797
Elevation In Robotics And Automation Using Level Structured
Cloud Cadre
Saumyo GhoshP
1
P
, Aksha MondalP
2
P
1
P
Mechanical, JSPM’s RSCOE, Pune -411033
P
2
P
Computer, SKN SITS, Pune –
Savitribai Phule Pune University, Pune, India
P
1
P
2
P
31T
Abstract
The rattling increase use of cloud in market propelled us to think on more possibilities of advancement in robotics and automation using cloud technology. Recently, robots and automation systems have been at the anterior of research with the preponderance of systems still operating independently using onboard computation, memory manipulation and communication. New approaches where robot and automation processing is performed remotely with access to large scale datasets, support a range of functions. Cloud Robotics supplements performance enhancement of robotics and autonomous systems by providing a global infrastructure in innovative ways. The paper talks about the recent research into Cloud Robotics for performance enhancement in robotics and autonomous systems: 1) Aloof Brain, 2) Big Data, 3) Collective Robot Learning. It has also insinuated some levels of segregation of cloud data to provide a more secured robot control system. The information is acquired according to the availability of contents corollary among the aplenty of stored relevant data, drifted to extract the better outcome of corollary and is made available through a various level structured cloud cadre (LSCC). LSCC is an assimilation of mother cloud, daughter cloud and granddaughter and various clouds integrating the gathered information required for controlling the robot.
31T
Keywords31T: level structured cloud cadre, segregation, mother cloud,
daughter cloud, granddaughter cloud, aloof brain, big data, collective robot learning.
I.
I
NTRODUCTIONThe reliable advancement of technology and its use in every field are increasing the dependencies of people on technology more. Living in a world without machines is almost impossible to imagine. Without even the simplest machines, many tasks that we do every day would be almost impossible. Machines make it easier for humans to perform everything from the simplest to the most complicated of tasks. Robot the red hot machine has banged its entry to the society, and has revolutionised the human way of living in the future. A robot is a mechanical or virtual artificial agent usually an electro-mechanical machine that is guided by computer program and electrical circuits [30]. Robots have replaced humanities in the assistance of performing those repetitive and dangerous tasks which humans prefer not to do due to size imitation or even those such as in outer space or bottom of the
sea where human cannot survive extreme environment. It can be humanity’s loyal servant, but it can also be mankind’s worst enemy, which may be our doom. Thus it has to be programmed very carefully and efficiently. Robots are attendant by intelligent chips indwelled in it. These chips are static and once programmed it can’t be changed, moreover the chips inhabit a huge space of robot.
With the recent popularity of “Cloud Computing” systems and the de facto increased standard in many areas of advanced information technologies there has been increasing interest in applying similar concepts to robotics and autonomous systems. Due to the limited capacities of on-board processing, storage and battery capacities, robotic devices are constrained to numerous limitations. With these increasing limitations of chips, the on-board processing concept is revamped to a concept of Robot Cloud i.e. robotics and automation using cloud. The concept can be infer as any robot or automation system that relies on either data or code from a network to support its operation, i.e., where not all sensing, computation, and memory is integrated into a single standalone system [29]. This definition is intended to include future systems and many existing systems that involve networked tele-operation or networked groups of mobile robots such as UAVs [26], [27] or warehouse robots [24], [25] as well as advanced assembly lines, processing plants, and home automation systems, and systems with computation performed by humans [21], [22]. Due to network latency, variable quality of service, and downtime, Cloud Robot and Automation systems often include some capacity for local processing for low-latency responses and during periods where network access is unavailable or unreliable.
The Google self-driving car exemplifies the idea. It indexes maps and images collected and updated by satellite, Street view, and crowd sourcing from the Cloud to facilitate accurate localization. Another example is the Kiva Systems pallet robot for warehouse logistics. These robots communicate wirelessly with a local central server to coordinate routing and share updates on detected changes in the environment [29].
