SenseTime Launches SenseMart OS, Bringing Embodied Intelligence to Retail

TechNode reporter Lucia wrote the original report. On Sept. 23, SenseTime launched SenseMart OS at its Shanghai headquarters, presenting it as a physical operating system for retail.

The system connects heterogeneous robots, retail equipment, and commercial operations. It aims to give operators, brands, and venues a deployable, operable, and replicable embodied-retail solution. At the event, the SenseMart Go robotic store showed how the OS coordinates these components in a real retail setting.

A robot retrieves a bottle of mineral water from a shelf after a TechNode reporter places an order.

The demonstration illustrates the system’s basic pick-and-deliver capability in a live store.

SenseMart OS offers a glimpse of how embodied intelligence could move beyond demonstrations into everyday commercial environments.

The launch also reflects a broader debate over what will define the next phase of competition in embodied intelligence: robot hardware, data, models, or systems?

SenseTime already had years of experience in retail, according to Dr. Yi Shuai, co-founder and chief scientist of SenseTime Smart Retail. Its product-recognition systems process millions of orders daily and its database covers more than 300,000 SKUs, or distinct product types and variants.

Unlike warehouses or factories, retail requires robots to interact with people. That tests perception, interaction, decision-making, and execution at the same time.

Retail also offers clear business metrics—sales, repeat purchases, and ROI—that can show whether the technology is working.

Dr. Yi said SenseMart OS focuses on three elements supporting four core capabilities: perception, interaction, decision-making, and execution. SenseTime aims to expand the system beyond retail into other offline service industries.

The system targets customer satisfaction, repeat purchases, and transaction completion, not just robot task success. It can connect wheeled dual-arm robots, humanoids, grippers, and other hardware depending on the task.

It combines visual, transaction, inventory, and voice-interaction data across the customer journey to train retail agents.