Over the years, the internet revolution has redefined how we do business today. This redefinition becomes even more relevant when we understand that by 2020, the number of internet connected devices would leapfrog and exceed by more than 50 billion units.
This is where Industry 4.0 comes in, i.e. evolving a ‘Smart Factory’ which will host automation and data exchange within the factory and to the outside world. In addition to IIoT or the Industrial Internet of Things, the ‘Smart Factory’ will encompass cyber physical systems and cloud computing. Further, robots will be playing an extremely important role in the coming years in helping humans to accomplish many tasks. In addition to the current industrial tasks being performed by robots like operator assistance, industrial assembly, etc, robots will also contribute significantly in rescue management systems, military applications and health care.
So, where do robots fit in the ‘Smart Factory’ approach? In fact, IIoT for robots has already been in vogue for many years now. The current research and application trends are promoting and nurturing this trend further and are leading to increased evolution of internet usage in robotics and thus, IIoT aided robotics applications. The current ecosystem provides sufficient scope for IIoT aided robotics applications to grow further, leading to a situation where humans, robots and IIoT nodes will interact & work on a co-operative basis.
IIoT of robots is not all hype and real investments are being done or are committed with IoT-ready robots, which will be soon available in the market. Most of the major robot manufacturers are having their strategy in place on how to address this trend now and for the future.
Internet of Robotic Things (IoRT)
The Internet of Robotics Things (IoRT) is the synergy between IIoT network & robots. Robots can monitor events, fuse sensor data from a variety of sources and use local and distributed intelligence to determine the best course of action. The robots can then act to control or manipulate objects in the physical world.
Networking architecture for IIoT
In a factory setting, the industrial ‘machine-to-machine’ (M2M) networks have been widely used to connect equipment across the floor to enable discrete automated processes. Since the 80’s, PLCs have used industrial protocols (like EtherNet/IP, PROFINET I/O, etc) to establish a line of communication from one machine to another by capturing and sending data from a central hub. Every device uses one or a set of proprietary protocols, so connecting one device to another means finding devices that speak the same language.
For instance, in order to connect to an EtherNet/IP network, the device (like a barcode reader, robot, or PLC) must be able to communicate over the EtherNet/IP protocol. Code developed to support these systems, data collected by them, and the interfaces that access them are unique to each particular system and therefore are not easily usable in other systems. Firstly, this is an obstacle to data sharing across the factory, since devices used for different purposes don’t always speak the same language. Secondly, these industrial devices do not speak the language of the internet. Data sharing over an industrial protocol like EtherNet/IP is largely limited to the local network and devices that use it, impeding access over the global internet for a true implementation of IIoT. The other overwhelming challenge is data security issues.
Current progress in resolving challenges
The European Union has conceived several research projects under FP7 (seventh framework programme for research) to address the issues related to Network Architecture. There are also other projects which are focused on cloud computing technologies, security issues, context aware approaches & semantic oriented design. To address the issue of interoperability, the European Telecommunications Standard Institute has defined a set of specifications that provide a restful architecture to standardise ways in which heterogeneous devices can offer services and be accessed seamlessly.
Some of the research projects which are a work in progress are PROBE-IT, IoT-I, ELLIOT, CALIPSO, etc. The technologies, which are being worked on through these research projects, are in different areas such as middleware, cloud, IPv6, user cognitive model, security & privacy, context awareness, resource abstraction & information centric networking.
Major gains from research projects
The most important gain from on-going research projects is real-time data acquisition and ability to respond in timely and adaptive ways to market demands. Businesses will start adapting more pervasive web-based technologies. All devices within a factory will start to seamlessly communicate with each other about quantity of material entering the line, getting rejected, etc., thus providing data to factory ERP system for inventory planning.
We will thus have Big Data from the outside world entering the factory to drive factory working. The current topology of the factory (leveraging industrial protocols like Ethernet/PROFINET) will change to internet based built on open standards where devices from multiple manufacturers will speak without barriers. Thus, the evolution of Cloud Robotics has emerged. Currently, the RoboEarth project does aim to develop a world wide web for robots – a giant network and data repository.
Modifications in network architecture
The new scenario will demand modifications in the current network architecture. Traditional industrial platforms like PLCs will be replaced by technologies that break free of device operating system or protocol dependency. Local networks will be replaced by cloud computing. Software will be replaced by web services and devices that use proprietary advantage to bind factories, which will be replaced by devices with internet protocols.
Does this mean that we have to junk our current factories? No. Businesses with existing networks using traditional Ethernet/IP sensors or PROFINET control devices are likely not looking to replace their entire factory floor with HTTP devices to achieve Internet compatibility. The investment in time and energy for such a migration would be akin to rebuilding the factory from scratch. Innovative industrial system manufacturers understand this and are engineering a number of new platforms that bring devices with and without HTTP communication capability onto unifying web services. This is a great opportunity for companies who aspire to be in this space and presents immense business potential in years to come for companies who will support undertaking modifications in the network architecture.
Data security is a critical issue that prevents the widespread adoption of IIoT technologies and applications. The following questions come up when IIoT is discussed – how to guarantee data confidentiality in an environment where even mobile objects are authorised to access data, how to model all IIoT fundamental entities and their relationships to cope with privacy issues and lastly, how to manage trust in a dynamic and flexible environment without established relationships among actors?
One of the best solutions for data security issues is opting for Role Based Access Control (RBAC). RBAC is a method of regulating access to computer or network resources based on the roles of individual users within an enterprise. In this context, access is the ability of an individual user to perform a specific task, such as view, create, or modify a file. The main advantage of RBAC in an IIoT perspective is that access rights can be dynamically modified by changing the role assignments. However, the IIoT context will require the introduction of new forms of RBAC-style solutions, in particular considering that IIoT data will likely represent data streams to be accessed in real-time, rather than being stored in static databases.
There are two categories of data security which need to be ensured – aiming at ensuring authenticity, confidentiality and integrity of the data streams during transmission & aiming at access control. Though mechanisms to guard against unauthorised access to streaming data are being devised, they are still a work in progress. The main challenge is to find solutions for handling the identity of robots and their related authorisation processes in a secure manner. The other fundamental IIoT research challenge is represented by privacy issues. The main reason that makes privacy a fundamental IIoT requirement lies in the envisioned IIoT-aided robotic application domain and in the adopted wireless technologies. Currently, limited solutions are available and much of the work is still work in progress.
Action steps for implementing IIoT
A true implementation of the IIoT connects the factory with other factories, suppliers, or customers, enabling the factory not only to meet today’s goals but to also predict tomorrow’s challenges. The major benefits from IIoT can be listed as follows:
Robots will anticipate opportunities and failures based on external events and plan production accordingly
Robots will perform self-monitoring and reporting
Robots will not wait for scheduled maintenance but have a maintenance run when they want it at the perfect time
Improved reliability of systems, optimised processes, reduced downtime and manufacturing flexibility