What do the Cyber Physical Systems (CPS) relate to?

In simplest terms, Cyber Physical System (CPS) refers to a platform comprising of a mechanical system that is controlled by computer algorithms, and tightly integrated with Internet and to its networked users. Here, the physical-mechanical components represented by smart sensors and actuators, and the software components represented by computing and networking devices are intrinsically intertwined in the platform.

Thus the building of CPS truly involves multidisciplinary fields advocating the merging of the theories of ‘mechatronics’ and ‘cybernetics’. And, therefore, it truly represents a new family of systems.

And in its implementation the mechatronic devices are augmented with spot sensor monitoring and communication capability, and their operation is monitored, coordinated and controlled by the software program that runs on the connected computing devices.

Evolution of CPS:

The factors that have led to the birth to the new disciplines, like that of CPS are;

• 21st century is seen as a water shed in technology development and innovation that have touched human lives in many ways. It seeks to  give birth to the new, more efficient, cost effective and secure systems, quite apart from the traditional systems that have existed for long.
•  Continuing research is being conducted to seek elegant but robust solutions to the needs of today’s vibrant economy and complex society.
• Technology is playing a pivotal role in meeting the requirements of our growing social and business settings.
•  There has been a continuing focus on industrial automation towards improving the efficiency of manufacturing and cutting costs, and rendering an effective consumer friendly digital supply-chain. 
• Leading developments of ingress of computing hardware and software into mechanical systems, like embedded devices, edge computing, ubiquitous computing, smart sensors and intelligent streamed bandwidth communication device technologies.

From the usability perspective, the technologies as the drivers for pushing this evolution of CPS are;

• Advances in the technologies of smart sensors, miniaturization, and making them more intelligent.
• Evolution of IoT and IIoT as data generating devices integrated with a range of physical systems.
• Use of embedded computers to speed up the processing at source.
• Evolution of edge computing to bring computing closer to data and the users where it is to be exploited.
• Ubiquitous computing enabling, however small, computing to take place at all key touch points in the process, duly supported by internet, middleware, microprocessors, mobile codes, I/O and user interfaces.
• Availability of smart mobiles with high bandwidth capabilities, advanced user interfaces, and integration with location and RFID and such other devices, and supporting a host of Web applications.
• Construction of robots for motion manipulation, performing critical tasks, and those requiring precision.
•  Emergence of broadband and secure networks.
• Fast and efficient processing with high storage capacities in user end computers. 
• Clouds that enable offering cost effective, trust worthy and secure  services using the computing resources, data and software tools.

Current speed of developments with the availability of these and relevant technologies are therefore expected to enhance CPS functionality, adaptability, scalability, security, and usability. As forecast therefore, CPS is poised to transform the way people will interact with engineering systems just as Internet has transformed the way people interact with Information.

The next stage of developments, therefore, that are further serving as the driving force for CPS introduction are:

•  Emergence of 5C Architecture as a pyramid to develop and deploy CPS, viz., connection, conversion, cyber, cognition, and configuration.
•  Mobile CPS due to emergence of smart phone platforms for an effective link between physical-mechanical and computational elements.
• CPS empowered by clouds paving the way to Industrie 4.0 standard, serving a host of societal applications, and digital supply chain.
• Integration of AI with CPS creating new opportunities with major social and business applications.

Popular Usage scenarios of CPS:

Gauging from the current status of developments, CPS is fast assuming reality of existence. The potential usage scenarios, as demand pullers, are:

• Process control systems using embedded computing devices. Examples are in pharmaceutical, chemical, textile manufacturing.
• Electrical power management using smart grids involving automated and equitable power distribution among the various users based on a given criteria.
• Autonomous vehicles incorporating smart sensors, cameras,  actuators, intelligent data analysis and motion manipulation.
• Robots for motion manipulation for applications in healthcare, hazardous environments, precision manufacturing etc.
• IoT- points of data pick up over the cloud, data analytics and resultant outputs represented as dash boards and 3D visualization in a range of application environments.
• End- to- end automated manufacturing line; example Automotives, with monitoring of assembly processes, assessing tool fatigue, and sounding an alarm before occurrence of a fault.
• Environmental monitoring – sensor networks - for collating data and performing analytics to decipher intrinsic values with representation in dashboard and 3D visualization forms.
• Supply – chain with edge computing at various touch points to deliver efficient customer services.

