Korea Institute of Mechanical Materials (KIMM) has developed a new robotic tasking artificial intelligence system designed to learn and perform everyday repetitive tasks by observing human demonstrations.
The technology was developed by a research team led by Dr. Jeong-Jung Kim from the KIMM AI Robotics Institute and allows robots to complete common activities such as organizing items, clearing tables, and manipulating objects. The system is designed to automate labor-intensive daily tasks across environments such as homes, offices, retail stores, and logistics facilities.
At the core of the platform is a hierarchical task execution framework that allows robots to break down complex jobs into sequential steps and execute them systematically. AI learns tasks by observing how humans perform them, and converts that demonstration into data that robots can use to reproduce the process.
The system integrates three key technologies: Task extraction to transform human demonstrations into usable datasets. A virtualized training environment that simulates real-world situations. Layered execution AI that ensures robots can perform tasks reliably.
KIMM said this approach differs from many existing robotic task systems, which often rely on single-task datasets or simulation-only testing. The new framework integrates a complete development pipeline, from dataset construction to real-world testing using physical robots.
During testing, the system achieved a success rate of over 90% across multiple tasks and demonstrated high reliability and adaptability even in changing environmental conditions.
KIMM researchers say the technology could expand the role of service robots in areas such as retail merchandising, warehouse logistics and general workplace support.
This study KIMM’s RoGeTA (Robot General Task AI) Framework Program It will run from 2024 to 2029 and aims to develop robot intelligence that can support a wide range of daily service tasks. The research team also plans to release task datasets and virtualized models of real environments to accelerate the development of future service robots.
KIMM research team developed a robot general-purpose task AI framework. Image provided by: Korea Institute of Mechanical Materials (KIMM)
