Conference PaperPDF Available

Fog Robotics: An Introduction

Authors:

Abstract and Figures

Cloud Robotics (CR) is an emerging and successful approach to robotics. The number of robots or other IoT devices may increase drastically in the future which might need enormous bandwidth and there might be security concerns. If robots in CR are not secured then robots can even become surveillance bot by hackers. Moreover, if an internet connection is lost due to network hitches then in that crucial moment robot may not be available to complete its given task. For example, a robot assisting a person can stop working unexpectedly or work with the instructions from hacker. In order to address such problems, we propose a new approach to robotics-Fog Robotics (FR) in this paper, so a network of robots can be used more securely and efficiently as compared to CR.
Content may be subject to copyright.
Fog Robotics: An Introduction
Siva Leela Krishna Chand Gudi, Suman Ojha, Benjamin Johnston, Jesse Clark and Mary-Anne Williams
Abstract Cloud Robotics (CR) is an emerging and success-
ful approach to robotics. The number of robots or other IoT
devices may increase drastically in the future which might need
enormous bandwidth and there might be security concerns. If
robots in CR are not secured then robots can even become
surveillance bot by hackers. Moreover, if an internet connection
is lost due to network hitches then in that crucial moment robot
may not be available to complete its given task. For example,
a robot assisting a person can stop working unexpectedly or
work with the instructions from hacker. In order to address
such problems, we propose a new approach to robotics - Fog
Robotics (FR) in this paper, so a network of robots can be used
more securely and efficiently as compared to CR.
I. INTRODUCTION
Cloud Robotics (CR) was coined by J. Kuffner claiming
robots could be independent with no limits towards process-
ing and computation power [1]. CR provides on demand
tasks/ services, large data storage related to maps, object,
data, images, libraries, trajectories etc. all together in one
place. It was also claimed that it could reduce burden on
robots as they share information from the cloud. But right
now, the situation is changing related to security and speed,
and a new method should be available to tackle this issue.
Using the well-established concept of Fog Computing [2],
we present the concept of Fog Robotics(FR) in this paper. It
can be defined as an architecture which consists of storage,
networking functions, control with decentralized computing
closer to robots. Since this is a short introduction, our
discussion will mainly revolve around the advantages of FR.
II. FOG ROBOTICS ARCHITECTURE
The Fog Robotics architecture consists of a Fog Robot
Server (FRS), a Robot, and a Cloud system(Fig. 1). If a
robot requests data then it will first query the FRS and upon
availability, it can be directly used without involving the
cloud. If not, then the cloud will be utilized. It can teach
robots by sharing dynamic data with neighboring robots,
such as to be careful while moving towards a room as
recently one robot got crashed. FR improves security by
allowing sensitive information to be shared without requiring
it to be sent over external networks.
An FRS contains a knowledge base, computing resources,
environment models, recent robot outcome results such as
*This research was supported by Innovation and Enterprise Research
Laboratory (The Magic Lab), Centre for Artificial Intelligence, University
of Technology Sydney, Australia
S.L.K.C. Gudi, S. Ojha, B. Johnston, J. Clark, M.A. Williams
are with the Innovation and Enterprise Research Laboratory (The
Magic Lab), Centre for Artificial Intelligence, University of Tech-
nology Sydney, 81 Broadway, Sydney NSW 2007, Australia. (e-
mail: 12733580@student.uts.edu.au,suman.ojha@uts.edu.au,benjamin.johns
ton@uts.edu.au, jesse.clark@uts.edu.au, mary-anne.williams@uts.edu.au).
maps, user details, deep learning models, etc. It works as a
bridge between the robots and the cloud. It reduces band-
width and processing burden on cloud servers by processing
data locally.
Fig. 1. Architecture of Fog Robotics.
Service is delivered with the help of a local FRS which
makes latency and delay in jitter low. The distance between
client and server is near (probably one hop) based on the
circumstances such as the location of robot. FRS nodes
will be high in number, providing mobility and wireless
connectivity for all kinds of real time interactions. It is more
secure and hard to hack as they are not directly connected
to cloud. This whole process involves observing, realizing,
sharing, and reacting based on the condition of robot.
III. APPLICATIONS
An important application of FR can be seen in airports.
For example, if we consider a person asking a robot about
his departure place, then it can guide him until the escalator
and hand over its job to another robot waiting on the other
side of the escalator. The other robot will guide the person
to their destination. This process involves saving the persons
name, face, gender, shirt color, and age; and this helps the
second robot to recognize the person. In a similar way, FR
techniques can be applied in hotels, universities, subways,
bus terminals, train stations, homes and the list goes on.
REFERENCES
[1] J. Kuffner, “Cloud-enabled humanoid robots,” in Humanoid Robots
(Humanoids), 2010 10th IEEE-RAS International Conference on,
Nashville TN, United States, Dec., 2010.
[2] A. V. Dastjerdi, H. Gupta, R. N. Calheiros, S. K. Ghosh, and R. Buyya,
“Fog computing: Principles, architectures, and applications,” in Internet
of Things. Elsevier, 2016, pp. 61–75.