The sectoral application areas where CPS is a potential candidate for effective application of its concept are mentioned below. Even as these applications are still in infancy, experience with the prototype developments shows promise for full scale realization in the next five years or so.

• Aerospace Engineering
• Chemical Processes flows
• Pharmaceutical industry
• Medical imaging and diagnostics
•  Farming and Agriculture
• Water distribution management
•  Intelligent Manufacturing
•  Engineering Services
• Transportation systems
•  Automotive Industries
•  Emergency response as mitigation services
•  Building controls – smart buildings/cities/hospitals/homes
•  Military and Defense systems
•  Energy Systems

Potentials and Challenges with CPS:

The economic and social potential of CPS is adjudged to be vastly greater than what has been realized so far. US National Science Foundation has marked it as an area of intense applied research. China and Japan notably have programs investing huge sums in its developments. In Germany it is emerging as interdisciplinary science & technology research for industrial automation. Major Investments being made are in smart manufacturing empowered by clouds. It is leading next computing revolution from mainframe computing to desk top computing to edge computing and ubiquitous computing.

However, the challenges encountered in development and deployment of CPS, as currently seen, are;

• Design principles of software and mechanical engineering are vastly different due to difference in architecture- spatial & temporal scales. Their integration, therefore, poses a challenge.
• Building trust in CPS is a major societal challenge as a full scale implementation of CPS is still to give sufficient experiential data.
• Prevention of accidents in physical systems is an important aspect- some of the past examples of famous accidents being Patriot, Ariene, Pathfinder, Airport Baggage handling, London Ambulance.
• Security and privacy are the relevant issues as a significant amount of data is exchanged across multiple system elements in CPS.
• Challenges in smart cities implementation - data heterogeneity, data cleaning/management, security and safety, privacy and trust are the major issues being faced.
• Integration of social, physical, enterprise & biological systems, as the main sub sets of the emerging standard Industrie 4.0, continues to be a challenge.

Research efforts are underway to deal with these and related challenges. Core research areas are in;

• Unified architecture of integrated mechanical & computational system.
•  Data analytics and deep learning, security and information management, building useable autonomous devices, IoT connected to clouds and the like.
• Information intensive transformation of manufacturing in a connected environment of data, processes, people, services, systems with a targeted improvement in productivity, minimization of failures and savings in costs.
•  Principles of AI applied to CPS in a way to reap the benefits of both the fields.

Outlook:

The tangibles of CPS deployments, as reported, indicate;

• Increasing productivity and efficiency by as much as 30%.
• Upto 10% savings through early detection of faults.
•  Reduction in pilferage, as for example, in petroleum products distribution.
• Optimizing on assets, particularly with the organizations having multiple layers of business activities.
• Implementation is within reach as software constitutes upto 70 % of the development cost.
•  Employment potential of new skills with organizations dealing with process control going to next echelon of information intensive smart controls, AI applications to engineering systems, business of embedded systems, digital supply chain interfacing with R&D, presales, manufacturing, marketing, sales and logistics, technology platforms of cloud, big data, IoT, data analytics, mobility induced smart devices and research organizations working with focus on creating Industrie 4.0 standard systems of future.

As a global market forecast of CPS, a growth of 10% is predicted from a current level of US $ 80 Billion, but on another more optimistic note a Euro 350 Billion revenue is predicted to reach in next five years. Technology push on one hand and demand pull on the other are seemingly the main contributing factors for this, as presented in this paper. 

1. https://en.Wikipedia.org/wki/cyber-physical
2. www.nist.gov/el/cyber-physical-systems
3. www.quora.com/what-is-a-cyber-physical-system
4. www.ceeri.res.in/development/cyber-physical-system