798
seemed to be one of the major issues in cloud robotics.Organizations are still not able to credence on cloud storage completely as it’s an open source and stores a huge data all together. To conceive a more secured cloud structure our system has introduced a level structured cloud cadre (LSCC) shown in fig.1
.
Fig.1 LSCC
In this level structured cloud cadre we will be using cloud hierarchy to provide a level wise security. A part from security this structure will also provide an easy way of segregated data according to the level of its use and availability.
II.
R
ELATED WORKCloud robotics and automation is not a new topic of discussion. The emergence of thought in the field started with the recognition of networking to connect machines in manufacturing automation system over 30 years ago. General Motors, in the 1980s developed the Manufacturing Automation Protocols (MAP) [20]. The first industrial robot was connected to the web was in 1994 with an intuitive graphical user interface that allowed visitors to tele-operate the robot via any Internet browser. Inaba [13] describes the advantages of remote computing for robot control under the name ‘ remote brained robot’. In 2009, the Roboearth a project that envisioned a ‘World Wide Web’ for robots, a giant network and database repository where robot can share information and learn from each other about their behaviour and environment [9] [10]. Systems have been proposed that address different aspects of the cloud robotics vision. Under a major European Union grant, a series of system architectures for service robotics by the Roboearth research team [2][3]. Some of them focus on remote sensor data processing [2] or on implementing computationally expensive algorithms in the cloud [1]. Other attempts were made to make existing web and cloud based resources available for the robots. In 2010 James kuffner introduced the term “Cloud Robotics”. A good
overview of cloud robotics is given by Goldberg [12]. The Ubiquitous Network Platform (UNR-PF) handles the task execution and supervision [18] on multiple robots and sensor devices at different locations. “Industry 4.0”, the term was introduced in Germany in 2011. It predicts an industrial revolution that will use networking [16]. In 2012, Industrial Internet was introduced to describe new effort where machines connect over to network to share data and processing [17] [19].
Many such related projects have shown up and many yet to come. In this paper we are not much focussing on any new robot system but we have tried to work on the existing systems to make it work better.
III.
ALOOF BRAIN
Robots are known for their capability of processing information using multiple processors. There are robots such as mobile robots or Unmanned Aerial Vehicles (UAV) has limitations in size and payload thus results in having limited capacities of on-board processing. Such robots can exploit the facilities of offloading the heavy computation task to an external computer. Cloud is capable of providing the robots this aloof or remote processing task allowing them to offload these heavy computations. Cloud provides remote computing by enabling massively parallel computation on demand. The European Union, Dream cloud Project is an achievement that enables dynamic resource allocation in many high performance and embedded system. It aims at enabling High Performance Computing (HPC) and cloud computing system to stabiles the workload and manages resources so that they can get a meaningful guarantee in terms of performances and energy.
Cloud computing is used as a platform for offloading and processing tasks remotely. The clouds permit running interactive, real time simulation tasks in parallel. It includes various tasks like image building using PCL (Point Cloud Libraries), video surveillance, sensor information tracking, evaluating performance evaluation tasks etc. Agostinho [54] created a cloud platform for robotics workflows using the Platform as a Service approach. Researchers have started more implication of these services in the scientific and technical HPC tasks.
Computing resources can be hired on rent for various computing task from public domain cloud service provider such as Amazon Web Services[55] Elastic Compute Cloud, known as EC2[56] , Google Compute Engine[57] , and Microsoft Azure[58]. The DAvinCi [59] project provided a platform that uses Map/Reduce in the cloud for solving FAST SLAM as parallelization. This helps robots to explore the chances of creating a global map by allowing the robots to interact with each other and share localization.
