(ICPS) A Cyber Physical System (CPS) is a mechanism controlled or monitored by computer-based algorithms, tightly integrated with internet and its users. It is an engineered system that are build from, and depend upon, the seamless integration of computational algorithms and physical components. In general Cyber means computation, communication, and control that are discrete and logical. Physical means natural and human-made systems governed by the laws of physics and operating in continuous time. Computing and communication systems bridges with the physical world are referred to as Cyber Physical Systems. CPS are physical and engineered systems whose operations are monitored, coordinated, controlled and integrated by a computing and communication core . Examples of CPS include medical devices and systems, aerospace systems, transportation vehicles and intelligent highways, defense system, robotics system, process control, factory automation, building and environmental control and smart spaces. CPS must interacts with the real world with proper security, safely and efficiently.
The term “Cyber Physical Systems” emerged around 2006, when it was coined by Helen Gill at the National Science Foundation in the United States. During World War II, Wiener pioneered technology for the automatic aiming and firing of anti-aircraft guns. Although the mechanisms he used did not involve digital computers, his control logic was effectively a computation, albeit one carried out with analog circuits and mechanical parts. Today, people are aiming to give an IoT backbone to CPS. This thought has emerged as a necessity driven concept. People imagined having automated cars which will have very less accident rates than human driven cars. The future road networks may also be connected with internet and therefore may reduce traffic congestion. These ideas for a better life gave birth to CPS.
In today’s context, CPS are emerging from the integration of embedded computing devices, smart objects, people and physical environments, which are normally tied by a communication infrastructure. These include systems like Smart Cities, Smart Grids, Smart Factories, Smart Buildings, Smart Houses and Smart Cars where every object is connected to every other object. They are aimed to provide an adaptive, resilient, efficient and cost effective scenario. Let us imagine the case where a road-accident patient is rushed to a hospital only to be asked to go to a police station first. If these systems are interconnected, then there will be very less chances of delay in the treatment. However, these objects should have a valid relationship in physical world too. For example, a traffic light can in no way be linked to a microwave. Connecting these two may overburden the data store as well as the network. So, we need a thoughtful connection between the objects in a CPS. CPS is like a reality to virtual mapping where the real physical world is linked to the virtual world of information processing through some sensors and actuators. The sensors are continuously spewing real time data. CPS is therefore a massive generator of data and needs real time processing.
These big words may sound like the futuristic wave of talking refrigerators and self-driving taxis. But, it means much more than that. CPS or anything related to smart objects is about devices, data and connectivity. Data – big and small – is the front and the center in the world of connected devices. The CPS as a whole along with its individual components and devices at its backbone will generate huge volumes of real time data. Storing these data for analytics may not always be feasible and immediately analyzing them will also be too difficult. Traditional analysis tools are not well suited to capture the complete essence of this massive data. The volume, velocity and variety is too large for comprehensive analysis and the range of potential correlations and relationships between disparate data sources are too great for any analyst to test all hypotheses and derive all the value buried in the data. A good machine learning systems in order to deal such Data require (i) data preparation capabilities (ii) algorithms – basic and advanced (iii) automation and iterative processes (iv) scalability (v) ensemble modeling and (vi) real-time decision making. This means we want the system to make all the decisions for us and take necessary actions quickly. For example, we want a smart hospital system to immediately send necessary information to a smart police station in case of road-accident. Machine learning algorithms already has some good capability of letting computers do the heavy thinking for us. But, we are striving for more to deal with large volumes of such data in a short time.
To capture such large scale concept, we will need millions of data generating smart objects. These will act like the building blocks of such a big network. Smart Objects (SOs) are Digital Entities (DEs), augmented with sensing, local processing, storing and networking capabilities. SOs may act as intelligent agents with some level of autonomy, cooperate with other agents and exchange information with human users and other computing devices within the CPS. DEs can be viewed as users in the context. A Physical Entity (PE) can be represented in the digital world by a Digital Entity which is in fact its Digital Proxy. There are many kinds of digital representations of PEs that we can imagine: 3D models, avatars, objects or instances; even a social network account could be viewed as such.
