NEO

Household chore capability represents one of the most anticipated applications of humanoid robots. The NEO is designed to assist with everyday tasks such as tidying rooms, moving objects to their proper locations, and performing routine domestic maintenance. This requires a combination of object recognition (identifying what needs to be picked up or moved), task planning (determining the sequence of actions), and manipulation skill (grasping and handling diverse household objects safely). The variety of tasks and objects encountered in a real home makes household chores one of the most technically challenging applications in robotics — far more unpredictable than structured industrial environments.

Tidying up requires the NEO to identify objects that are out of place, determine where they belong, and physically move them. This seemingly simple task involves sophisticated AI capabilities: recognizing hundreds of different household objects, understanding the concept of where things "should" be based on learned household norms, planning efficient tidying routes through cluttered rooms, and manipulating objects of varying size, weight, shape, and fragility. 1X Technologies approaches this challenge by combining computer vision for object recognition with learned manipulation policies that generalize across object types. The robot improves over time as it learns the specific preferences of its household.

Safe human interaction capability means the NEO is designed to operate in shared spaces with people rather than being confined behind safety cages. This involves multiple engineering approaches: force-limited actuators that restrict the power the robot can exert during unexpected contact, real-time proximity sensing that slows or stops the robot when humans are nearby, compliant mechanical design that absorbs impact energy, and software-level safety monitoring that enforces behavioral constraints regardless of task instructions. For humanoid robots, safe human interaction is essential because the intended operating environments — warehouses, factories, hospitals, homes — all involve close coexistence with people.

Adaptive learning means the NEO improves its performance over time based on experience and user feedback. Rather than operating with a fixed set of behaviors, the robot can refine its understanding of its environment, learn new tasks, adjust to its owner's preferences, and recover more gracefully from mistakes. This capability is powered by machine learning algorithms that continuously update the robot's internal models based on real-world interaction data. For a home-oriented robot, adaptive learning is particularly valuable because every household is different — the robot must adapt to unique room layouts, object collections, daily routines, and individual preferences that cannot be pre-programmed at the factory.

Gentle manipulation capability means the NEO is designed to handle objects with appropriate delicacy — applying enough force to maintain a secure grip but not so much as to damage fragile items. This requires sophisticated force sensing and control at the gripper and arm level, where the robot monitors the forces it applies in real time and adjusts accordingly. For a household robot, gentle manipulation is critical because homes contain a wide range of object fragilities — from sturdy dishes to delicate glass, from rigid books to soft clothing. The ability to modulate grip strength and manipulation speed based on the detected properties of each object is what makes a household robot practically useful rather than a liability.