Humanoid robots are no longer confined to research labs or carefully controlled demonstrations. They're now working alongside humans in warehouses and factories, performing repetitive tasks like ferrying bins and delivering packages. Agility Robotics' Digit robot is operating at a Schaeffler automotive parts factory in South Carolina, carrying 25-pound baskets from stamping presses to conveyor belts, a job previously done by a human worker who was promoted to a supervisory role. The company plans to deploy hundreds of these robots across its U.S. and European plants by 2030, signaling that humanoid robots have moved from experimental to economically viable.

Why Are Warehouses the Testing Ground for Home Robots?

Warehouses and factories serve as the ideal environment for robots to learn how to coexist with humans before they ever enter homes. These spaces offer controlled complexity: they're built for human workers, feature predictable layouts, and present manageable safety challenges compared to the unpredictability of residential life. Digit works three shifts a day in Amazon warehouses, navigating the same stairs and hallways as human colleagues without requiring expensive facility modifications. This real-world deployment generates invaluable data about how robots behave around people, what can go wrong, and how to prevent it.

The progression is deliberate.

"Humanoids in households, for instance, are over 10 years away," said Jonathan Hurst, co-founder of Agility Robotics.

Jonathan Hurst, Co-founder and Chief Robot Officer at Agility Robotics

Homes present vastly more complexity than warehouses: unpredictable children, pets, narrow hallways, wet floors, and countless edge cases that no robot has encountered before. Warehouses, by contrast, offer a structured environment where robots can safely fail, learn, and improve before facing the chaos of domestic life.

How Do Robots Stay Safe Around Human Workers?

Safety in shared human-robot spaces requires multiple layers of protection, not just physical design. Humanoid robots like Digit face a unique challenge: they balance constantly, which means they could fall. To prevent injuries to nearby workers, Digit operates behind a plexiglass barrier at the Schaeffler factory. But barriers are only the first line of defense.

The second layer is the robot's brain: its perception and decision-making systems. Different robots take different approaches to safety:

• Digit's Approach: Uses NVIDIA's Isaac Sim application to train a whole-body control foundation model on decades of simulated time in just days, then deploys it "zero-shot" to create an "always on" safety layer that instinctively manages disturbances like bumps and pushes.
• Locus Origin's Approach: Combines LiDAR sensors to detect people and objects with depth-sensing 3D cameras to spot hazards that standard sensors miss, from dropped items to shifting floor levels, plus AI-driven object recognition to fill blind spots like raised forklift forks.
• Kachaka's Approach: Uses deep learning to analyze camera feeds pixel-by-pixel, identifying walkable areas and obstacles that LiDAR sensors often miss, while keeping high-stakes movement decisions under deterministic control rather than AI decision-making.

When multiple robots work together, coordination becomes critical.

"Without predictive path planning, where robots essentially share their intended routes and adjust proactively, they end up constantly stopping or rerouting on the fly. This creates unpredictable movements that can unsettle nearby workers," explained Kane Edwards, business development manager at Locus Robotics.

Kane Edwards, Business Development Manager at Locus Robotics

The Origin robot fleet uses multi-agent reinforcement learning through Locus Robotics' LocusONE platform, which acts like an air traffic controller, monitoring traffic, predicting congestion, and adjusting paths before dangers form.

The third safety layer involves formal boundaries and regulatory frameworks. Before Kachaka could enter KDDI's Tokyo offices, it underwent Failure Modes and Effects Analysis (FMEA), a process that analyzes every potential failure and its consequences. Agility Robotics is leading the development of ISO 25785-1, the first international safety standard for bipedal, dynamically stable robots like Digit, while building a certification scheme for insurers. These standards will eventually govern robots in homes, not just warehouses.

What Does This Mean for Workers and Employment?

The deployment of humanoid robots in factories raises legitimate questions about job displacement. At Schaeffler, the human worker whose job Digit took was promoted to a supervisory position, suggesting that robots may restructure rather than eliminate jobs. However, the scale of future deployment could change this dynamic significantly. McKinsey consultants estimate that around 200 humanoids are currently working in factories, but that number could grow to 5 million by 2040 without incurring substantial reductions in the manufacturing workforce. This projection assumes that robots displace humans in specific tasks, driving a restructuring of jobs and upgrading of remaining roles, though it's too early to evaluate the actual impact of humanoid robots specifically on employment.

The economic case for humanoid robots is becoming clearer. Agility did not disclose Digit's price, but stated that each robot costs $10 to $25 per hour to operate, while an entry-level job at the Schaeffler factory pays $20 per hour. This cost parity means humanoid robots are now economically competitive with human labor for specific, well-defined tasks, making them attractive to manufacturers seeking to optimize operations.

What Makes Humanoid Robots Different From Traditional Industrial Robots?

Traditional industrial robots are fixed to workstations and require expensive facility modifications. Humanoid robots, by contrast, are designed to fit directly into human-driven activities in environments already built for humans. Digit is built to human scale: 5 feet 9 inches tall and 143 pounds, with legs featuring inverted knees for lifting, arms designed for lifting parcels and maintaining balance, four-fingered grippers, and a torso housing processing, batteries, and sensors. This human-centric design means Digit can navigate stairs, hallways, and uneven terrain without requiring a million-dollar facility retrofit.

The robot's physical capabilities are matched by sophisticated perception systems. Digit's sensors include RGB depth cameras, LiDAR, a motion-sensing inertial measurement unit (IMU), and encoders that measure the position and velocity of its joints. Walking control is dynamic, allowing the robot to manage uneven terrain, recover from disturbances, and climb stairs and inclines. Before deployment, Agility engineers map work environments and configure specific tasks on-site, formulating tasks as structured workflows rather than joint-motor commands, specifying variables like pickup location, drop-off location, and object type.

Currently, real-world industrial use of humanoid robots is limited to a small number of early, narrow deployments in warehouses and factories. Most other humanoid systems in industry remain in pilot or trial phases. Agility robots have been tested at Amazon and GXO Logistics, while BMW has trialed Figure's humanoids. These deployments represent a step beyond simple pilots, indicating that humanoid robots are capable of economically useful work and may well take on labor currently performed by humans.

When Will Humanoid Robots Be Ready for Homes?

The path from warehouse to home is long and deliberate. Every sensor, every algorithm, and every edge case encountered in a Tokyo office building or South Carolina factory becomes data that makes the next generation of robots safer in living rooms. Warehouses are where robots learn to detect obstacles, coordinate with other robots, recover from falls, and operate safely around unpredictable humans. Only after mastering these challenges will humanoid robots be ready for the far greater complexity of domestic life.

The timeline is measured in years, not months. Warehouses, construction sites, and offices will continue to be the proving ground for humanoid robots, solving safety and reliability challenges one edge case at a time. The home is the final exam, and robots are still in their first semester of preparation.