Robot Fleet Management

In order to develop, operate, and manage the various robots and services developed at Robotics LAB through a network, we are developing a robot fleet management solution in 3 areas based on a web technology stacks: ① a robot control system, ② robot client software for user-robot interaction and connection, ③ a DevOps pipeline for the distribution, operation, and management of solutions, and more.

Robot Control System

This control system is for defining workflow according to the purpose of the robot service, and for the monitoring, management, and operation of multiple robots through Scheduling.

[Patrol Robot Control System Video]

To ensure real-time performance, we’re developing an optimal communication solution by combining WebRTC, Socket, REST API, HTTP/2, and other various protocols with MessageQueue to provide seamless control services indoors, outdoors, and in unstable network environments such as underground spaces.

We’re developing a control system that is provided through intellectualized robot services by analyzing a vast amount of logs and data during the service operation process to acquire new insights, developing data visualization methods and Big Data analysis engines for better robot service quality, and researching and solutionizing abnormal data detection technology through machine learning. In addition, we have also been researching Monolithic, MSA, and other service architecture designs from the initial development phase to ensure the stable operation of the system. 

[Hotel Service Robot Control System]

2. Robot Client Software

This software is being developed to control robots in real time using backend technologies such as Node, JAVA, Python and to link with the servers that update the robots. We are also developing a management application for linking the robot function execution modules and to monitor the client systems of the robots.

In frontend development, various forms of visualization are being achieved using Vuejs, WebGL, etc., and screen control for touch and speech interaction. 2D and 3D animation for the expression of robot emotions are currently under development. 

[Agent Screen]

3. CI/CD Pipeline

CFor efficient development, CI/CD automation is an important concept that cannot be neglected. The Fleet Management solution of Robotics LAB organizes and automates its pipeline for all processes from design/development/build/testing/distribution/management to ensure efficient development.

This can rapidly generate a variety of different projects for each robot, service, and technology, and it also can run tests and services under various environments such as Cloud, On-premise, Embedded, etc. We plan to continue our research to create an environment that supports quick development cycles and to establish the optimal CI/CD for members to develop various different ideas and tasks efficiently.