The combination of IoT technologies and Data Science and predictive technologies in any organization may at first seem like a “egg-chicken” dilemma. Those which do not have the infrastructure of IoT may have sophisticated algorithms for data analytics. For other companies who have adopted IoT technologies with the ultimate goal of optimizing physical processes (CPS) or providing predictive analytic solutions still have an opportunity to use data and analytics to advance their business. In some industries like fleet management, the useful life of the data may be too short to justify keeping it. In others, like broadcasting, files may be too large to store for long periods. Microsoft research has found that about 23 percent of enterprises use data for basic reporting functions – and that seems to be a good start for the optimists. Keeping data on hand leaves room for more sophisticated analytics later, as business needs and capabilities grow. Typically, the progression of data analysis mirrors the same evolution we see from our IoT customers, from reactive to proactive and ultimately to predictive analytics using machine learning.
So, as a concluding remarks, CPS systems are generally built with IOT devices, data from IOT devices are processed through in-network computation or in cloud, which result in actuation - actuation is directed by the control scheme. Hence research issues of CPS system are design of IOT platforms, machine learning and intelligence in IOT devices, security for data exchanges, analytics and machine learning on cloud; appliance based social sensing, Hybrid and latency sensitive control scheme, web-services architecture. So CPS scheme can have more cohesive research agenda defined across different subspaces. In a wider perspective, CPS are integrations of computation, communication and physical processes. Embedded computers, networks monitor and control the physical processes, with feedback loops where physical processes affect computations and vice versa. The Data Science, IoT and cyber security are fundamental and basic pillars on which CPS are normally built. The economic and societal potential of such systems is vastly greater than what has been realized, and major investments are being made worldwide to develop the technology. Despite the fact that the drivers for CPS come from different sectors, the technology gaps in the sectors stem from a common set of fundamental challenges . The key cross-cutting platform technologies needed to overcome these challenges and accelerate the development of CPS applications in all sectors.
1. General characteristics of a Cyber Physical Systems
a) Cyber capability in every physical component.
b) Networked at multiple and extreme scales.
c) Dynamically reorganizing/re-configuring termed as open systems.
d) Constituent elements are coupled logically and physically.
e) High degrees of automation, control loops closed at many scales.
2. Research challenges in CPS: The large-scale nature of CPS raises a number of specific challenges ranging from system-level management and control to data analytics.
a) Robustness, Safety and Security of CPS.
b) Controlling Hybrid Systems: Apply to hierarchies of asynchronous dynamics at different time scale(from months to microsecond) and geographic scope (from on-chip to planetary scale).
c) Computational Abstractions: Physical properties such as laws of physics and chemistry, safety, resources, real time power constrained etc must be captured by programming abstractions.
d) Sensor and mobile networks: Essential requirement to increase system autonomy in practice requires self-organization of mobile and Adhoc CPS networks.
e) System-level challenges include novel scalable methods for global system control, effective development of large-scale management platforms, well-defined control interfaces for IoT technologies and various IoT standards.
f) Data related challenges include effective data collection, cleaning and storage, data latency and real-time analytics. IoT and mobility are driving more data into enterprises and Big Data Analytics has become an essential component for extracting value from these data.
g) Abstractions, modularity and composability – to enable CPS system elements to be combined and reused while retaining safety, security and reliability.
h) Systems engineering based architectures and standards – to enable efficient design and development of reliability systems while ensuring interoperability and integration with legacy systems
i) Adaptive and predictive hierarchical hybrid control – to achieve tightly coordinated and synchronized actions and interactions in systems that are intrinsically synchronous, distributed and noisy
j) Integration of multi-physical models and models of software–to enable co-design of physically engineered and computational elements with predictable system behaviors
k) Distributed sensing, communications and perception – to enable flexible, reliable and high performance distributed networks of CPS that provide an accurate and reliable model of the world and enable time - aware and time-critical functionality
l) Diagnostics and prognostics – to identify, predict and prevent or recover from faults in complex systems
m) CPS Security – to guarantee safety by guarding against malicious attacks on CPS systems
n) Validation, verification, and certification – to speed up the design cycle for bringing innovations to market while ensuring high confidence in system safety and functionality.
o) Autonomy and human interaction –to develop models of autonomous CPS systems and humans interacting with them to facilitate model-based design of reactive systems that are used by humans.
p) Interoperable System of Systems:The ability of CPSs to interoperate across systems in complex tasks and environments is a key challenge. Additionally, approaches to modularity and composability that enable reliable and verifiable assembly of individual CPSs into interacting systems of systems are needed.