799
disruption in objects and environmental pose, shape and robotresponse to sensors and commands. Rapyuta proposed by Hunziker [31] in the Roboearth Cloud engine that allows heavy computing of expensive algorithms and their implementations to run in the cloud. Cloud Enabled Robotics System (CERS) [29] allows robots to form a local networked system. A distributed service framework for cloud based robot services is being presented by Jeeves. The basic mechanism of the framework is a robot assignment function, which determines the robot capability and accordingly assigns the proposed tasks to the end users to the suitable robot. In general cloud computing and robotics can interact and mesh up to minimize the various constraints of robots by offloading the task for the improved performance thus there come the realization of the Aloof Brain.
IV.
BIG DATA
Big data basically consists of data sets with sizes beyond the ability of commonly used software tools to 30Tcapture30T, 30Tcurate30T,
manage, and process data within a tolerable flow of time [35]. The huge calibrated data’s that cannot be managed, transformed, moved or analysed efficiently without using high performance computing resources, is termed as Big Data[36]. The Cloud has the capability to provide the robots and automation systems to an access of vast resources of data that are not possible to maintain in on-board memory. “Big Data” describes “data that exceeds the processing capacity of conventional database systems” [37] including images, video, maps, real-time network and financial transactions [38], and vast networks of sensors [39].
Autonomous robots are typically equipped with multiple sensors, cameras etc. Data collected from these sensors is normally used for taking decisions on localization, movement, path determination etc. for the robot. Sharing this data among multiple robots increases the volume to an extent that managing it would overwhelm limited capacities onboard a robot. Thus Cloud makes the Big data i.e. the enormous amount of data available on demand.
The Columbia Grasp dataset [11] has one of the large dataset of pre-computed hold on thousands of 3D models that have been used in many research studies [40]. Grasping is an interminable challenge in robotics: determining the optimal way to grasp a newly encountered object.
Fig.2 Google's object recognition system
Google Goggles [41], a free image recognition service for mobile devices (see Fig. 2), has been incorporated into a Cloud based system for robot grasping [42], as illustrated in Fig.3
Fig.3 System Architecture for Cloud-based object recognition for grasping.
It uses the network-based image recognition service for mobile devices. It has also been used as a service for robotics research. Various datasets have been reported in the literature specific to research topics. The Willow Garage Household Objects Database [43] is another dataset that contains thousands of 3D models that can be used to evaluate various aspects of grasping. The new advancement in the field is the RoboEarth [44].
It basically acts as a Cloud engine for the robots. The
Roboearth Cloud engine also called as RAPYUTAP P[45].
RAPYUTA allows Robots to have a powerful computation. Robots can offload their heavy computation to secure computing environment in the cloud with least configuration. Roboearth gets a high bandwidth access in the knowledge repository enabling robots to benefit from the experience of other robots.
Roboearth is a European Union’s project. It develops a web based knowledge which lets a robot to share information and obtain. Its main vision is to provide a World Wide Web for the robots. A web based Wikipedia acting as a platform for sharing knowledge about objects, actions and ambiance between robots. In simple words Roboearth is a knowledge base for the robots. All the information accumulated in the knowledge base is furnished with their requirements in terms of robot ability which are checked when downloading them. Using these requirement specifications, a robot is able to understand whether it can use the information as it is or it needs some extra information to utilize it.
V.
COLLECTIVE ROBOT LEARNING
800
opportunities in the field of robot learning [46]. Besides usingWWW for the faster communication, a fascinating opportunity is to allow robots to creatively and collaboratively update share knowledge repositories. Such knowledge base for robots allows robots to cope with the complexities of human environments and offer a simple though powerful way for long time robot learning. The cloud provides sharing of data for robot learning by harnessing data from many instances of physical trials and environment.
An example of collective learning influenced from a disaster recovery scenario where discrete mobile robots are sent to build a map of location. These robots interacting through a central cloud service, post the localization information that can be shared by the fellow robots which helps in enhancing the map discovery. Robots and automation systems can post initial and desired conditions, post control policies and trajectories, sensor information of physical traits of an environment, motion coordinates, tracking data and updated localization data for collective learning. Kiva Automated Material Handling Systems [47] is another practical example of collective learning. Here hundreds of mobile robots are used to move packages in warehouse using local networking system to coordinate motion, update tracking data and to have control on flow of traffic.
