Technology vendor Simbe Robotics Inc. has added radio frequency identification (RFID) and machine learning technologies to the latest version of its autonomous inventory-counting robot, in an upgrade that could accelerate product tracking and inventory audits in retail settings, the company said July 18.

San Francisco-based Simbe says its Tally robot can improve supply chain visibility by tracking goods after they leave the warehouse with more speed and accuracy than standard, manual counts.

In earlier versions, the mobile robot used high-resolution image sensors to scan store shelves as it maneuvered around crowded retail floors. By adding RFID scanners to the platform, Simbe says the robot can now capture in-store data on all RFID-tagged merchandise, capturing over 700 product tags per second with higher than 99 percent accuracy.

Simbe has been testing the RFID-enabled platform in pilot programs for nine months, and has now deployed Tally in scale deployments with several international retailers, Simbe CEO and Co-founder Brad Bogolea said in a statement.

The price of RFID tags and scanners usually restricts the technology to applications on crates and boxes, not low-cost individual goods. But Simbe says the system is designed for high-value items that justify the investment, such as apparel, electronics, sporting goods, and home décor.

By using a mobile robot to scan their shelves, retailers can free up store employees from carrying RFID guns, while improving real-time data collection on product availability and placement, Simbe says. That data can help retailers avoid expensive problems with merchandise that is out of stock, over stocked, or lost through product location errors, the company said.

"Brick and mortar retail is changing rapidly," Simbe's Vice President of Research and Development for Hardware, Durgesh Tiwari, said in a statement. "Innovative technologies, like RFID and computer vision, are starting to become more mainstream as retailers think about how they can optimize in-store operations."