THE
Technology

SELECT A CATEGORY

Robotics

robotok-osszes-NICE technology

Localization & Navigation

Technologies like LiDAR (Light Detection and Ranging), visual SLAM (Simultaneous Localization and Mapping), 2D code systems and inertial navigation provide robots with reliable positioning and navigation references through feature maps, without the need for intrusive infrastructure.

Perception

The fusion of multi-sensor information such as RGBD camera, LiDAR, and safety sensors, combined with deep learning algorithms to achieve accurate detection of the target and understanding of the environment, enhance the autonomous mobile robot’s adaptability, robustness and operational safety to complex environments.

Control

The motion control parameters adaptive algorithm, based on the high-efficiency control model, realizes autonomous and efficient movements of different types of intelligent robots (including autonomous mobile robots and automated guided vehicles) and is equipped with soft and agile start-stop, high-speed and stable operation.

Software

laptop 03 software technology

Robot Management System (RMS)

Geek+ Robot Management System (RMS) is a multi-agent scheduling and task management platform system. It supports both cloud and local deployment, simultaneously prepossessing path planning, traffic management, task allocation, capacity optimization, safety emergency stops, and other tasks of large-scale mobile robot cluster, in real-time with high concurrency and reliability. Geek+ RMS provides APIs and SDKs with open standards making it easier for customers to develop and deploy business systems.

Data Platform (DP)

Geek+ Robot Management System (RMS) is a multi-agent scheduling and task management platform system. It supports both cloud and local deployment, simultaneously prepossessing path planning, traffic management, task allocation, capacity optimization, safety emergency stops, and other tasks of large-scale mobile robot cluster, in real-time with high concurrency and reliability. Geek+ RMS provides APIs and SDKs with open standards making it easier for customers to develop and deploy business systems.

Intelligent Warehouse Execution System (WES)

The flexible ARK moving system deals with multiple moving scenarios such as loading and unloading goods, storage on the dock, and factory production line feeding, etc. The sorting system can dynamically adjust the route, improve the sorting efficiency, suitable for parcel sorting, store distribution sorting, and cross-warehouse sorting. The forklift system is applied to the whole-pallet storage and high-level storage of the warehouse, and realizes the loading and unloading of whole-pallet goods in combination with the AGV. 

Each module has the flexibility to combine processes to form a variety of composite systems. At the same time, the system integrates into the equipment process control system MFC and the partner’s control system to form a complete intelligent execution system.

Simulation Platform

Geek+ Simulation Platform (SP) is a 1:1 simulation of real robot systems designed to help find the best plan and configuration before a project begins, validate the plan effectiveness and the algorithm, and further support on project evaluation and management, eventually creating reasonable ROI.

Intelligent Warehouse Management System (iWMS)

Based on the Geek+ Robot Management System and the Geek+ Intelligent Warehouse Execution System, Geek+ Intelligent Warehouse Management System (iWMS) effectively integrates all links of warehouse operation with robot picking, sorting, cross-warehousing, handling, access, and manual management, providing a complete solution for warehousing operations, and helps to meet the high market demand for speed and flexibility.

Geek+ iWMS has been tested by several large e-commerce sales such as the Singles Day sales in China, proven the capability to process orders at 30 times the amount processed on an average day. During the Singles Day sales in 2019, the system processed 8.11 million orders delivery.

Algorithm

Asset 12@4x do technology

Task Matching

Driven by the historical matching experience data, the shelves to be moved are matched with the mobile robot (AMR) one by one, and the current and future reward values are maximized by the combination of online and offline learning. Then the optimal matching strategy is obtained through continuous iterative learning. 

Path Planning

The fusion of multi-sensor information such as RGBD camera, LiDAR, and safety sensors, combined with deep learning algorithms to achieve accurate detection of the target and understanding of the environment, enhance the autonomous mobile robot’s adaptability, robustness and operational safety to complex environments.

Shelf Adjustment

The motion control parameters adaptive algorithm, based on the high-efficiency control model, realizes autonomous and efficient movements of different types of intelligent robots (including autonomous mobile robots and automated guided vehicles) and is equipped with soft and agile start-stop, high-speed and stable operation.