Path Planning of Multiple Quadrotors with Ultrawideband Localization for Warehouse Inventory Management

Abstract

The paper aims to investigate the effectiveness of ultrawideband localization in the path planning of multiple quadrotors performing inventory management in warehouses. As studies investigating warehouse automation became prevalent, researchers have looked into integrating quadrotors to aid inventory management with their ability to equip a wide array of sensors and reach high altitudes. Indoor localization has been one of the main foci of research where simultaneous localization and mapping (SLAM) and radio frequency identification (RFID) were commonly used due to their reliability and accuracy. Ultrawideband (UWB) can be viable for quadrotor navigation with their effectiveness even in nonideal lighting conditions and its relatively inexpensive infrastructure. A path planning algorithm was designed which optimizes quadrotor energy consumption and yaw movement. Simulations were performed to verify the feasibility of the system, and field experiments were carried out to assess the accuracy of UWB to localize the quadrotor. Results have shown that UWB maintained an accuracy of 0.43 m with 95% reliability, even if two quadrotors were used. Moreover, the quadrotor energy consumption was minimized successfully, with the maximum energy consumption difference below 23% for multiple inventory management.