1.1 Introduction

1.1.1 Brief Introduction to Networked Control System

The advent of communication networks introduced the concept of remotely controlling a system which gave birth to Networked Control Systems (NCSs). The classical definition of NCSs [1] can be as follows: when a traditional feedback control system is closed via a communication channel, which may be shared with other nodes outside the control system, then the control system is called a Network Control System (NCS). An NCS can also be defined as a feedback control system wherein the control loops are closed through a real-time network [2, 3]. The defining feature of NCS is that information (reference input, plant output, control input, etc.) is exchanged using a communication network among control system components (sensors, controllers, actuators, etc.). The conceptual model of NCS is shown in Fig. 1.1 [2]. The networked medium can be wired or wireless depending on the type of the applications. In NCS, when any form of data that is transmitted through wires, then such medium is called as wired network, while any form of data that is transmitted without use of electrical conductor, then such medium is called as wireless network medium. The main advantage of using wired medium is data security. However, the main advantage of using the wireless network medium is to get rid of wires. In NCS, the wired communication is carried out through CAN, Switched Ethernet, Ethernet, Profibus and Profinet networked medium, while wireless communication is done through wireless LAN, wireless PAN, wireless MAN or wireless WAN [3].

Fig. 1.1
figure 1

Conceptual model of NCS [2]

Full size image

1.1.2 Advantages and Applications of Networked Control System

For many years now, data networking technologies have been widely applied in industrial and military control applications. These applications include manufacturing plants, automobiles and aircraft. Connecting the control system components in these applications such as sensors, controllers and actuators via a network can effectively reduce the complexity of systems with nominal economical investments. Furthermore, network controllers allow data to be shared efficiently without bulk wiring. It also allows easily to add more sensors, actuators and controllers with very little cost and without heavy structural changes to the whole system. Most importantly they connect cyberspace to physical space making task execution from a distance easy. These systems are becoming more realizable today and have a lot of potential applications including space explorations, terrestrial exploration, factory automation, remote diagnostics and troubleshooting, hazardous environments, experimental facilities, domestic robots, automobiles, aircraft, manufacturing plant monitoring, nursing homes or hospitals, telerobotics, smart grid.

Fig. 1.2
figure 2

Shared structure of NCS [2]

Full size image

1.1.3 Structure of Networked Control System

In general, there are two major types of control systems that utilize communication networks: (1) shared network control system and (2) remote control system. Figure 1.2 shows the architecture of shared network control system [2]. It can be noticed that with shared network control system, the transfer of information from sensors to controllers and control signals from controllers to actuators greatly reduces the complexity of connections and provides more flexibility in installation, ease of maintenance and troubleshooting. Moreover, it also provides the communication among control loops [2, 4, 5]. This feature is extremely useful when a control loop exchanges information with other control loops to perform more sophisticated controls, such as fault accommodation and control. Similar structures for network-based control have been applied to automobiles and industrial plants. The other control system that utilizes the network medium is remote control system. In remote control system, the place where central controller is installed is called a local site and the place where plant is installed is called a remote site. The data transfer between local site and remote site is carried out through communication network. Sometimes the remote control system is also defined as teleoperation control system. There are two general approaches to design an NCS using remote control system: (i) hierarchical structure and (ii) direct structure. In hierarchical structure, there are several subsystems that are connected to central controller through communication network. Each subsystem contains sensor, actuator and controller by itself as depicted in Fig. 1.3 [2]. In this case, a subsystem controller receives a set point from the central controller. The subsystem then tries to satisfy this set point by itself. The sensor data or status signal is transmitted back via network to the central controller. The block diagram of direct structure is shown in Fig. 1.4 [2]. In this case, the sensor and actuator are attached to a plant, while a controller is separated from the plant by a network connection. The sensor transmits the signal to the controller through the network medium, and controller sends back the processed control signal to the plant via actuator through the network medium. This type of configuration is used mainly in the process industries and haptic surgery. Many complex network control systems use the combination of both the structures known as hybrid structure.

Fig. 1.3
figure 3

Hierarchical structure of NCS

Full size image
Fig. 1.4
figure 4

Direct structure of NCS

Full size image

1.1.4 Concerns in Networked Control System

The presence of communication medium in Networked Control System leads to several natural issues such as:

  • Time delay: The time required for the data to travel within the network is defined as time delay. The nature of time delay depends on the various factors such as network configuration, distance of communication between plant and controller, baud rate, network characteristics and network topology. The time delay can affect the performance of the system in all the structure of NCSs (shared, hierarchical and direct).

