Rastko Selmic

Louisiana Tech University | Louisiana Tech · Department of Electrical Engineering

This article studies cyberattack detection in discrete-time leader-following nonlinear multiagent systems subject to unknown but bounded system noises. The Takagi–Sugeno fuzzy model is used to approximate the nonlinear systems over the true value of the state. A new method is developed for simultaneous distributed cyberattack detection and leader-f...

The paper focuses on formation control with constant communication delays for second-order, uncertain, nonlinear multi-agents with a nonsymmetric control gain matrix and an unknown control direction. The multi-agent system is modeled using an undirected graph. A three-layer neural network (NN) is employed to approximate an unknown nonlinearity. Unl...

Cybersecurity is a critical process that safeguards networks, systems, and applications against cyber-attacks, wherein digital information is targeted for unauthorized access, manipulation, or destruction. As attackers continually evolve their tactics, addressing cybersecurity challenges has become paramount, especially in sensitive domains like th...

This paper considers a leader-following formation control problem for a class of second-order, uncertain, input-affine, nonlinear multi-agent systems modeled by a directed graph. A three-layer neural network (NN) is proposed with an input layer, two hidden layers, and an output layer to approximate an unknown nonlinearity. The NN weights tuning law...

One of the main features of autonomous control systems is solving complicated optimization problems without human intervention in the presence of uncertainty in real time [1] . Autonomous systems (ASs) must have the recognition and discretion potency, evaluation and estimate authority, and decision-making power to independently perform various ta...

This paper studies the gradient-based adversarial attacks on cluster-based, heterogeneous, multi-agent, deep reinforcement learning (MADRL) systems with time-delayed data transmission. The structure of the MADRL system consists of various clusters of agents. The deep Q-network (DQN) architecture presents the first cluster’s agent structure. The oth...

In this paper, a formation control problem of second-order holonomic agents is considered, where agents navigate around obstacles using proximal policy optimization (PPO)-based deep reinforcement learning (DRL). The formation is allowed to shrink and expand, while maintaining its shape, in order to navigate the geometric centroid of the formation t...

This paper proposes a distributed, false data injection (FDI) cyber-attack detection method in communication channels for a class of discrete-time, nonlinear, heterogeneous, multi-agent systems controlled by our formation-based controller. A distributed neural network (NN)-based observer is proposed that generates the residual signal which is used...

This paper is concerned with cyberattack detection in discrete-time, leader-following, nonlinear, multi-agent systems subject to unknown but bounded (UBB) system noises. The Takagi-Sugeno (T-S) fuzzy model is employed to approximate the nonlinear systems over the true value of the state. A distributed cyberattack detection method, based on a new fu...

This paper considers a leader-following formation control problem for heterogeneous, second-order, uncertain, input-affine, nonlinear multi-agent systems modeled by a directed graph. A tunable, three-layer neural network (NN) is proposed with an input layer, two hidden layers, and an output layer to approximate an unknown nonlinearity. Unlike commo...

A heuristic dynamic programming (HDP) algorithm for trajectory tracking and formation control of multi-agent systems (MAS) is presented in this paper. The selected HDP method allows for an online optimal control design. The multi-agent control problem is formulated as a leader-follower, where it is necessary for followers to maintain the assigned f...

This paper studies distance-based formation control of a set of nonlinear multi-agent systems over directed graphs. We propose a distributed, distance-based formation control scheme for a set of heterogeneous, nonlinear agents over a particular class of minimally, structurally persistent, directed graphs in a 3-D space, namely directed trilateral...

This paper proposes a distance-based formation control and target tracking for multi-agent systems where agents are modeled using second-order nonlinear systems in the presence of disturbance. By applying a rigid graph theory, we develop a neural network-based backstepping controller to address the distance-based formation control problem of nonlin...

This work establishes properties of the normalized rigidity matrix in two-and three-dimensional spaces. The upper bound of the normalized rigidity matrix singular values is derived for minimally and infinitesimally rigid frameworks in two-and three-dimensional spaces. We prove that the transformation of a framework does not affect the normalized ri...