q) Privacy issues: CPS technologies and techniques that enhance privacy and enable the appropriate use of sensitive and personal information while protecting personal privacy are needed.
r) Collaborations, Innovation and Entrepreneurship: Addressing the R&D gaps will require close collaborations between industry, R&D systems/Academics/ University and Government. Private Public Partnerships are expected to play a central role in bringing these stakeholders together. Such partnerships should be designed to optimize the movement of people across sectors and intellectual capital flow across organizational boundaries and be structured to ensure that IP concerns do not impede progress, while promoting industrial competitiveness. Models for strategic management of Intellectual Property Rights (IPR) will need to be developed by each agency consistent with National Policy. There should be mechanisms to reinvigorate Human Resource by including education and training in the mix of engineering and computer science skills that are critical to CPS.
s) Verification, Validation and Certification of CPS: Compositional verification and testing methods that explore the heterogeneous nature of CPS models are essential. Biomedical and Healthcare Systems: PS research is revealing numerous opportunities and challenges in medicine and biomedical engineering. These include intelligent operating rooms and hospitals, image guided surgery and therapy, fluid flow control for medicine and biological assays, and the development of physical and neural prostheses. Standardized abstractions and architectures that permit modular design and development of cyber physical systems are urgently needed. CPS applications involve components that interact through a complex physical environment. To achieve Reliability and security we need new frameworks, efficient tools and algorithms.
3 Broad areas of research under ICPS:
1) Theoretical foundations of CPS
1. Modeling, Analysis and Synthesis Techniques
2. Architectures for CPS
3. Building blocks for CPS
4. Systems Abstractions, Services and OS Support
5. Evaluation approaches and metrics
6. Novel CPS applications
7. Ubiquitous and pervasive computing for enhanced user interactions with CPS
8. Control of CPS
9. Mobile computing and devices for CPS
10. Cloud computing and distributed systems to support scalability and manage complexity of CPS
11. Analysis, verification, and synthesis of hybrid systems
12. Data Science & Technologies for CPS
13. Simulation of CPS applications
14. Security and privacy of CPS
15. Networking systems for CPS applications
16. Experimental prototypes of CPS
17. Use case and user study of CPS
18. Sensors and actuators for CPS applications
19. Cyber-physical multimedia systems and applications
20. Wearable cyber-physical systems and applications
21. Emerging applications in CPS, including social space, crowd sourcing, art, healthcare and human computer interactions
2) Application of CPS in sector specific areas and research: Being a developing country, India can be one of the best candidates for research in IoT and CPS on the following issues.
a) A smart city concept has been initiated by the Prime Minister of India. The core infrastructure elements in an Indian smart city would include:
• adequate water supply
• assured electricity supply
• sanitation, including solid waste management
• efficient urban mobility and public transport
• affordable housing, especially for the poor
• robust IT connectivity and digitalization
• good governance, especially e-Governance and citizen participation
• sustainable environment
• safety and security of citizens, particularly women, children and the elderly
• health and education
b) Smart water management systems would regulate water supply depending on the demand in the area. Wastage of water could be monitored and reduced.
c) Smart electricity connections would ensure energy conservation in urban area, and help light up remote rural places.
d) Indian population is a huge concern in today’s scenario. This leads to numerous problems; the number of vehicles on the roads are increasing more than what can be handled. This in turn leads to the traffic jams during the busy hours that cause delays especially in case of emergencies. Such congestion along with faulty driving cause serious road accidents in the country every minute and 16 die every hour. Over 1,37,000 people were killed in road accidents in 2013 alone, that is more than the number of people killed in all our wars put together. Automated cars and smart road management networks can help this regulate traffic and also reduce accidents.
e) Medical facilities could be given immediately to emergency case patients if there could be some far visioned smart ambulances; and smart hospitals.
f) Public safety demands for an interconnected smart surveillance cameras that will send alerts to nearby police stations if they sense any crime.
g) There can be some SOS alarm triggers at the roads in case of emergency.
h) Illegal wildlife trade in India has also increased. If there are some sensors which will send continuous data about the animals and alert the forest department in case of any ambiguity, then poaching can be prevented.
i) Swachh Bharat Abhiyan (Clean India Mission) launched by the Prime Minister in 2014 aims to make the surroundings clean and better for the citizens. The unnecessary littering on streets can be prevented by smart garbage collection methods; thereby making clean cities.
