Commercial cleaning robots have traditionally been designed around large floors and relatively predictable environments. Pudu Robotics is taking the technology in a smaller direction with the PUDU ET1, an AI-native compact scrubber-dryer designed for spaces of roughly 100 to 800 square meters. The company says the new robot combines autonomous cleaning, adaptive AI, automated docking and narrow-space navigation for convenience stores, restaurants, pharmacies, hotels and offices.

The next battleground for commercial cleaning robots may not be the warehouse or airport. It could be the neighborhood convenience store.

Those environments present an awkward automation problem. They need frequent cleaning, but their narrow aisles, customers, shelving and constantly changing floor conditions can make larger autonomous machines impractical. At the same time, smaller cleaning machines often require more human supervision.

Pudu Robotics is attempting to close that gap with the PUDU ET1, the first product in its new PUDU E Series.

The compact scrubber-dryer is designed for commercial spaces measuring between 100 and 800 square meters. Pudu is targeting convenience stores, drugstores, chain restaurants, business and economy hotels and offices—locations where cleaning has to happen around people and business operations rather than in empty facilities.

The company’s proposition goes beyond making a smaller floor-cleaning robot. ET1 is designed as an AI-native cleaning system, combining perception, navigation, cleaning controls and automated maintenance into a single workflow.

That distinction matters as service robotics moves from simple task automation toward more adaptive machines.

ET1 combines sweeping, scrubbing, vacuuming and dust-mopping. Pudu says its front-sweep and rear-scrub configuration can handle dry debris and wet spills in a single pass. For more demanding cleaning, the system uses a roller capable of reaching 800 revolutions per minute and suction rated at 20 kPa.

The robot also supports 85-degree-Celsius hot-water scrubbing for stubborn or sticky contamination, particularly in environments such as restaurants and convenience stores.

But the more interesting technology is what happens when floor conditions change.

Pudu’s AI Magic Cleaning system uses perception and motion-control capabilities to adapt the cleaning process to the environment. The robot can recognize floor surfaces and stains, generate cleaning heatmaps and alter its cleaning behavior according to what it detects.

For example, the system can identify carpet edges and lift the scrubbing roller to avoid wetting carpet. It can also lift brushes when encountering liquid spills, reducing the risk of spreading contamination.

For difficult stains, the company’s AI Adaptive Scrubbing system can reduce the robot’s travel speed while increasing roller speed and water output.

That represents a broader shift in commercial robotics: the robot is not simply following a predetermined route. It is attempting to change how it performs a task based on what its sensors detect.

The practical value of that capability may be greater in small retail environments than in large, controlled facilities.

A convenience store can have customers walking through an aisle, spilled drinks appearing without warning and merchandise temporarily blocking a previously mapped route. A restaurant can move tables and chairs between cleaning cycles. An office can have people working throughout the day.

Autonomy in these settings therefore depends on more than navigation.

ET1 is also designed to minimize the human labor required after a cleaning cycle. Its 8-in-1 docking station can automatically charge the robot, refill clean water, drain wastewater, dispense detergent, rinse the roller and squeegee, spin-dry the roller and heat water.

The robot returns to the dock after completing a task and performs a self-cleaning cycle. It can also automatically collect dust, hair and debris.

That creates a potentially important distinction between robotic cleaning and genuinely automated cleaning.

A robot that still needs employees to empty tanks, wash components and prepare it for every shift may reduce physical cleaning work without eliminating much operational overhead. Automating those peripheral tasks makes the business case more compelling.

Pudu is also emphasizing deployment simplicity.

The ET1 supports automatic mapping and one-click starting, allowing store employees to begin operations without extensive technical setup. The company says its software includes separate interfaces for managers and daily operators, as well as maintenance reminders.

For retailers operating hundreds or thousands of locations, that could matter as much as cleaning performance.

The economics of commercial robotics are increasingly shifting toward fleet-scale deployment. A chain cannot realistically train every store manager to configure a complicated robot. It needs machines that can be installed, operated and maintained with relatively little specialist intervention.

