In the entire process of the supply chain, parcel sortation has always been a very important step. With the development of science and technology, the sorting methods have become more and more intelligent and diverse. From manual sortation, to traditional automatic cross-belt machine, then finally large-scale autonomous sorting robot system coming into existence , the efficiency is continuously increasing while the cost keeps going down. Now sorting robot is no doubt to be one of the most popular mainstream solutions in the market.
However, sorting robot working in a cluster will face a problem of path convergence, the more robots will work together, the more path convergence it will cause, consequently lead to efficiency reduction. Therefore in traditional sorting robot system, it normally needs to build a two-level steel platform in order to disperse path convergences. That is, dozens or even hundreds of mailbags or cage trolleys are placed on the ground floor, a fleet of robots patrol on the top floor and drop the parcel into corresponding slot which is connected to the mailbag, thus, a parcel sortation is completed. The problem is, it’s quite expensive to build the steel platform, let alone lots of issues caused by warehouse reconstruction.
Geek+: Interweaving Sorting Robot System
As a professional intelligent logistics robotics company, Geek+ continues exploring and innovating the application of sorting solution,and independently developed an Interweaving sorting system without steel platform, which greatly minimizes the path convergence and increases the efficiency through highly intelligent and flexible system scheduling.
This brand new Interweaving-sorting system provides a strategic solution to express industry with low investment, fast ROI, simple implementation, strong scalability, and high intelligence.
The S20 robot'sInterweaving- sorting system is more flexible and more standardized.
--Reduce the difficulty of deployment and cost
--High standardization and fast implementation. The entire system can be completed in 20 days.
--Site modifications can be minimized as much as possible by the flexible layout and application. It can be used as an automated upgrade for manual sortation, or as a fully automated sorting system.
The payload of S20 is 20KGS of parcel with dimension > 50cm*50com, which is equivalent to the cross-belt’s capacity. With good flexibility and scalability, the sorting system is capable of adapting to different scenarios and reaching an efficiency rate in either 1000-2000 pcs/hour or tens of thousands of pcs per hour.
Geek+ Interweaving sorting system is able to cover almost packages sorting scenarios in such as e-commerce, courier sorting, and will be used more widely in the future.
Robot S20 developed by Geek+
The S20 robot is developed by Geek+ independently. Its bottom robot carrier is equipped with a bundle device and roller at a height of 700mm, and can be directly connected to the cage trolley. Both S20 robot and the original flap sorting robot are navigated by QR code. After scanning the QR code on the parcel, the system identifies the destination of each parcel then assigns a robot to move it to the corresponding cage trolley.
Accessible Machine-Learning through reinforcement learning of system application
The robot will receive feedback from the external environment when making decisions. While constantly absorbing feedback, the robots will make a decision instructed by the algorithm and maximize the performance. When robots work in a cluster, the system is able to make immediate decisions to maximize cluster efficiency according to real-time environment.
Less robot intersections by dynamic adjustment on path end point
The dynamic adjustment of end point will increase the complexity of manual parcels collection management, thus raise the risk of error.
However, Geek+ Sorting System can further liberate the work of manual parcel collection, and arrange the location of sorted parcel containers. This will facilitate workers on manual inventorying and loading then greatly reduce the chance of errors. The essence of these two operations is to reduce system complexity by adding intermediate nodes in proper place.