... Fog Robotics: The concept of Fog Robotics (FR) is mainly inspired from the concept of Fog Computing (FC)/Edge Computing (EC). It was first introduced in the year 2010 by Gudi et al., as an extension to Cloud Robotics [4]. Fog computing and Edge computing appear similar since they both push intelligence and processing closer to the data sources. ...
Conference Paper
The growing demand of industrial, automotive and service robots presents a challenge to the centralized Cloud Robotics model in terms of privacy, security, latency, bandwidth, and reliability. Especially, mobile robots have limited on-board computational power which restricts their mission planning in autonomous applications. With the evolution of Fog computing, computations may be offloaded to Fog devices and/or smart gateway devices which together form a distributed computing platform in close proximity to the mobile robot. In this work, we demonstrate the application of Fog computing for mobile robots with a specific case study of color-based object detection, tracking and mapping in a confined area. The computations required for image processing are offloaded to the Fog devices via Fog nodes and the results are acquired back in real-time. The control algorithms for tracking predefined paths and mapping a pre- defined area are validated using a controlled mobile robot with an on-board camera and processing unit. Also, the effects of improvement in latency due to fog environment as compared to on-board computation on the mobile robot is demonstrated.
Chapter
Network latency is a major problem in Cloud Robotics for human robot interaction such as tele-operation. Routing delays can be highly variable in a heterogeneous computing environment, imposing challenges to reliably tele-operate a robot with a closed-loop feedback controller. By sharing Gaussian Mixture Model (GMM), Hidden Semi-Markov Models (HSMMs), and linear quadratic tracking (LQT) controllers between the Cloud and the robot, we build a motion recognition, segmentation, and synthesis framework for Cloud Robotic tele-operation; and we introduce a set of latency mitigation network protocols under this framework. In experiments with a dynamic robot arm using this framework to perform learned hand-written letter motions, we study the motion recognition errors, motion synthesis errors, and the latency mitigation performance.KeywordsMotion segmentationTele-operationLatency mitigationCloud and fog robotics
Chapter
Full-text available
Smart factories call for a new system architecture to control massive industrial robots and sensors. Right after the concept of Cloud Robotics, Fog Robotics provide an appropriate architecture for this type of control system because of its low latency. Based on Fog Robotics, aiming at scalable resource management, we propose a Kubernetes-based microservice architecture for smart factories, including Terminal, Fog, and Cloud Layer. Kubernetes is utilized to settle the resource management of the architecture. Considering smart factory applications, we describe functions for each layer. To be specific, the warehouse scenario is studied for launching the architecture in reality. The requirements and difficulties are analyzed and settled by utilizing Robot Operating System (ROS) and microservices. ROS is used to provide the interfaces of robots and data. Warehouse tasks are separated into different microservices to develop and update conveniently. A case study in warehouses is provided, which considers the standardization, industrial requirements, and microservices. Based on path planning microservices and ROS, experiments are conducted to evaluate the performance while expanding the architecture. The results demonstrate the feasibility of launching the architecture in real applications.
Article
Fog robotics is an emerging topic that derives from cloud robotics, but similarly as fog computing, the applications require low latency connections in order to be useful in real life environments. This article presents a new perspective to the topic in which not only robotics takes advantage of the fog computing paradigm, but fog computing is able to leverage the robotics technology in order to enhance its features. This work highlights the benefits obtained by each technology when it is mixed with the other and sketches the relevant topics to research in order to make this partnership possible.
Article
With the increasing use of the Internet of Things (IoT) in various fields and the need to process and store huge volumes of generated data, Fog computing was introduced to complement Cloud computing services. Fog computing offers basic services at the network for supporting IoT applications with low response time requirements. However, Fogs are distributed, heterogeneous, and their resources are limited, therefore efficient distribution of IoT applications tasks in Fog nodes, in order to meet quality of service (QoS) and quality of experience (QoE) constraints is challenging. In this survey, at first, we have an overview of basic concepts of Fog computing, and then review the application placement problem in Fog computing with focus on Artificial intelligence (AI) techniques. We target three main objectives with considering a characteristics of AI-based methods in Fog application placement problem: (i) categorizing evolutionary algorithms, (ii) categorizing machine learning algorithms, and (iii) categorizing combinatorial algorithms into subcategories includes a combination of machine learning and heuristic, a combination of evolutionary and heuristic, and a combinations of evolutionary and machine learning. Then the security considerations of application placement have been reviewed. Finally, we provide a number of open questions and issues as future works.