Fig.4 the Lightning path planning framework
The “Lightning” Framework (fig4) proposes a framework in which trajectories from many robots over many tasks and using Cloud computing for parallel planning and trajectory adjustment can be used for Collective Robot Learning [48]. For example consider grasping, grasping stability of the fingers in contact can be learned from the previous grasp on an object [49]. Capabilities of the robots can be improved with limited computational resources using sharing data through Collective Robot Learning. For the distributed task coordination and control The Ubiquitous Networked Robot platform (UNR-PF) framework was introduced by Kamei [50]. The framework deals with various tasks supervision on multiple robots and sensor devices at different locations. The UNR-PF takes away the concrete hardware from the robots and offers a generic interface which is used by the application developers to create hardware independent robotic services. Because of the UNR-PF devices and robots can be controlled remotely. The MyRobots project [51] [52] from Robot Shop presents a ‘social network’ for the robots where users can connect, monitor, and share robots. Collective learning
environment for ubiquitous robots is described by Chibani and Sandygulova [53]. Sensors embedded can provide large amount of information that can be beneficial in the future processing and collective learning.
VI.
PRELIMINARY
The existence of a robot world is an over-the-hill story but the trend of implementation of this concept is new. Many researchers and scientist have come out with so many new proposals of robot. Some were good to implement and some failed to show up. The different designs and its usage are making human life simpler and also increasing the efficiency of work [3]. Newly quite a few works have been reported in the literature bringing about the various benefits of using cloud computing in supplementing robotics and autonomous systems operations. Datasets for various images, maps etc. along with libraries and modules are being developed and made available on the Internet for research in cloud robotics. The cloud already houses programming constructs, libraries and frameworks that can help computers recognizing the type’s and classes of objects from images. The power of Big Data and Cloud Computing are extensively being used by many industries [28], [15], and are being harnessed to optimize water usage for irrigation [7]. Hunter et al. [23] presents algorithms for a Cloud-based transportation system called Mobile Millennium, which uses the GPS in cellular phones to gather traffic information, process it, and distribute it and also to collect and share data about noise levels and air quality (see Fig. 5).
Fig.5. Millennium, a Cloud-based transportation system that combines streaming data from taxis, maps, and road-based sensors.
801
functionality or cause damage [29]. ABE has shown its
promising future in fine-grained access control for outsourced sensitive data [32], [33], [34]. Typically, data are encrypted by the owner under a set of attributes. The parties accessing the data are assigned access structures by the owner and can decrypt the data only if the access structures match the data attributes.
To overcome with all these challenges we have focused more to provide a secured storage of data in cloud. Our system is divided into regions. The robot manufacturer belongs to these lower regions. The regional robot manufacturer will provide the local and regional information to the robot. It will also be provided with the regional cloud security. In the next region there comes the country region. Each country will be monitoring and connecting this information with other regional robots. It also provided an interface between the information sharing process with other countries. This level we can be called as mother region. This region checks the authentication of the robot controller. It also filters the nonessential and unwanted data from the incoming data source.
VII.
SECURITY REQUIREMENTS
In this paper, we endeavored to meet the following main security requirements for practical privacy-preserving of information stored in cloud.
1) Repository Privacy: Storage on the public cloud is subject to 4 privacy requirements.
a) Data Privacy: un-justified parties (e.g., public
cloud and outside attackers) should not get the hang of the content of stored data.
b) Anonymity: no particular user can be fraternized with the storage and retrieval process, i.e., these processes should be anonymous.
c) Decoupled Ability: unauthorized parties should
not be able to link multiple data files to profile a user. It indicates that the file identifiers should appear erratically and leak no useful information.
d) Keyword Privacy: the keyword used for search
should remain aplomb because it may contain sensitive information, which will hinder the public cloud from searching for the required data files.
