California technology company Simbe Robotics recently released a tally robot that supports RFID technology. According to the company, this is one of the few robots on the market that uses computer vision and RFID-based data and built-in machine learning capabilities.
Simbe Robotics reports that this combination of technologies enables companies such as retailers of sporting goods, electronics or apparel to capture highly accurate inventory data while also capturing images displayed in stores. Through machine learning, the robot can predict the position of the label, the type and quantity of the product attached to the label, based on the location of the label in the store. Brabe Bogolea, founder and CEO of Simbe, said that three companies are currently using RFID on their robots.
Tally is an ultra-thin robot designed for customers and shop assistants. Durgesh Tiwari, vice president of hardware development at Simbe, said it is located on a circular base that is about 62 inches high and 18 inches wide. He added that with the addition of RFID, the shape factor of the robot has not changed. The company has built a card reader that uses Impinj technology and a series of antenna arrays that can be oriented according to the direction in which the tag is read.
Tiwari said Tally is slimmer than most RFID-reading robots, making it more convenient to use in front of the store. He said: “The shops in the shops we have worked for are very narrow, and there are clothes hanging on the aisles.” For this reason, its ultra-thin size makes it suitable for reading in front of the store. It comes with a rechargeable battery that lasts for 9 to 12 hours and can be transported day and night in the store’s shelves.
Typically, Tally moves on the aisle at a walking speed equivalent to humans, depending on the density of the labels on the shelf. The tag can be read at speeds of up to 700 tags per second. It goes through one side of the shelf and then back to the other side of the shelf, capturing all the products on the shelf. It can read tags from the ground to 5 meters (16.4 feet) high and has built-in smart identification signal strength and tag position based on that intensity. In this way, Tally prevents stray readings from products in adjacent channels.
When Tally collects data, it forwards this information to cloud-based services, and Simbe Robotics software determines which items have been captured and where each product is located. The system then transmits this information and any images of the area to the store’s own software. Bogolea said that the robot will not be noticeable on the way. It provides ample space for anyone on the robot path.
The company reports that Tally’s RFID capabilities allow it to capture inventory data with 99% accuracy or greater, so stores can quickly update inventory and warehouse inventory levels. The system not only allows stores to replenish when inventory levels are low, but also helps locate products when requested by customers. For example, if someone orders goods online and plans to pick up the goods in a physical store, employees can access Simbe inventory data to see the final location of each product and then pick up the goods in the past.
This technology also reduces the need for handheld RFID tags, which is time consuming and requires the involvement of sales clerk. Bogolea said, “The core value of Tally is to liberate the salesperson and focus on customer service.” Although the system requires less manual work by store staff, it is also better than those who wish to access RFID inventory data without requiring the salesperson to use a handheld reader. The fixed card reader infrastructure required is cheap.
Tiwani said the company may choose to offer tally solutions to customers outside the retail market in the future.