  • Packet Loss: Whenever the data transmitted from sensor or controller through the network and fails to reach the destination, then such condition is defined as packet loss condition. The packet loss is mainly caused due to congestion, network traffic and jitter problems. There are two types of packet loss: (i) single packet loss and (ii) multiple packet loss. The packet loss situation occurs in all the structure of NCSs (shared, hierarchical and direct).

  • Packet disorder: The packet disorder issue is generally caused in wireless NCS due to heavy traffic, congestion or jitter. In wireless NCS, the communications take place in the form of small packets. So in order to have secure communication, each packet is provided with a unique identification number in the header. During transmission, if any packet is lost and fails to reach at the destination, the packet disorder situation takes place. If this disorder is not corrected, then it severely affects the performance of the closed-loop system. This situation also takes place in all the structure of NCSs connected wirelessly.

  • Bandwidth Sharing: This issue occurs in both the shared structure as well hierarchical structure of NCSs. Both these structures provide the flexibility of connecting large number of devices (such as plant, controller, sensor and actuator) through a common network medium. As the number of devices increases, the bandwidth sharing is also increased which in turn causes reduced transmission speed, congestion, jitter or networked traffic problem. This may further deteriorate the system performance.

  • Security: The security is one of the major concern in NCS when the communication is carried out without wires. In wireless communication, there are chances of hacking due to which the false data is generated at the controller side and may cause the instability in the system. This issue needs to be handled very appropriately when the communication is carried out through shared structure or hierarchical structure of NCS.

1.2 Literature Review on Networked Control System

Due to its distinct advantages and wide industrial applications, NCS has become popular among the control engineers and also it has become an active research topic among international researchers fraternity. As mentioned earlier, NCS generally possesses a dynamic nature which results in various challenges for researchers in terms of random time delay, packet loss, multiple packet loss, packet disordering, resource allocation and bandwidth sharing. If these challenges are not handled properly, they may result in degradation of the system’s performance. Among these challenges, time delay and packet loss are considered to be crucial issues in NCS that causes potential instability.

The next section presents the concise literature survey on the compensation of network-induced delay and packet loss in continuous-time domain as well discrete-time domain.

1.2.1 NCS in Continuous-Time Domain

Various researchers [6,7,8,9,10,11,12,13,14,15,16,17,18,19] have laid their sincere efforts for designing different control algorithms that compensates the effect of network delay. In the early stages of NCS, when modelling of random time delay was difficult to obtain, the most appropriate approach was to treat the random time delay as constant which is called as deterministic delays. Luck and Ray [6, 7] introduced the concept of compensating the time delay in continuous-time domain. They compensated the effect of time delay by introducing the receiver buffer at the controller and actuator side. The size of the buffer was equal to sensor to controller delay and controller to actuator delay. The proposed methodology was tested under IEEE 802.4 network test bed considering the deterministic types of delays. Later on, Luck and Ray [8, 9] also designed predictor-based compensator in which observer was designed to estimate the plant states and predictor was used to predict the control sequences based on the past input signals. The FIFO buffer was set at the controller side and actuator side that stores the past output measurements as well as control measurements. The size of the buffer was set according to the upper bounds of sensor to controller delay and controller to actuator delay. They also tested the efficacy of the proposed algorithm on IEEE 802.4 networked medium. Chen [10] designed conventional form of memory feedback controller based on delay compensation method. Yu et al. [11] designed multiple step delay compensator for NCS in the presence of dynamic noise and measurement noises. Yang [12] proposed the state feedback controller in the presence of network delays in continuous-time domain. They proposed ZOH model at controller and actuator side to compensate the effect of network delay. They also assumed that the sensor is time-driven device, while actuator is event-driven device. Montestruque and Antsaklis [13] designed state feedback controller that compensates the effect of deterministic network delays in continuous-time domain. Kim et al. [15] modelled an NCS as a switched system with constant input delays and derived the sufficient conditions for the system stability using piecewise continuous Lyapunov methods. Godoy et al. [16] designed PID controller to compensate the network delay and validate the feasibility of controller through DC motor as plant and controller area network (CAN) bus as networked medium. Li et al. [17] designed a method for Internet-based network control system in a dual rate configuration to achieve load minimization and dynamic performance specifications. The remote PID controller was designed which regulates the output according to desirable reference and adopts the lower sampling rate to reduce the load on the network. The performance of the system was validated for fixed network delays.