2) Scrutiny: In emergency data access, the users may be physically unable to grant data access or without the perfect knowledge to decide if the data requester is a legal programmer. We require authorization to be fine grained and authorized parties’ access activities to leave crypto- graphic evidence.
VIII.
PROPOSED ALGORITHM
The coded information required to operate the robot has to be integrated by using models [5] [6] that provides effortless and secured accessibility of contents by authorized personnel. There are two things that needs to be considered in this model; security and accessibility. This model is based on attribute based encryption in which cipher text are labeled with a set of the Robots various attributes like country code, date of manufacture etc. Along with the attributes, the private key will be generated by using two languages that will correspond to the letters of the robots attribute. As encryption squander less resources than decryption, the process will be enhanced by using only a constant number of pairings to decrypt any cipher text. The data will be decrypted only if the cipher text attributes match with that of private key.
Setup (λ, U) → (Pk, Mk): The setup algorithm takes as input a security parameter λ and a universe description U, which defines the set of allowed attributes in the system. It outputs the public parameters Pk and the master secret key Mk. Encrypt (Pk, M, A) → CT. The encryption algorithm takes as input the public parameters PK, a message M and a set of attributes A and outputs a cipher text C associated with the attribute set.
KeyGen (Mk, S) → Sk: The key generation algorithm
takes as input the master secret key Mk and an access structure S and outputs a private key SK associated with the attributes.
Decrypt(Sk,C) → M. the decryption algorithm takes private key Sk associated with access structure S and cipher text C associated with attribute set A and outputs message M to indicate if A satisfies S else an error message will be given
LangChange (Pk, M, A): This algorithm takes public key PRkR, message M and attribute A and converts the attribute to
cipher text for private key generation using two local languages. Each letter of the attribute will be substituted by a letter taken from one of the two languages chosen earlier by the developer.
Substitution algorithm is as follows:
1. Consider English alphabets. From some language
L we will take 26 letters ZR1R ZR2R…ZRn.
2. For i=1 to 26 ORiR=ZRiR+7
If ORiR>26
Then ORiR=ORiR-26
Else Save ORi
KeyGen (MRkR, S, O) → SRkR.
The master secret key MRk, Rencrypted letter ORiR and an access
structure S are taken as input and outputs a private key SRkR
802
Decrypt (SRkR, C) → M.The decryption algorithm takes private key SRkR
associated with access structure S and cipher text C associated with attribute set A and outputs message M to indicate if A satisfies S else an error message will be given. Thus the performance will depend on the number of attributes involved.
XI.
DISCUSSION AND CONCLUSION
In this paper, we discussed how a system for knowledge enabled distributed task execution can be built on top of existing, common platforms in order to increase modularity, flexibility, and adaptability of robot applications. By assimilating the data into a hierarchical structure from different region the programmers can operate the robots more efficiently and can provide a more secured structure. Consequently separating generic functionality from environment-, domain- or task-specific knowledge, we intend to achieve a high degree of reusability as well as improved adaptability to novel situations. Different kinds of knowledge are explicitly represented and can be edited independently by the respective domain experts. Control decisions are formulated as inference tasks that are evaluated on the robot’s knowledge during execution. Provided with a level wise security makes the system more trustworthy. Controlling the system with respect to the manufacturer of the robot makes it easy to modify. This exceeds the grade of security and alleviates the complicacy of robot handling as each regional robot will be provided with different type of security and will be handled in different zone.
REFERENCES
[1] D. Hunziker, M. Gajamohan, M. Waibel, and R.
DAndrea, “Rapyuta: The roboearth cloud engine,” in IEEE International Conference on Robotics and Automation (ICRA), 2013.
[2] R. Arumugam, V. Enti, L. Bingbing, W. Xiaojun, K.