This paper proposes a distributed cyber-attack detection method in communication channels for a class of discrete, nonlinear, heterogeneous, multi-agent systems that are controlled by our proposed formation-based controller. A residual-based detection system, exploiting a neural network (NN)-based observer, is developed to detect false data injecti...

This paper introduces an adaptable, human-computer interaction method to control multi-rotor aerial vehicles in unsupervised, multi-subject environments. A region-based convolutional neural network (R-CNN) first detects subjects in a frame and their faces’ regions of interest (RoIs), which are then fed to a facial recognition module to search for t...

This paper proposes an adaptive neural network-based backstepping controller that uses rigid graph theory to address the distance-based formation control problem and target tracking for nonlinear multi-agent systems with bounded time-delay and disturbance. The radial basis function neural network (RBFNN) is used to overcome and compensate for the u...

This paper proposes an adaptive neural network-based backstepping controller that uses rigid graph theory to address the distance-based formation control problem and target tracking for nonlinear multi-agent systems with bounded time-delay and disturbance. The radial basis function neural network (RBFNN) is used to overcome and compensate for the u...

This paper proposes a neural network-based backstepping controller to address the distance-based formation control problem and target tracking for a class of nonlinear multiagent systems in Brunovsky form using rigid graph theory. The radial basis function neural network (RBFNN) is used to ensure the system stability in the presence of unknown nonl...

Positive definite (PD) kernels are an important tool in machine learning that enable efficient solutions to otherwise difficult or intractable problems by implicitly linearizing the problem geometry. In this paper, we develop a set-theoretic interpretation of the earth mover's distance (EMD) and propose earth mover's intersection (EMI), a PD analog...

Unordered feature sets are a nonstandard data structure that traditional neural networks are incapable of addressing in a principled manner. Providing a concatenation of features in an arbitrary order may lead to the learning of spurious patterns or biases that do not actually exist. Another complication is introduced if the number of features vari...

Security in wireless sensor networks (WSNs) is centered on six fundamental requirements namely, authentication, confidentiality, integrity, reliability, availability and data freshness [4, 39, 52, 62]. In this chapter, we describe these requirements, the different kinds of attacks that aim to compromise these requirements (and hence the security of...

A major advantage of wireless sensor networks (WSNs) over wired networks is the potential for ad hoc deployment of the network. If the monitoring of a dangerous environment is required, then one may not be able to deploy a wired network.

Localization—the process by which the positions of the nodes of a Wireless Sensor Network (WSN) are found with respect to some absolute or relative frame of reference—is fundamental to how the WSN performs at executing its functions.

A wireless sensor network (WSN) is a system that consists of multiple sensing elements distributed spatially with a specific objective to measure different physical quantities and communicate those measurements between themselves and the central gateway.

Previous chapters describe sensor network architecture, communication protocols and various characteristics related to security of WSNs.

WSNs are used for a wide range of applications such as environmental sensing, clinical monitoring, and military surveillance. As in traditional data networks such as the Internet and Mobile Ad hoc Networks (MANETs), the different applications of WSNs demand that certain service requirements be met for optimal performance.

The development of microcontrollers, communication technology, micro-electromechanical systems (MEMS) and nanotechnology allowed for research and development of new systems for sensing and communication called wireless sensor networks (WSNs). Such networks are characterized as ad hoc (no previous setup or supporting infrastructure is required), uti...

We present a solution to the nonplanar, multiagent formation control problem using rigid graph theory. First we provide an overview of rigid graph theory, which is an abstraction of the rigidity of civil and mechanical structures and can be used in designing frameworks that are rigid or flexible (ie, nondeformable or deformable). The rigidity of a...

This paper provides a solution to the nonplanar multiagent formation control problem using graph rigidity. We consider a 3-D multiagent formation control where multiple agents are operating in one plane and some other agents are operating outside of that plane. This can be referred to as a layered formation control where the objective is for all ag...

This paper addresses the decentralized formation control problem of unmanned vehicles in three-dimensional (3D) space. We propose a new Neural Network-based control law that uses graph rigidity and relative positions of the vehicles. The novel control law for each vehicle consists of a nonlinear component that is dependent on the closed-loop error...