In view of this, an increasing number of researchers began to investigate different control methodologies for NCSs with random or time-varying delays. Vardhan and Kumar [18] used smith predictor algorithm to compensate the effects of time-varying network delays in continuous-time domain. Urban et al. [19] studied the effect of network delays in wired and wireless networked medium using PID controller. They used CAN protocol for the wired communication and Zigbee protocol for wireless networked medium. Zhang et al. [20] proposed the stability criteria for NCS having network delays shorter as well as longer than sampling interval. They also proposed state feedback controller using conventional estimator technique that compensates the effect of network delays having time-varying nature. Similarly, Walsh et al. [21] proposed the mathematical model of NCS considering time-varying network-induced delay. They derived the stability criteria for general NCS in continuous-time domain based on TOD (try-once-discard) algorithm. Yue et al. [14] designed state feedback controller in the presence of time-varying network delays. They assumed that the network delays are lesser than sampling interval. Tipsuwan and Chow [22] proposed the concept of external gain scheduling via GSM. The GSM was used to adjust the controller gains externally at the controller output with respect to the current network traffic conditions without interrupting the internal design of controller. The network delays in the forward channel and feedback channel were modelled using RTT approach. Ji Kim [23] proposed state feedback control with estimator to compensate the effect of time-varying network delay in the presence of matched uncertainty. They tested the efficacy of the proposed controller using Ethernet as a network medium. Ma and Zhao [24] derived the stability criteria for closed-loop NCS using the average dwell time approach and piecewise Lyapunov function method. They designed state feedback controller with estimator that takes care of sensor to controller delay. Peng and Yue [25] designed the state feedback controller for NCS considering time-varying network delay in the states and matched uncertainty. Gao et al. [26] proposed a new time delay system approach which contains multiple successive delay components in the plant states, and based on that, they designed the \(H_{\infty }\) controller to overcome the effect of these state delays. Liu et al. [27] designed network predictive controller to overcome the effects of random network delay in continuous-time domain. The effects of random delays were compensated through network delay compensator placed on the actuator side. The network delay chooses the control input values from the control latest prediction sequence. Cuellar et al. [28] proposed an observer-based predictor using the Pade approximation technique for time lag processes. Sun and Xu [29] modelled the random time delays using stochastic approach in continuous-time domain. They used Markov jump linear systems approach to model sensor to controller random delay, while controller to actuator delay was assumed to be constant. Yuhong and Yeguo [30] designed state feedback controller considering time-varying network delay in the states and proved the closed-loop NCS stability using LMI approach. Ono et al. [31] designed a state feedback controller based on a modified Smith predictor which stabilized the plant in the presence of dead time. Similarly, Ridwan and Trilaksono [32] designed the \(H_{\infty }\) state feedback controller assuming all state variables are measurable in the presence of time-varying network delays. Vallabhan et al. [33] have used the analytical framework approach for compensation of random time delay and packet loss. Hikichi et al. [34] worked on continuous-time delay compensation using predictors and disturbance observer for designing a PID controller. Hu et al. [35] designed a sliding mode intermittent controller for bidirectional associative memory (BAM) using neural networks with delays. Cac et al. [36] used a pole placement method for compensating the time delay in the continuous-time domain. The algorithm was designed for the CAN-type deterministic networked medium. Yi et al. [37] solved the time delay problem by using the Smith predictor algorithm. The method was verified over wireless sensor networks (WSN) connected between the controller output and plant input. Recently, Khanesar et al. [38] modelled the random time delays using a uniform probability distribution function in continuous-time domain. Saravanakumar et al. [39] proved the stability using a Markovian jump approach for neural networks having varying time interval delays.