Baskaran, F.Kong, A. Kumar, K. Meng, and G. Kit, “DAvinCi: A cloud computing framework for service robots,” in Proc. Int. Conf. Robot. Autom. (ICRA), 2010, pp. 3084–3089.
[3] Z. Du, W. Yang, Y. Chen, X. Sun, X. Wang, and C. Xu,
“Design of a robot cloud center,” in Proc. Int. Symp. Auton. Decentralized Syst., Mar. 2011, pp. 269–275.
[4] C. Schmitt, “Security and privacy in the era of big data,”
2014. [Online]. Available: 30Thttp://www.renci.org/wpcontent/uploads/2014/
02/30T 0313WhitePaper-iRODS.pdf
[5] Sajjad Haider and Farhan Bashir Shaikh “Security
Threats in Cloud Computing” 2011 Dec
[6] Yue Tong, Jinyuan Sun, Sherman S. M. Chow, and Pan
Li,” Cloud-Assisted Mobile-Access of Health Data With Privacy and Auditability”, March 2014
[7] Droplet, “Cloud Powered Water Sprinkler System,” 2014.
[Online].Available: http://smartdroplet.com/
[8] K. Ren, C. Wang, and Q. Wang et al., “Security
challenges for the public cloud,” IEEE Internet Comput., vol. 16, no. 1, pp. 69–73, 2012.
[9] “What is RoboEarth?” [Online].
Available: 30Thttp://www.roboearth.org/what-is-roboearth30T
[10] M. Waibel, M. Beetz, J. Civera, R. D'Andrea, J. Elfring, D. Gálvez-López, K. Häussermann, R. Janssen, J. Montiel, A. Perzylo, B. Schießle,M. Tenorth,O. Zweigle, and R.DeMolengraft, “RoboEarth,” IEEE Robot. Autom. Mag., vol. 18, no. 2, pp. 69–82, Jun. 2011.
[11] C. Goldfeder, M. Ciocarlie, H. Dang, P. Allen, The
columbia grasp database, in: Robotics and Automation, 2009. ICRA ’09. IEEE International Conference on, 2009, pp. 1710–1716.
[12] K. Goldberg and B. Kehoe, “Cloud robotics and
automation: A survey of related work,” EECS Department, University of California, Berkeley, Tech. Rep. UCB/EECS-2013-5, Jan 2013.
[13] M. Inaba, “Remote-brained robots,” in Proc. Int. Joint
Conf. Artif. Intell., 1997, pp. 1593–1606.
[14] E.-J. Goh, “Secure indexes,” IACR Cryptology ePrint
Archive, vol. 2003, p. 216, 2003.
[15] McKinsey Inc., “Big data and the opportunities it creates for semiconductor players,” 2012. [Online].Available: 30Thttp://www.mckinsey.com/~/media/
McKinsey/dotcom/client_service/Semiconductors/Issue\ Autumnn\2012/PDFs/Big_data_and_the_opportunities_it _creates_for_semiconductor_players.ashx30T
[16] “Industry 4.0.” [Online].Available:
http://www.bmbf.de/en/19955.php
[17] P. C. Evans and M. Annunziata, Industrial Internet:
Pushing the boundaries of minds and machines General Electric, Tech. Rep., 2012.
[18] J. Furrer, K. Kamei, C. Sharma, T. Miyashita, and N.
Hagita, “Unrpf: An open-source platform for cloud networked robotic services,” in System Integration (SII), 2012 IEEE/SICE International Symposium on, dec. 2012, pp. 945 –950.
[19] J. Kelly, “The industrial internet and big data analytics: Opportunities and challenges,” 2013. [Online]. Available: http://wikibon.org/wiki/v/The Industrial Internet and Big Data Analytics%3A Opportunities and Challenges
[20] J. D. Irwin, the Industrial Electronics Handbook. Boca
803
[21]A. J. Quinn and B. B. Bederson, “Human computation: A
survey and taxonomy of a growing field,” in Proc. SIGCHI Conf. Human Factors Comput. Syst., 2011, ser. CHI '11, pp. 1403–1412.
