Scaling Robot Intelligence With Cloud-Based Cloning Platforms

1. Introduction

Robotics and artificial intelligence (AI) have experienced a complete world over the past decade. It is an age where machines are more than smart tools they must learn and adapt just like we humans. These technologies coming together thus offer great potential for revolutionizing many industries, from manufacturing to healthcare. But as we explore the frontier of AI embodied robotics, there is an immediate problem-scalability. In today's blanket of robotic systems, the bottleneck to widespread application and efficiency lies in the limitations of individual devices (Kumar and R. Singhal, 2020). These limitations, from basic computational limitations to the low capabilities of memory itself, affect robots' ability to cope with more and more complex tasks. Enter the paradigm of cloud-based cloning platforms, a visionary concept that addresses the scalability dilemma and opens new horizons for the deployment of intelligent robotic systems (Nayyar and A. Nagpal, 2018). The idea of cloud robotics is a new paradigm that has opened up a lot of doors for robotics research. Without a doubt, the availability of storage, ease of access, and remote computing assistance provide an appealing domain of study that aims to improve the on-board robot's capacity to do complicated jobs. The first operational cloud robotics project, RoboEarth, was announced in the middle of 2009, however networked robotics and online assistance in robotic systems may be dated to the end of the 1990s. Cloud computing is expanding at a rapid pace, which is attracting more and more academic attention to the idea of cloud robots. One such definition of “cloud computing” is a “cloud service” that facilitates a distributed virtualized computing environment with the purpose of improving a robot's ability to do complicated tasks. Some of the current issues for on-board robots include, for instance, creating a 3D representation of an item or determining the journey direction in unstructured environments (P. S. Kumar and A. Sharma, 2019). The shift away from on-board or local network-based services and towards cloud-based ones is being driven by the effects of the cloud on various computer systems and services (S. El-Gendy, 2020). With the cloud's scalability, big storage capacity, cheap maintenance cost, and high processing power, the robotics field might potentially benefit from these features. In most cases, the on-board robot still has storage and processing limitations (K. N. V. Srinivas and P. K. Lakineni, 2023). So, it's more important than ever to load the robot's sensor data into an external computer system that can handle the processing, store, and share of knowledge, and then send the results back to the robot. Figure.1 shows that distributed computation and storage resources are the backbone of cloud robotics infrastructure (P. S. Kumar, S. Gangadharan and S. C. N, 2023). These resources are made available to different kinds of robots so that they may carry out activities in the cloud or communicate with one another.

Figure 1.

A comprehensive overview of the architectural framework and diverse applications in cloud robotics