1.2.2 NCS in Discrete-Time Domain

Like continuous-time domain, many researchers [40,41,42,43,44,45,46,47,48,49,50,51,52] have also tried to focus their work in discrete-time domain. Jacovitti and Scarano [40] proposed various time delay estimation techniques for discrete-time systems. Nilsson et al. [41] used stochastic approach to design state feedback controller for time-varying network delays in discrete-time domain. Similarly, Shousong et al. [42] also used stochastic approach for designing optimal controllers for NCS. They assumed that the random network delays are greater than sampling interval. Zhivoglyadov et al. [43] proposed state feedback observer technique for linear network control system to compensate the effect of random delays. Yue et al. [44] provided the model of NCSs with random network-induced delay in discrete-time domain. They designed \(H_{\infty }\) controller to compensate the effect of random delays in the presence of matched uncertainty. Zhao et al. [45] designed integrated predictive controller for Networked Control System. The predictive controller is applied to generate the control predictions for each delayed sensing data and previous control information. They also designed the time delay compensator at actuator side that actively compensates the forward channel delay when control action is taken. Gou [46] designed the state feedback controller in discrete-time domain based on variable-period sampling approach for random network delays in NCS. Xiong et al. [47] introduced the concept of ZOH model at controller and actuator side that compensates the effect random network delays in discrete-time domain. The proposed ZOH model has an capability of choosing the newest control input. Li et al. [48] designed a sliding mode predictive control for compensation of delay in a Networked Control System using a Kalman predictor. They considered networked delays are random in nature with an integral multiple of sampling interval. Guo et al. [49] considered the state estimation problem for wireless NCS. The sliding mode observer was designed to solve the state estimation problem considering stochastic uncertainty and time delay. Yao et al. [50] designed a robust model predictive control (RMPC) and state observer for a class of time-varying systems under input constraints such as matched uncertainty. Yang et al. [52] proposed discrete-time sliding mode observer that estimates the random delay and compensates its effect in the presence of matched uncertainty. Argha et al. [51] designed stochastic-type sliding mode controller that compensates the effect of random networked delay with values lesser than sampling interval.

Various researchers [41, 43, 46, 51, 53,54,55,56,57] have also laid their sincere efforts to model random time delays in last decade. Among them, Nilsson et al. [41] introduced the time stamp technique to model the random time-varying networked delay. Shousong et al. [42] used stochastic approach to model the random network delays. Gou [46], Zhang et al. [53] as well as Dong and Kim [54] modelled random time delay using the concept of Markov’s chain process in discrete-time domain. They have used two-state Markov’s chain model to describe sensor to controller delay and controller to actuator delay. While Yang et al. [55] modelled random networked delay using Bernoulli’s distributed white sequence approach. Shi and Yu [56] modelled random delays using Markov’s chain process and designed output feedback controller to handle the effects of random delay. Ge et al. [57] used an independent and identically distributed approach to model the time-varying networked delay and proposed state feedback controller. Recently, Argha et al. [51] proposed Bernoulli’s white sequence approach for modelling the random time delay and proposed sliding mode controller in the presence of random time delay and matched uncertainty.

1.2.3 Packet Losses in NCS

As mentioned above, there are also possibilities of packet loss during the transmission of data packets from sensor to controller as well as controller to actuator. The packet loss takes place due to heavy network load, network congestion and node competition. In NCS, there are two types of packet losses (i) single packet loss and (ii) multiple packet loss. The single packet loss situation generally occurs when the communication of data transfer is done over a shorter distance, and multiple packet loss situations generally occur when communication of data transfer is done over a longer distance. In the research works [20, 38, 44, 48, 51, 53,54,55,56,57,58,59,60,61,62,63,64], mathematical model is proposed assuming that the packet loss within the communication medium occurs when network delay is greater than sampling interval. Zhang et al. [20] consider the deterministic single packet loss model with packet dropouts occurring at an asymptotic rate. Khanesar et al. [38] derived the packet loss model using the concept of uniform distributed probability function with single packet loss assumption. Yue et al. [44] designed single and multiple packet loss model in context with random network delays. They assumed that whenever the controller and actuator are not updated, the data packet loss takes place for that sampling interval. Li et al. [48] designed sliding mode predictive controller under multiple packet transmission policy. Zhang et al. [53] considered the packet loss in correspondence with random time delay model. They also made a generalized assumption that when the delays are greater than sampling interval, the data packets will be lost at the controller side. Dong and Kim [54] used Dirac delta probability function to derive the mathematical model of packet loss assuming the single packet loss situation. Similarly, Yang et al. [55] also considered the same packet loss approach while modelling the random time delays. Shi and Yu [56] assumed the packet loss situation while modelling the random time delays using Markov’s chain process. Argha et al. [51] included the random packet loss situation while modelling the random time delays using Bernoulli’s distribution. Hespanha et al. [58] used Bernoulli’s probability distribution function to derive the mathematical model of single packet loss as well as multiple packet loss. In both the cases, the situation of packet loss was considered when the network delay is greater than sampling interval. Gupta et al. [59] designed optimal LQG controller that compensates the effect of packet loss occurring within the network. Similarly, Wu and Chen [60] used the concept of the state estimation to compensate the effect of packet loss in discrete-time domain in NCS. Zhu and Yang [61] designed state feedback controller with multiple packet transmission. They proposed the model of NCS with multiple packet transmission and given mathematical model of packet dropout in sensor to controller channel and controller to actuator channel. Niu and Ho [62] designed the compensator using probability function that compensates the effect of packet loss within the network. Wen and Gao [63] proposed \(H_{\infty }\) controller for NCS in multiple packet transmission with random delays. They modelled multiple packet using Markov’s chain process. Recently, Song et al. [64] have proposed the packet loss model using Markov’s chain process. The model was validated for a single packet drop as well as successive packet drops. They proposed discrete-time integral sliding mode controller using the proposed model to compensate the effects of packet loss.