[22]L. von Ahn, “Human computation,” Ph.D. dissertation,
Carnegie Mellon Univ., Pittsburgh, PA, USA, 2009.
[23] T.Hunter, T. Moldovan, M. Zaharia, S. Merzgui, J.
Ma,M. J. Franklin, P. Abbeel, and A. M. Bayen, “Scaling the mobile millennium system in the cloud,” in Proc. 2nd ACM Symp. Cloud Compu. (SOCC), Cascais, Portugal, 2011, pp. 28:1–28:8.
[24] R. D'Andrea, “Guest editorial: A revolution in the
warehouse: A Rretrospective on KIVA systems and the grand challenges ahead,” IEEE Trans. Autom. Sci. Eng. (T-ASE), vol. 9, no. 4, pp. 638–639, Oct. 2012.
[25] KIVA, “KIVA Systems,” 2014. [Online]. Available:
www.kivasystems.com
[26] V. Kumar and N. Michael, “Opportunities and challenges with autonomous micro aerial vehicles,” Int. J. Robot. Res. (IJRR), vol. 31, no. 11, pp. 1279–1291, 2012.
[27] Michael, D. Mellinger, Q. Lindsey, and V. Kumar, “The
GRASP multiple micro UAV testbed,” IEEE Robot. Autom. Mag., vol. 17, no. 3, pp. 56–65, 2010
[28] Accenture Inc., “A New Era for Energy Companies:
Cloud Computing Changes the Game.” 2012. [Online]. Available:
http://www.accenture.com/SiteCollectionDocuments/PDF
/Accenture-New-Era-Energy-Companies-Cloud-Computing-changes-Game. pdf
[29] Ben Kehoe, Sachin Patil, Pieter Abbeel, and Ken
Goldberg“A Survey of Research on Cloud Robotics and
Automation” IEEE journal November 06, 2014
[30]Wikipedia of robot and automation system
[31] D. Hunziker, M. Gajamohan, M. Waibel, R. D’Andrea,
Rapyuta: The roboearth cloud engine, in: Robotics and Automation (ICRA), 2013 IEEE International Conference on, 2013, pp. 438–444.
[32] M. Chase and S. S. M. Chow, “Improving privacy and
security in multiauthority attribute-based encryption,” in Proc. ACM Conf. Comput. Commun. Security, 2009, pp. 121–130.
[33] S. S. M. Chow, “New privacy-preserving architectures
for identity -/attribute-based encryption” Ph.D. dissertation, Courant Inst. Math. Sci., New York University, New York, NY, USA, 2010.
[34] S. Yu, C. Wang, K. Ren, and W. Lou, “Achieving
secure, scalable, and fine-grained data access control in cloud computing,” presented at the IEEE Conf. Comput. Commun., San Diego, CA, USA, Mar. 2010.
[35]9TSnijders, C.; Matzat, U.; Reips, U.-D. (2012). 9T30T"'Big Data':
Big gaps of knowledge in the field of Internet"9T30T.
International Journal of Internet Science 7: 1–5.
[36] F. J. Alexander, A. Hoisie, A. Szalay, Big data,
Computing in Science Engineering 13 (6) (2011) 10–13.
[37] E. Dumbill, “Defining Big Data,” [Online]. Available:
Forbes Data Driven Blog, 2014
[38] Q. Le,M. Ranzato, R. Monga,M. Devin,K.Chen, G.
Corrado, J.Dean, andA.Ng, “Building high-level features using large scale unsupervised learning,” in Proc. Int. Conf. Mach. Learn., Vancouver, BC, Canada, 2012, pp. 8595–8598.