ET1’s compact dimensions are part of the same strategy. Pudu says the robot can operate in spaces with minimum clearances of 50 centimeters in width and 60 centimeters in height, allowing it to navigate narrow aisles and low-clearance areas.

That makes the product less of a miniature version of a conventional commercial scrubber and more of a machine designed around the constraints of small-format retail.

There is another unusual component: advertising.

Pudu says ET1 can optionally be fitted with a 21.5-inch display that retailers can use for promotions, product campaigns and brand content. Cloud-based content management and customizable body wraps could turn the cleaning robot into a mobile digital-signage surface.

That feature brings the product into territory familiar to retail technology and MarTech. The robot becomes not just an operational asset but potentially an in-store media touchpoint.

Whether retailers will see meaningful advertising value in a cleaning machine remains to be demonstrated. But the idea reflects a wider trend in service robotics: machines operating in public spaces can become platforms for additional customer-facing services.

Pudu is entering the market with substantial commercial-robotics experience. According to Frost & Sullivan’s 2025 Global Embodied Intelligence and Commercial Service Robotics Independent Market Research Report, cited by Pudu, the company ranked first globally in commercial cleaning robotics revenue with a 29% market share.

That claim is company-reported and should be viewed in that context, but it illustrates the scale of the competitive market Pudu is targeting.

The broader service-robotics ecosystem includes companies such as Keenon Robotics, Gaussian Robotics and LionsBot, while major technology companies including NVIDIA, Google and Microsoft are supplying AI, robotics and cloud infrastructure that increasingly underpins autonomous machines.

Pudu’s challenge is now moving from technical capability to deployment economics.

For a small store, the relevant question is not whether a robot can scrub a floor. It is whether the robot can reliably do so with minimal supervision, operate around customers, avoid disrupting employees and reduce enough labor or cleaning overhead to justify its cost.

That makes ET1 an interesting test of where commercial robotics is heading.

The first generation of cleaning robots proved that machines could automate repetitive floor care. The next generation is trying to make that automation adaptive, self-maintaining and viable in environments too small or too unpredictable for conventional autonomous systems.

If Pudu can deliver that combination at chain scale, the humble convenience store could become an important proving ground for AI-native physical automation.

Market Landscape

Commercial service robotics is moving from large, controlled environments toward distributed deployments across retail, hospitality, healthcare, restaurants and offices.

The market challenge is increasingly not whether robots can perform a task, but whether they can operate autonomously enough to deliver a positive return on investment.

Large cleaning robots remain attractive for airports, warehouses, shopping malls and other high-square-footage facilities. Smaller machines such as ET1 target the much larger number of locations where floor area is limited but cleaning frequency is high.

Pudu’s approach also illustrates the convergence of AI, robotics and enterprise operations software. Perception systems allow robots to react to changing environments, while cloud management can potentially give multi-location operators greater visibility into fleets, maintenance and performance.

For retailers, the strongest value proposition may ultimately be operational consistency. A chain could standardize cleaning procedures across hundreds of locations rather than relying entirely on local staffing and manual routines.

However, adoption will depend on reliability, maintenance costs, battery performance, navigation accuracy and the amount of human intervention required in real-world stores. AI-native capabilities are useful only when they translate into fewer interruptions and lower total operating costs.

Top Insights

• Pudu ET1 targets 100–800-square-meter commercial spaces, bringing autonomous scrubbing and AI-based cleaning adaptation to convenience stores, restaurants, pharmacies and offices.
• AI Magic Cleaning adapts operations to floor conditions, recognizing surfaces and stains while adjusting speed, water output and brush behavior during cleaning.
• An 8-in-1 docking station automates maintenance tasks, reducing employee involvement in charging, refilling, wastewater handling, detergent management and roller cleaning.
• Retail chains are a major deployment opportunity, because compact dimensions, automated mapping and simplified controls can make robotics easier to scale across locations.
• ET1 also introduces a retail-media angle, with an optional 21.5-inch display allowing stores to combine cleaning automation with promotional and advertising content.

Power Tomorrow’s Intelligence — Build It with TechEdgeAI