1.2.4 Output Feedback Control Algorithms for NCS

The above all literature discusses about design of controllers based on the state information method. The major disadvantage of these controllers is that its performance depends upon the availability of state information [65]. In various applications of NCS such as missile guidance control, aircraft control, chemical industries and automobile sectors, it is not mandatory that all the state information is available. In such cases, it is better to design the controller based on output feedback method. The main advantage of this method is that the performance of controller depends on the availability of output information which is always available. Recently, many researchers have paid much attention to designing the controllers based on output feedback approach in NCS. Mu et al. [66] proposed Luenberger output feedback based controller for discrete-time networked systems. The controller consists of two parts: a state observer that estimates plant states from the output when it is available via network and a model of the plant that is used to generate the control signal when plant output is not available. Similarly, Hespanha and Naghshtabrizi [67] designed observer based Luenberger output feedback to deal with these problems for anticipative and non-anticipative control unit in continuous-time domain. Shi and Yu [56] proved the stability of NCS with random time delays using output feedback method. Zhang and Xia [68] also designed predictive controller that compensates the effect random delays in the presence of matched uncertainty using output feedback approach. Yu and Antsaklis [69] introduced the concept of event triggered for designing output feedback controller in NCS in the presence of time-varying network delays. Zhang et al. [70] designed output feedback sliding mode controller to study the effect of variable time delay in the presence of matched uncertainty. Jungers et al. [71] proved stability of NCSs including global time-varying networked delay. They designed controller based on dynamic output feedback approach dependent on estimation of time-varying delay. Similarly, Wang et al. [72] designed output feedback \(H_{\infty }\) controller for NCSs with packet dropouts, network-induced delays and data drift. They introduced polytopic uncertainty-based data drift to model closed-loop NCSs which include random time delay and packet loss. Qui et al. [73] designed robust output feedback controller for T-S fuzzy-based affine systems with unreliable communication links with multiple packets dropout. Later, Hong et al. [74] designed conventional observer using output feedback approach for wireless NCSs with time-varying network delays and packet dropouts. They modelled wireless NCS using asynchronous dynamic system with assumption that time-varying network delays can be more or less than sampling interval. Yu et al. [75] designed multiple dynamic output feedback controllers for Networked Control Systems in the presence of random time delays and packet loss.

It is worth to mention here that the time required for the data packets to travel from sensor to controller and controller to actuator is defined as total network delay. When such delay is transformed into discrete-time domain, it mostly possesses non-integer type of values. Such network delays in discrete-time domain are defined as fractional delays [76,77,78,79,80,81,82]. The networked control system has sensor to controller fractional delay present in the feedback channel and controller to actuator fractional delay present in the forward channel. The nature of both these fractional delays depends on the type of the communication medium. In NCS, when the data packets are exchanged through real-time communication medium, the network delay always have the fractional delay. So, it is important to compensate the effect of deterministic and random type of fractional delay in discrete-time domain at each sampling instant in the presence of packet loss and matched uncertainty.

