According to Jackie Wu, CEO at Corvus Robotics, “Most warehouses in the US still perform manual inventory audits, which is a time consuming and labor-intensive process.”

In recent years, however, many warehouse operations have started to replace their forklifts, scissor lifts, and cherry pickers with light, fast, versatile drones. “Inventory management is basically a cost center for companies,” Wu told Commercial UAV News. “Warehouse labor is either pulled from more valuable tasks like order picking or replenishment to perform the audits, or temporary labor must be hired and trained to perform the audits. And of course, it requires using and maintaining material handling equipment, which are typically gas powered.”

To address the problems associated with traditional inspections, Corvus Robotics has developed a fully autonomous drone-based solution that saves time, reduces labor costs, and produces more accurate inventory counts. “Our solution provides sustainability benefits, as well,” Wu asserted. “Our drones are electric, and they can fly at night with no lights.” 

Manufactured at the company’s headquarters in Mountain View, California, each Corvus One drone is equipped with 14 cameras designed with custom optics to see and understand the environment using a proprietary AI large-world model. In addition, “all compute for the AI large-world model runs onboard the drone in real-time, without needing a trunk full of GPUs for autonomy like a self-driving car would,” Mohammed Kabir, CTO at Corvus Robotics explained. 

“No beacons, reflectors, markers, stickers, or even WiFi are required for localization for us to navigate inside a warehouse,” Kabir stated.      

To enable autonomous navigation, the AI large-world model uses a learning based, neural network approach to autonomy to understand what inventory is where, what to look for, and how to interact with the environment. “We don't rely on any sort of high-fidelity mapping,” Kabir said. “We're just using the cameras we have and the data sourced from the fleet to do our learning.”

One company that has benefited from the Corvus Robotics approach is MSI Surfaces. A California-based distributor of flooring, countertop, wall tile, and hardscaping products, MSI turned to Corvus for help with managing its vast inventory. MSI’s previous approach had led to lost or misplaced inventory, and it put a strain on company resources. Since introducing Corvus One drones to perform daily cycle counting flights, MSI has seen drastic reductions in inventory loss, faster and more accurate order fulfillment, and greater overall efficiency. MSI now uses Corvus drones in its four largest and busiest distribution centers across the US. 

“Before we deploy at a customer facility, we typically see their current inventory accuracy hovers around 60-70%,” Wu reported. “With the Corvus One autonomous inventory management system, we boost that number to 99%, with no human intervention whatsoever. The drone flies, scans, charges, and transfers the data it collects completely autonomously.”

Wu and Kabir look forward to helping more companies with inventory control projects, and they  see the Corvus One’s successful track record with many different companies as central to further adoption. “Accurate inventory helps companies better plan for manufacturing their goods or products, forecast demand, and avoid overstock or stockouts,” Wu  said. “With a real-time accurate data set, a company can also look internally and improve its operational processes.”