Design And Implementation Of Robot Container Cloud Platform System Based On Mircroservice

The term cloud robotics is a combination of cloud computing and robotics first proposed by James J.Kuffner at an academic conference at Carnegie Mellon University in 2010.Early cloud robot studies are mostly aimed at a particular application scenario or special category of robots and are not versatility.In this paper,Kubernetes and containerization technology are used to design and implement a robot container cloud platform based on Microservices Architecture(MA).The main contents of this paper include the following:(1)Clarify what are the functional requirements and nonfunctional requirements of the current robot cloud platform;propose a general intelligent robot cloud platform technical architecture,and divide the system.(2)Build an open source community artificial intelligence cloud platform independently as a resource pool for terminal equipment such as robots,and finally realize the onestop development,deployment and management of algorithm applications,and realize the effective distribution and recovery of server resources.At last,based on Kubernetes,an elastic expansion mechanism for algorithm applications is implemented.(3)Design and implement the access layer of the robot cloud platform,using the microservice architecture design method to split each subservice,including the data management microservice DeviceCenter and the task execution engine microservice ControlCenter,which reduces the coupling between modules.At the same time,in order to facilitate robot application developers to build robot applications according to different scenarios,a behavior tree module has been added.(4)Verify the functionality of the entire system by use case scenario experiment,the expected system goals can be achieved in experimental environment.