j) Agriculture: CPS technologies will play a key role in helping to increase efficiency throughout the value chain, improving environmental footprint and creating opportunities for a skilled and semi-skilled workforce.
k) Infrastructure Management: Infrastructure of a country, city or area refers to the structures, systems, and facilities necessary for the economy to function. It typically includes roads, bridges, tunnels, railways, water supply, sewers, electrical grids, communication network, power plants, industrial units and so forth. Hence, infrastructure is defined as "the physical components of interrelated systems providing commodities and services essential to enable, sustain, or enhance societal living conditions”. In the context of India, technological intervention in the management of infrastructure is critical for national economy. A key element of infrastructure management is the challenge of maintenance of existing structures some of which may be 100 years old. Lack of maintenance action at right time and right place can lead to accidents and disruption of essential services leading to uncalled for financial losses. IOT based CPS can provide the key technology for infrastructure management. It can be used to provide technology for condition monitoring and predictive maintenance of infrastructural elements (e.g. Bridges, Roads & railway network) involving electronic sensing, domain models and data analytics. Distinctive nature of physical model of each infrastructural element introduces specific challenges in designing intervention models for predictive and preventive maintenance. Involvement of electronic sensing, physical modeling of the infrastructure element and machine learning based data analytics require an integrated interdisciplinary approach for development and deployment of CPS for predictive maintenance of critical infrastructure components.
l) Internal & External security: Complex, networked systems are increasingly critical for meeting our national security needs. In these, CPS Science & Technology plays a fundamental role. The broad areas in which CPS can play role is expediting design and delivery of trustworthy, adaptable and affordable systems, operations in cyberspace and autonomous systems to augment security operations.
m) Disaster Management: CPS technologies including next generation public safety communications, sensor networks, and response robotics can dramatically increase the situational awareness of emergency responders and enable optimized response through all phases of disaster events such as earthquakes, floods and bombing attacks etc. Cyber enabled systems uniquely facilitate resilience, robustness and flexibility
n) Energy: Clean renewable energy resources such as solar, wind and tidal are expected to advance significantly due to environmental issues. The integration of intermittent and uncertain wind and solar sources and plug - in devices necessitates not only - new sensors, switches and meters, but also a smart infrastructure for realizing a smart grid - an adaptive, resilient, efficient, and cost - effective electricity distribution system. CPS technologies are essential for the creation of this infrastructure, enabling the optimization and management of resources and facilities and allowing consumers to control and manage their energy consumption patterns.
o) Healthcare: The healthcare challenges arising from our ever growing population combined with the opportunities provided by inexpensive sensing, ubiquitous communication and computation and the demand for 24/7 care will lead to an explosion of cyber-physical medical products. CPS correct-by-construction design methodologies are needed to design cost-effective, easy-to-certify, and safe products.
p) Manufacturing and Industry: The complexity of what the industry are able to design and build, and what consumer society wants is constantly increasing, while the time scale for products and the lead time are decreasing, even as product variety is increasing. CPS technologies are vital to increase national competitiveness in manufacturing. CPS enables the convergence of the global industrial system with the power of advanced computing, analytics, low-cost sensing, and new levels of connectivity permitted by the Internet, and the new paradigm called the Industrial Internet is emerging. Predictive maintenance models can be developed for plant machinery based on sensors on IOT platforms. This data from IOT platforms can be analysed at the Cloud so that higher availability of plant machinery can be ensured by detecting problems before they result in a major machine malfunction or breakdown.
q) Transportation: There is an urgent need to improve efficiency and safety in public transportation systems. CPS technologies can (potentially) eliminate accidents caused by human error, Congestion control, traffic based grid jams it includes road, air and highway networks. CPS has immense potential to address some of transport related issues in most of metro cities.
r) Cyber Society: Cyber Space that seamlessly network with the physical world are gradually becoming part of societies across the globe. Consumers increasingly want special purpose software for automating tasks and expect interoperability across different hardware devices. Investments in basic research are needed to create the foundations for innovation and jobs growth across the commercial applications sector. Success in deploying CPS requires leveraging the interaction between people and technology, and between existing complex infrastructures and human behavior. It is important to approach such systems from a sociotechnical perspective, to ensure that CPS satisfy the needs, want, and aspirations of the stakeholders in a positive economic, social and environmental manner.