ramón Barber, director of the Mobile Robots Group and a professor in the UC3M Department of Systems Engineering and Automation, emphasized the growing demand for such technology. “We all know people for whom simple gestures, such as someone bringing them a glass of water with a pill or setting the table for them, represent a remarkably significant help. That is the main objective of our robot,” he stated. He also highlighted the demographic shift occurring in many societies, with an increasing elderly population and a decreasing number of caregivers, making assistive robotics increasingly vital.
“Every day Notice more elderly people in our society and fewer people who can care for them, so these types of technological solutions are going to become increasingly necessary,” Barber added. “Assistive robots are emerging as a key tool to improve the quality of life and autonomy of people.”
From Lab to Living Room: The Future of Assistive Robotics
Currently, ADAM remains an experimental platform with a cost estimated between 80,000 and 100,000 euros. However, researchers are optimistic that the technology is mature enough to suggest that similar robots could become commonplace in homes within the next 10 to 15 years, and at a significantly more affordable price point. The development represents a significant step toward more natural and easily teachable service robots. The research team’s approach to robotic manipulation, focusing on adaptation and understanding rather than mechanical repetition, is a key component of this progress.
The techniques developed by the UC3M team address the problem of robotic adaptation by allowing learned movements to adjust smoothly to changing target positions, maintaining the core action. This “rubber band” effect ensures the robot can handle real-world variability without losing functionality.
The team’s work highlights the potential of imitation learning, where robots learn by observing human demonstrations. This contrasts with traditional programming methods that require extensive coding for even simple tasks. By allowing robots to learn from observation, researchers are paving the way for more intuitive and user-friendly robotic assistants.
The next step for the UC3M team involves refining ADAM’s capabilities and exploring ways to reduce its cost, bringing the promise of accessible assistive robotics closer to reality. Further research will focus on expanding the range of tasks ADAM can perform and improving its ability to interact safely and effectively with people in everyday environments.
What are your thoughts on the role of robots in elder care? Share your comments below, and let’s continue the conversation.