[39]L. Wang, M. Liu, and M. Q.-H. Meng, “Real-time
multi-sensor data retrieval for cloud robotic systems,” IEEE Trans. Autom. Sci. Eng. (T-ASE): Special Issue on Cloud Robotics and Automation, to be published.
[40]C. Goldfeder, M. Ciocarlie, H. Dang, P. Allen, The
columbia grasp database, in: Robotics and Automation, 2009. ICRA ’09. IEEE International Conference on, 2009, pp. 1710–1716
[41] “Google Goggles.” [Online]. Available:
http://www.google.com/ mobile/goggles/
[42] B. Kehoe, A. Matsukawa, S. Candido, J. Kuffner, and K.
Goldberg, “Cloud-based robot grasping with the google object recognition engine,” in Proc. Int. Conf. Robot. Autom. (ICRA), Karlsruhe, Germany, 2013, pp. 4263– 4270.
[43]The willow garage, personal service robotics with tiered human-in-the-loop assistance (2013)
[44] The robo earth project. URL 30Thttp://roboearth.org/cloud
robotics/30T
[45]D. Hunziker, M. Gajamohan, M. Waibel, R. D’Andrea,
Rapyuta: The roboearth cloud engine, in: Robotics and Automation (ICRA), 2013 IEEE International Conference on, 2013, pp. 438–444.
[46] http://robohub.org/analysis-robot-learning-in-the-cloud/
[47] Kiva automated materials handling systems.
URL 30T
http://www.kivasystems.com/solutions/system-overview/30T
[48] D. Berenson, P. Abbeel, and K. Goldberg, “A robot path planning framework that learns from experience,” in Proc. Int. Conf. Robot. Autom. (ICRA), May 2012, pp. 3671–3678.
[49] H. Dang and P. K. Allen, “Learning grasp stability,” in Proc. Int. Conf. Robot. Autom. (ICRA), May 2012, pp. 2392–2397.
[50] K. Kamei, S. Nishio, N. Hagita, M. Sato, Cloud
networked robotics, Network, IEEE 26 (3) (2012) 28–34.
[51]MyRobots.com.” [Online].
804
[52] What is MyRobots?” [Online]. Available:
http://myrobots.com/wiki/About/myrobots.
[53] A. Chibani, Y. Amirat, S. Mohammed, E. Matson, N.
Hagita, M. Barreto, Ubiquitous robotics: Recent challenges and future trends, Robotics and Autonomous Systems 61 (11) (2013) 1162 – 1172, ubiquitous Robotics.
[54] L. Agostinho, L. Olivi, G. Feliciano, F. Paolieri, D.
Rodrigues, E. Cardozo, E. Guimaraes, A cloud computing environment for supporting networked robotics applications, in:(DASC), 2011 IEEE Ninth International Conference on, 2011, pp. 1110–1116.
[55] “Amazon Web Services.” [Online].
Available: 30Thttp://aws.amazon.com30T
[56] Amazon elastic compute cloud platform (ec2). URL
http://aws.amazon.com/ec2/
[57] “Google Compute Engine.” [Online].
Available: 30Thttps://cloud.google30T.
com/products/compute-engine
[58] Microsoft azure cloud platform.
URL 30Thttp://azure.microsoft.com/en-us/30T
[59] R. Arumugam, V. Enti, L. Bingbing, W. Xiaojun, K.
Baskaran, F. Kong, A. Kumar, K. Meng, and G. Kit, “DAvinCi: A cloud computing framework for service robots,” in Proc. Int. Conf. Robot. Autom. (ICRA), 2010, pp. 3084–3089.
AUTHORS PROFILE
Mr Saumyo Ghosh is mechanical engineer from JSPM’s RACOE affiliated from Savitribai phule pune University, Pune, India. He is very social in nature and presently a part of matrubhumi engineer pariwar.
Miss Aksha Mondal is pursuing her computer engineer from SKN Sinhgad Institute of
Technology and science. She is a part of
Microsoft student Paetners.
