Design of a household robot with autonomous navigation for object detection and sorting

Researchers have designed an intelligent household robot capable of autonomous navigation, object detection, and sorting. This robot, equipped with advanced technologies like depth cameras, YOLOv11 object recognition, and a flexible gripper, can perform various tasks, including retrieving objects and interactive entertainment. The study, published in Robot Learning, highlights the robot’s ability to navigate indoor spaces, recognize objects with high accuracy, and grasp items of different sizes and weights. This innovation could transform household robots into more efficient and interactive family assistants.

In the quest to make household robots more efficient and interactive, a team of researchers from Suzhou Industrial Park Institute of Vocational Technology and Xi’an Jiaotong-Liverpool University has developed an intelligent robot that can autonomously navigate, detect, and sort objects. This breakthrough, published in the journal Robot Learning, brings us closer to having robots that can perform a variety of household tasks with ease.

"This robot is designed to be a comprehensive family assistant, capable of performing tasks that make daily life more convenient and enjoyable,"says Bingjie Xu, one of researchers on the project. "From picking up toys to fetching items, it can handle a wide range of objects with precision and care."

The robot integrates a movable chassis, robotic arm, lifting platform, and flexible gripper, allowing it to grasp objects of varying sizes and types. The key to its functionality is the YOLOv11 object recognition system, which, combined with a depth camera, enables the robot to detect target objects with high accuracy. The YOLOv11 algorithm significantly enhances the robot's ability to recognize and locate objects in real-time, this makes it highly effective for tasks like cleaning up or fetching items.

One of the most significant features of this robot is its autonomous navigation capability. Using 2D LiDAR and the Navigation2 framework in ROS2, the robot can generate a 2D radar map of its environment and navigate indoor spaces without human intervention. "The robot can move seamlessly through rooms, avoid obstacles, and reach designated locations on its own," says Yangzesheng Lu. "This feature is crucial for tasks like picking up toys scattered around the house or fetching items from different rooms."

The robot’s ability to interact with humans is another highlight. It uses a speech recognition system to understand and respond to voice commands, making it easier for family members to communicate with it. The speech recognition system allows the robot to understand and execute commands from family members. This makes it a more engaging and useful assistant.

In a series of experiments, the robot demonstrated its ability to grasp objects of different sizes and weights, including an orange, a bowl, a bottle, a bear, a book, an umbrella, a handbag, and a potted plant. The robot's flexible gripper and advanced vision system ensure that it can handle objects with precision and care, makes it highly adaptable to different household scenarios.

The researchers believe that this robot could be a game-changer in the field of household robotics. "We are excited about the potential applications of this robot in everyday life,"  says Dr. Bu. "Future work will focus on enhancing the robot's object detection accuracy and integrating large language models to improve its semantic understanding capabilities."  These advancements will further bridge the gap between humans and robots, fostering more seamless and intuitive interactions. "This robot represents a significant step towards creating more efficient and interactive family assistants,"says Dr. Sun. "We hope it will enhance daily life in modern households."  

This paper, titled “Design of a household robot with autonomous navigation for object detection and sorting,” was published in Robot Learning. Xu B, Lu Y, Wang J, Bu Q, Leach M, et al. Design of a household robot with autonomous navigation for object detection and sorting. Robot Learn. 2025(1):0005, https://doi.org/10.55092/rl20250005.