Designing Voice Picking System In Warehouse Based On Indoor Positioning

With the rapid development of China’s economy, the rapid expansion of e-commerce, and customer orders become multi-species and multi-batch, the logistics industry is facing unprecedented challenges. Order picking of warehousing is one of the most critical aspects of logistics and distribution business, the traditional ways of picking are under great pressure and have shorts in some cases, when facing the growing logistics industry, and a more efficient way of picking is necessary, and therefore, this article proposed a new way, voice picking, to solve the problem that traditional picking method is facing, which can help operators free their hands and eyes, and focus on picking operations, improve picking efficiency. Meanwhile, to improve picking efficiency further, we introduce indoor positioning technology, so that we can know operators’ real-time location information, and choose the operator who is the most close to the cargo to be picked to complete order picking.In this paper, common indoor positioning technologies are analyzed, for the complex indoor environment of warehousing, selecting RFID indoor positioning technology. As for classic RFID indoor positioning algorithms, LANDMARC algorithm is the most mature, in this artic, we use improved algorithm based on LANDMARC algorithm. At first, introduce the theory of LANDMARC algorithm, and then, analyze factors which impact the accuracy of the algorithm, including selecting the value of "k", setting coordinate weighting value of "k neighbors" and selecting the position of "k neighbor", and making improvements of LANDMARC algorithm. LANDMARC algorithm often occur errors when selecting neighbors and reduce the positioning accuracy, this article analyzes the specific environment of picking system, taking the best fixed neighbor label to reduce the positioning error conditions that are due to choose the wrong neighborhoods. Finally, simulating and analyzing LANDMARC algorithm and improved algorithm, the result shows that the improved algorithm can improve the positioning accuracy of the system.In this paper, we design hardware for the system. Hardware platform includes the following modules:master chip application unit, RFID reader module, speech recognition module, Ethernet communication module. In this article, each module is analyzed, and system diagram is drawn.Finally, we design software for the system. Linux operating system is chosen for software platform, and we customize Linux system kernel. Software programming modules include: LD3320 speech recognition module, RFID signal reading module, Ethernet communication module. In this paper, we analyze the driver programming for each module under Linux system and draw main flow diagrams of software programming modules.