This book presents a comprehensive overview of wireless sensor networks (WSNs) with an emphasis on security, coverage, and localization. It offers a structural treatment of WSN building blocks including hardware and protocol architectures and also provides a systems-level view of how WSNs operate. These building blocks will allow readers to program...

The development of microcontrollers, communication technology, microelectromechanical systems (MEMS), and nanotechnology allowed for research and development of new systems for sensing and communication called wireless sensor networks.

Positive definite kernels are an important tool in machine learning that enable efficient solutions to otherwise difficult or intractable problems by implicitly linearizing the problem geometry. In this paper we develop a set-theoretic interpretation of the Earth Mover's Distance (EMD) that naturally yields metric and kernel forms of EMD as general...

Wireless sensor network (WSN) coverage completeness is an important quality of service (QoS). It is frequently assumed that events occurring in the sensor field can always be detected. However, this is not necessarily the case, particularly if there are holes in the sensor network coverage. This paper introduces a novel centralised method for detec...

This paper is concerned with the evaluation and comparison of several methods for the classification and recognition of static hand postures from small unlabeled point sets corresponding to physical landmarks, e.g. reflective marker positions in a motion capture environment. We compare various classification algorithms based upon multiple interpret...

New Unmanned Ground Vehicle (UGV) navigation algorithms in coordinate-free and localization-free Wireless Sensor and Actuator Networks (WSANs) are presented. The algorithms are distributed and designed for multi-UGV multi-destination navigation in a sensor network environment. The proposed method eliminates possible multiplicity of destination node...

We consider a Micro-Aerial Vehicle (MAV), used as a mobile sensor node, in conjunction with static sensor nodes, in a mission of detection and localization of a hidden Electromagnetic (EM) emitter. This paper provides algorithms for the MAV control under the Position-Adaptive Direction Finding (PADF) concept. The MAV avoids obstructions or location...

In this paper we describe plans for a dynamic hand gesture recognition system based on motion capture cameras with unlabeled markers. The intended classifier is an extension of previous work on static hand gesture recognition in the same environment. The static gestures are to form the basis of a vocabulary that will allow precise descriptions of v...

This paper provides a summary of the development of a three state machine-based cooperative control algorithm that is applied to multiple Unmanned Aerial Vehicles (UAVs) or Micro-Aerial Vehicles (MAVs) control. We use MAVs for cooperative search of a hidden electromagnetic source (emitter) in a controlled environment. MAVs are equipped with wireles...

We consider a navigation problem in a distributed, self-organized and coordinate-free Wireless Sensor and Ac- tuator Network (WSAN). We rst present navigation algorithms that are veried using simulation results. Con- sidering more than one destination and multiple mobile Unmanned Ground Vehicles (UGVs), we introduce a distributed solution to the Mu...

Wireless sensor networks are being used in a variety of ways - from reconnaissance and detection in military to biomedical applications and a wide variety of commercial endeavors. We introduce a position-adaptive direction finding method using mobile sensor networks and present recent experimental results in localization of a non-cooperative sensor...

This paper provides a summary of preliminary RF direction finding results generated within an AFOSR funded testbed facility recently developed at Louisiana Tech University. This facility, denoted as the Louisiana Tech University Micro- Aerial Vehicle/Wireless Sensor Network (MAVSeN) Laboratory, has recently acquired a number of state-of-the-art MAV...

This paper provides a summary of recent results on a novel multi-platform RF emitter localization technique denoted as Position-Adaptive RF Direction Finding (PADF). This basic PADF formulation is based on the investigation of iterative path-loss based (i.e. path loss exponent) metrics estimates that are measured across multiple platforms in order...

This paper presents an application of a neural network control scheme in a thermopower generator control problem. Presented adaptive neural network control scheme for power systems frequency control does not require offline training. We apply Lewis' neural network controller [Lewis, Jagannathan, & Yesildirect, Neural network control of robot manipu...

Trust is very important in wireless sensor networks to transfer the data from source to destination. The Dynamic Source Protocol calculates the alternate path, if any node fails to transfer the data. The Dynamic Source Protocol does not have any built-in functionality to calculate an alternate path if the path has a malicious node. With the expense...