1.3 Contribution of Book

This book contributes mainly following:

  • Firstly, a novel discrete-time sliding surface is proposed using the compensated state information and proposed discrete-time sliding mode control algorithm that encompasses deterministic type network-induced fractional delay and single packet loss. The Thiran approximation technique is used for compensating the fractional delay. There are two types of DSMC proposed, namely switching type and non-switching type. The conditions for stability of the closed-loop system are derived using the Lyapunov approach in both the algorithms. The efficacy of the algorithms is endorsed in simulation as well validated experimentally. Further, the proposed algorithms are compared with conventional sliding mode controller with CAN and Switched Ethernet as network medium.

  • The algorithms are further extended for output feedback. A multirate output feedback (MROF) technique is used to estimate the state variables. The proposed multirate output feedback discrete-time sliding mode controller performance is checked under the networked environment.

  • Next, the monograph proposes the discrete-time sliding surface design for the random fractional delay and single packet loss that occur within the sampling period. The random delay is compensated using Thiran’s approximation technique in the presence of packet loss situation. The random fractional delay is modelled by Poisson’s distribution function, and packet loss is modelled by Bernoulli’s function. The closed-loop stability is established using the Lyapunov function. The efficacy of proposed non-switching type of DSMC is endowed by simulation results and also experimentally validated with servo system.

  • Further, the proposed algorithms are extended for the random fractional delay with multiple packet loss situation. The multiple packet loss policy is defined for the development of DSMC. The simulation as well as experimental results with various fractional delay situation and matched uncertainties are shown to prove the efficacy of the proposed algorithms.

1.4 Organization of Book

The outline of book is as follows:

  • Chapter 1 briefs Networked Control System and the literature survey on various control algorithms proposed in continuous-time as well as discrete-time domain. This chapter also discusses various issue of NCS.

  • Chapter 2 discusses the preliminaries of Networked Control System and sliding mode control technique. In this chapter, a basic block diagram of NCS with different types of time delays that affect the performance of the system are discussed. The origin of sliding mode controller in continuous-time domain and discrete-time domain is also briefly discussed. Lastly, the challenges in designing DSMC for NCS are discussed.

  • The main contribution of monograph, design of discrete-time sliding mode control for deterministic type fractional delay is discussed in Chap. 3. Firstly, sliding surface is proposed with compensated delay. The network-induced delay is compensated using Thiran’s approximation. The discrete-time sliding mode control law is derived using proposed sliding surface with switching type reaching law. Further, the stability of the closed-loop NCS is proved through Lyapunov approach. The efficacy of the proposed algorithm is tested under simulation environment and experimental environment.

  • Chapter 4 presents design of non-switching type discrete-time sliding mode controller in the presence of deterministic fractional delay and matched uncertainty. In this chapter, the design of control law is based on sliding surface derived using Thiran’s approximation. Further, the stability of the closed-loop NCS is proved through Lyapunov approach that ensures the finite-time convergence of system states within the specified band. The efficacy of the proposed algorithm is tested under simulation and experimentation with real-time networks like CAN and Switched Ethernet.

  • Chapter 5 describes the design of discrete-time sliding mode control using multirate output feedback approach with fractional delay compensation. In this chapter, the MROF-based estimator is placed at the plant side using the control input signal and fast measured output. The stability of the closed-loop NCS with derived control law is proved using Lyapunov approach. The simulation results are carried out in the presence of network delay and matched uncertainty in order to prove the effectiveness of proposed algorithm.

  • Chapter 6 describes the design of discrete-time sliding mode controller for random communication delay and single packet loss. In this chapter, the compensation of random fractional delay is discussed using Thiran’s approximation with packet loss condition. The mathematical models of random fractional delay and single packet loss are derived using stochastic approach. The derived discrete-time sliding mode control law is verified through simulation and implementation results in the presence of random fractional delay, packet loss and matched uncertainty.

  • Chapter 7 discusses design of discrete-time sliding mode control for multiple packet transmission. The multiple packet loss policy is defined and used for DSMC design. Further, a second-order disturbance estimator is incorporated at the plant side to estimate the disturbance \(O(h^{3})\) that occur on the plant. This improves the robustness properties of closed-loop NCS. The efficiency of the proposed algorithm is verified through simulation results.

  • The concluding remarks along with future scope and challenges are mentioned in Chap. 8. The final comments and future scope of discrete-time SMC algorithms are discussed in this chapter. Lastly, various challenges are also listed that are still remain unsolved in network control system domain.