Article
Full-text available
Robotic applications nowadays are widely adopted to enhance operational automation and performance of real-world Cyber-Physical Systems (CPSs) including Industry 4.0, agriculture, healthcare, and disaster management. These applications are composed of latency-sensitive, data-heavy, and compute-intensive tasks. The robots, however, are constrained in the computational power and storage capacity. The concept of multi-agent cloud robotics enables robot-to-robot cooperation and creates a complementary environment for the robots in executing large-scale applications with the capability to utilize the edge and cloud resources. However, in such a collaborative environment, the optimal resource allocation for robotic tasks is challenging to achieve. Heterogeneous energy consumption rates and application of execution costs associated with the robots and computing instances make it even more complex. In addition, the data transmission delay between local robots, edge nodes, and cloud data centres adversely affects the real-time interactions and impedes service performance guarantee. Taking all these issues into account, this paper comprehensively surveys the state-of-the-art on resource allocation and service provisioning in multi-agent cloud robotics. The paper presents the application domains of multi-agent cloud robotics through explicit comparison with the contemporary computing paradigms and identifies the specific research challenges. A complete taxonomy on resource allocation is presented for the first time, together with the discussion of resource pooling, computation offloading, and task scheduling for efficient service provisioning. Furthermore, we highlight the research gaps from the learned lessons, and present future directions deemed beneficial to further advance this emerging field.
Article
Industrial robots are widely adopted in modern industries, including automobile manufacturing, warehousing, and so on. The industrial robot with cellular network connection will be more flexible and intelligent than that without a network connection or with a conventional industrial communication network connection (e.g., fieldbus, real-time Ethernet, and WiFi). This is seen as the direction of development of the next generation of industrial robots. As a promising candidate for the next generation of cellular networks, fog radio access network (F-RAN) can provide artificial intelligence (AI) on the network edge, which is equipped with some computing and storage facilities, thus promoting the development of industrial robots in the future. In this article, we first propose a local-network cooperative control architecture for industrial robots in the F-RAN environment, which divides the robot controller into an onboard controller and a network controller, in order to decouple the basic control functions and application-dependent functions. The network controller is implemented on the fog access node (F-AP), and it possibly uses AI to bring advanced capabilities to the robot control. Then, based on the proposed architecture, a testbed for multiple automated guided vehicle (AGV) coordination in F-RAN has been designed and implemented, and a recurrent neural network (RNN)-empowered multi-AGV coordination policy is also developed. Finally, its performance is examined by several experiments. The results show that network intelligence deployed in F-RAN can improve the scheduling efficiency up to 35 percent, and the proposed control architecture can save huge backhaul traffic compared to the conventional cloud-based control architecture for industrial robots.
Article
Full-text available
Robotların üretimde kullanımı son yıllarda hızla artmaktadır. Robotların birçok sektörde insanların yerini alması ise istihdam endişelerini beraberinde getirmiştir. Robotların işsizliğe neden olacağı görüşü mali tartışmaları da alevlendirmiştir. Bu görüşe göre birçok devlet emek üzerinden vergi almaktadır ve emek üzerinden alınan vergi gelirleri toplam vergi gelirlerinin büyük bir kısmını oluşturmaktadır. Robotlar nedeniyle artan işsizlik ise bir yandan vergi gelirlerini azaltırken diğer yandan işçilerin eğitimine yönelik yapılan kamu harcamalarını arttırabilir. Bu anlamda, devletlerin mali problemler yaşamasının önlenmesi için ise robot vergisi önerilmiştir. Bu noktadan hareketle çalışmada robot vergilerinin Türkiye’deki potansiyeli araştırılmıştır. Çalışmanın sonucuna göre, Türkiye’de kısa dönemde robot vergileri gerekli değil iken uzun vadede artacak otomasyon ile robot vergilerine yer vermek çözüm olabilecektir. The use of robots in production has been increasing rapidly in recent years. The fact that robots replace people in many sectors has brought employment concerns. The idea that the robots will cause unemployment has fueled fiscal debates. According to this idea, many states taxes labor and tax revenues received from labor constitute a large part of total tax revenues. Unemployment which is increased by robots would reduce tax revenues and increase public spending on workers' education. In this regard, robot tax has been proposed to prevent the fiscal problems of the states. From this point on, this study investigates the potential of robot taxes in Turkey. According to the results of the study, in Turkey, while robot taxes are not necessary in the short-term period, with the automation that will increase in the long term, it will be possible to include robot taxes.
Chapter
The Internet of Everything (IoE) solutions gradually bring every object online, and processing data in a centralized cloud does not scale to requirements of such an environment. This is because there are applications such as health monitoring and emergency response that require low latency, so delay caused by transferring data to the cloud and then back to the application can seriously impact the performance. To this end, Fog computing has emerged, where cloud computing is extended to the edge of the network to decrease the latency and network congestion. Fog computing is a paradigm for managing a highly distributed and possibly virtualized environment that provides compute and network services between sensors and cloud data centers. This chapter provides a background and motivations regarding the emergence of Fog computing, and defines its key characteristics. In addition, a reference architecture for Fog computing is presented, and recent related development and applications are discussed.
Cloud-enabled humanoid robots
  • J Kuffner
J. Kuffner, "Cloud-enabled humanoid robots," in Humanoid Robots (Humanoids), 2010 10th IEEE-RAS International Conference on, Nashville TN, United States, Dec., 2010.