Cooperation in wireless sensor networks to detect the malicious node without any infrastructure is a recent trend in research. The current models need more storage, computation, security tools, and communication requirements. They fail in wireless sensor networks due to limitation of resources. Trust-based approach does not need high-end resource r...

Using a dynamical model of happiness in humans, proposed in, we develop a model of happiness for a network of people using dynamical elements with interconnecting coupling factors. The network model of happiness is derived from a leader-followers network model proposed by Wang and Slotine. Such networks with interconnected dynamical elements can be...

In this paper, we present an algorithm for navigating an unmanned ground vehicle (UGV) through coordinate-free Wireless Sensor and Actuator Networks. The navigation algorithm proceeds in two phases. In the first phase, for each node in the sensor network we compute a hop-level distance from the target node, which indicates the number of communicati...

This paper provides a discussion and a summary of the latest results on a novel sensor-node based approach to emitter localization denoted as Position - Adaptive Direction Finding (PADF). PADF is based on the formulation and investigation of iterative path-loss based (i.e. path loss exponent) metrics estimates that are measured across multiple plat...

A number of potential advantages associated with a new concept denoted as Sensor Agnostic Networks are discussed. For this particular paper, the primary focus is on integrated wireless networks that contain one or more MAVs (Micro Unmanned Aerial Vehicle). The development and presentation includes several approaches to analysis and design of Sensor...

We discuss the development, design, implementation, and demonstration of a robotic UGV (Unmanned Ground Vehicle) system for networked and non-line-of-sight sensing applications. Our development team is comprised of AFRL Summer Interns, University Faculty, and Personnel from AFRL. The system concept is based on a previously published technique known...

Ultra-wideband (UWB) signals exhibit different characteristics upon propagation through matter compared with narrow-band signals. The latter keeps a sinusoidal shape during different forms of signal propagation. The behavior of narrow-band signals does not apply to UWB signals in many cases. Presently, the possibilities for development of UWB signa...

Ultra-wideband (UWB) signals exhibit different characteristics upon propagation through matter compared with narrowband signals. The latter keeping a sinusoidal shape during different forms of signal propagation. The behavior of narrowband signals does not apply to UWB signals in many cases. Presently, the possibilities for development of UWB signa...

We present a set of algorithms for the navigation of Unmanned Ground Vehicles (UGVs) towards a set of pre-identified target nodes in coordinate-free and localization-free wireless sensor and actuator networks. The UGVs are equipped with a set of wireless listeners that provide sensing information about the potential field generated by the network o...

Wireless sensor network coverage completeness is an important Quality of Service measure. It is frequently assumed that events occurring in the sensor field will always be detected. However, this is not necessary the case, particularly if there are holes in the sensor network coverage. This paper introduces a novel method for detection and relative...

We discuss the development of Position-Adaptive Sensors [1] for purposes for detecting embedded chemical substances in challenging environments. This concept is a generalization of patented Position-Adaptive Radar Concepts developed at AFRL for challenging conditions such as urban environments. For purposes of investigating the detection of chemica...

We discuss the development of Position-Adaptive Sensors [1] for purposes for detecting embedded chemical substances in challenging environments. This concept is a generalization of patented Position-Adaptive Radar Concepts developed at AFRL for challenging conditions such as urban environments. For purposes of investigating the detection of chemica...

Detection and patching of coverage holes in Wireless Sensor Networks (WSNs) are important measures of Quality of Service (QoS) for security and other applications that emphasize sensor network coverage. In this paper, we model a WSN using simplicial complexes based on its communication graph by which the network can be represented as connections of...

This paper presents a model for fault detection of a power leader in nonlinear leader-follower networks. The fault detection method is developed for the network model proposed by Wang and Slotine. Every follower is coupled with a nonlinear, neural net based observer for fault detection. Neural net tuning algorithms are derived and fault identifiers...

This paper presents a concept of feedback control for smart actuators that are compatible with smart sensors, communication protocols, and a hierarchical Integrated System Health Management (ISHM) architecture developed by NASApsilas Stennis Space Center. Smart sensors and actuators typically provide functionalities such as automatic configuration,...

This paper presents a dynamic model of wireless sensor networks (WSNs) and its application to sensor node fault detection. Recurrent neural networks (NNs) are used to model a sensor node, the node's dynamics, and interconnections with other sensor network nodes. An NN modeling approach is used for sensor node identification and fault detection in W...

This paper presents the real-time implementation of a neural network-based fault detection for wireless sensor networks (WSNs). The method is implemented on TinyOS operating system. A collection tree network is formed and multi-hoping data is sent to the base station root. Nodes take environmental measurements every N seconds while neighboring node...

We have formulated a series of position-adaptive sensor concepts for explosive detection applications using swarms of micro-UAV's. These concepts are a generalization of position-adaptive radar concepts developed for challenging conditions such as urban environments. For radar applications, this concept is developed with platforms within a UAV swar...

We have formulated a series of position-adaptive sensor concepts for explosive detection applications using swarms of micro-UAV's. These concepts are a generalization of position-adaptive radar concepts developed for challenging conditions such as urban environments. For radar applications, this concept is developed with platforms within a UAV swar...

This paper presents a dynamic model of wireless sensor networks (WSNs) and its application to a sensor node fault detection. Recurrent neural networks (RNNs) are used to model a sensor node, its dynamics, and interconnections with other sensor network nodes. The modeling approach is used for sensor node identification and fault detection. The input...

This paper presents a concept of a feedback control for smart actuators that are compatible with NASA's smart sensors, communication protocols, and overall hierarchical system architecture. The smart actuators are needed in order to embed intelligence, self-learning, and self-diagnosis into every actuator that is part of a complex system such as a...

Wireless Sensor Networks (WSNs) collect information about the physical environment aiding to a wide variety of applications ranging from target detection to monitoring of harmful chemical gases. The drive to scale down the system size and cost has resulted in constraints in the quality of components. Reliable and accurate performance of sensors is...

This paper presents a dynamic model of wireless sensor networks (WSNs) and its application to sensor node fault detection. Recurrent neural networks (NNs) are used to model a sensor node, the node's dynamics, and interconnections with other sensor network nodes. An NN modeling approach is used for sensor node identification and fault detection in W...

This paper describes actuator fault identification in unknown, input-affine, nonlinear systems using neural networks. Neural net tuning algorithms have been derived and identifier have been developed using the Lyapunov approach. The paper defines and analyses the fault dynamics i.e., the dynamical properties of a failure process. A rigorous detecta...

As markets continue to become global and firms become more international, corporate joint ventures provide firms with opportunities to rapidly create economies of scale and learn new skills and technologies that would be very difficult for them to obtain ...

A neural net (NN)-based actuator saturation compensation scheme for the nonlinear systems in Brunovsky canonical form is presented. The scheme that leads to stability, command following, and disturbance rejection is rigorously proved and verified using a general "pendulum type" and a robot manipulator dynamical systems. Online weights tuning law, t...

An indirect adaptive neural network (NN) saturation compensator is presented for a class of nonlinear systems. Output feedback control is considered where only the system output is assumed to be measurable. The imposed actuator saturation is assumed to be unknown and treated as the system input disturbance. A NN-based state observer estimates deriv...

This paper presents an analytical modeling of a piezoelectric multi-layer cantilever used as a micro-electro-mechanical-system (MEMS) chemical sensor. Selectively coated microcantilevers have been developed for highly sensitive chemical sensor applications. The proposed piezoelectric chemical sensor consists of an array of multi-layer piezoelectric...

A robust neural network (NN) composite saturation compensation scheme is presented for a trajectory tracking and a vibration suppression of a flexible link robot arm. The scheme is based on a singular-perturbation technique and can accommodate an unknown disturbance and the saturation constraints. The saturation compensator is composed of a robust...

A neural network-based saturation compensation signal is inserted into the actuator control, effectively preventing it from being saturated. The proposed neural network (NN) saturation compensation scheme presents a form of intelligent anti-windup saturation where NN adjusts its output to prevent saturation of the control signal. On-line weights tu...

This article describes the design and characterization of an ultra-wideband (UWB) pulse generator suitable for biomedical research. The pulser generates nanosecond-long electromagnetic pulses of adjustable rise time, amplitude, and duration, permitting analysis of possible bioeffects over a range of pulse characteristics. Nanosecond rise-time pulse...