As cyber-attacks on web applications evolve, machine learning models for Web Application Firewalls (WAFs) should become more sophisticated. Gradient Boosting and Neural Networks are applied to a dataset of HTTP traffic (CIC-IDS2017), which includes features such as request methods, timestamps, response codes, and user agents, extracted for anomaly detection. Gradient Boosting, an ensemble method that builds sequential decision trees, effectively handles class imbalance and complex patterns in the data. Neural Networks, while highly flexible and capable of capturing complex relationships, face challenges with overfitting due to the relatively small dataset. The performance of these two advanced methods is compared with previous results from Random Forest and Support Vector Machines, demonstrating clear improvements in accuracy and F1-score with Gradient Boosting, while Neural Networks demonstrate potential, especially with access to larger datasets. This study highlights the benefits of using more advanced techniques for HTTP traffic analysis and provides a comparison framework to guide future WAF model development.

Stimulated by Pharma 4.0, pharmaceutical industry is undergoing a transition from batch to continuous manufacturing (CM). There are manifold of benefits of continuous tablets production such as: increased product quality; improved efficiency; reduced production costs. Even more, CM could accelerate the product development resulting in faster delivery of medicine to patients, and this without affecting product quality. To exploit the advantages of CM, the mainly separated process units need to be integrated to form end-to-end systems from raw material to final product. Such integration requires a deep understanding and a holistic approach toward process development and optimization. In this paper, a feasibility study of using model predictive control for the multi-variable process of tablet production is presented. The simulations have been performed on a novel and unique simulation environment where the main unit processes involved in the manufacturing of tablets are implemented. Preliminary results indicate that the proposed methodology is feasible and implementable on the benchmark simulation.

In the context of the development of wireless networks in the field of connected vehicles, platooning becomes more common on the road, and vehicular safety communication is crucial to ensure driver and passenger safety. Communication between vehicles within platoons relies on Dedicated Short-Range Communication (DSRC) and Cooperative Adaptive Cruise Control (CACC). The DSRC allows vehicles to communicate by exchanging safety information like location, speed, and acceleration. As a result, the CACC control solution maintains a speed-independent inter-vehicle distance of the followers’ vehicles based on this information. The platoon systems are susceptible to data packet loss that determine delayed and incomplete responses which can lead to collisions or other safety issues. This paper analyzes indicators such as the distance between platoon members, the front delay, and the relative speed of vehicles that have an impact on vehicular safety communication.

In the context of continuous digitalization in all business areas, cyber-attacks became a major challenge. The data security component is a priority in the development of this changing environment. There have been many types of cyber-attacks such as Distributed Denial-of-Service, Malware Attack, Password Attack, or SQL Injection Attack that have affected different countries around the world. These security vulnerabilities led to data exfiltration or unavailability of certain critical services for long periods of time, while causing significant material damage. In order to properly manage the problems that may occur, it is necessary for any entity offering services on the Internet to have a cyber security component. Moreover, for a better coordination in the response to cyber incidents, it is necessary to train the teams providing these services. The development of capabilities can be achieved in a controlled environment, where different types of real-life attacks are simulated. A classic training method is considered to be a Red vs. Blue competition. In this paper, a different strategy is proposed, in which the two types are combined in a Red and Blue scenario. To develop it, a central system is needed to generate the information to be captured, to validate its accuracy and to monitor the availability of services. Also, through this, it was possible to observe the behaviour of the players and how they improve their knowledge. Compared to the Red vs. Blue version, the proposed one in which the teams are combined offers more advantages for understanding how a hacker acts and for reducing the response time in case of a cyber security incident.

The results obtained so far from the industry and academia illustrate that fully automated vehicles are increasingly closer to be deployed in series on the roads, but they need to be connected between them and with the infrastructure for optimal performances. The roundabout type of intersection contributes to decreasing the number of accidents, but it's performance is reduced when the traffic volume increases. In this case, the intersection becomes an area of bottlenecks. To minimize the travel time of connected and automated vehicles through roundabouts by reducing the waiting time before entering it, in this paper, a centralized controller for a roundabout was proposed. It has the task of computing and sending the waiting time for the vehicles that want to travel through the intersection such that the total time to traverse it is minimized and the collisions are avoided.

Nowadays, the wind power industry development and the increasing size of wind farms pose a couple of technical challenges. The most fundamental problem is caused by the dynamics of the air turbulence created by wind turbines (i.e., turbines mix wind into chaotic vortices, termed wakes). The power production of a wind turbine that operates in the wake of another is reduced up to 40%, the turbine having also a higher unsteady loading and a shorter mean time between failures. But, considering the wind turbines inter-dependencies and making them cooperative by exchanging information, the resulting wind farm performances would be much greater than those obtained while controlling individual wind turbines and neglecting their interactions. Moreover, the particular feature of wind farms, i.e., the interconnected structure of the involved wind turbines (seen as sub-systems) through the wakes effect, make them suitable as distributed model predictive control (DMPC) applications. Thus, this paper makes a literature review regarding the recent results on DMPC for the cooperative wind farm control problem, starting with the wind turbines and wind farms modeling and control, then presenting the DMPC solutions and concluding with the challenges and opportunities for future research directions.

It is known that nowadays there are too many vehicles present on the roads that lead to an increased risk of accidents, fuel consumption, and pollution, vehicle components wear, more traffic jams, driving stress and discomfort for the passengers, longer journey duration, large vehicle-to-vehicle gaps. One solution is represented by vehicle platooning, which brings many advantages, but is naturally based on single-lane scenarios. Since vehicles drive on multiple-lane roads and highways and considering the optimal space utilization, the following concept is to treat vehicles as multiple-lane groups, i.e., individual vehicles that travel closely together on different lanes and are electronically connected through communications, by using swarm intelligence that borrows ideas from swarm robotics and dynamical multi-objective optimization. As such, this paper makes a survey of the state-of-the-art related to vehicle platooning, swarm robotics concepts, swarm path planning and traffic simulators, and presents their links in the proposed concept.

The control of human hands and fingers represents one of the most complex functions of the motor cortex. In order to implement anthropomorphic hands that mimic accurately the motion ability of the human hands, the basic biological mechanisms of the natural muscle control should be modelled. This paper presents the design of a significantly improved control system based on analogue neural networks that can be used to replicate the biological control mechanisms of the natural muscles. In order to demonstrate the proposed concept, experiments were performed using a single-joint robotic arm that can be flexed as the human elbow by an artificial muscle connected as the biceps. In order to bring more biological plausibility to the robotic arm, the artificial muscle is implemented using a shape memory alloy wire which actuates by contraction as the natural muscles. Moreover, the contraction force of the actuator wire is directly determined by the spiking frequency of the electronic neurons as the motor neurons determines the contraction strength of the natural muscles. The experimental results show that using excitatory neurons and several inhibitory neurons unevenly distributed on the inputs of the artificial motor neurons that drive the shape memory alloy actuator, the spiking neural network is able to control with high precision the rotation of the arm mobile lever to random target positions even if the arm is slightly loaded. These results validate the control method of the robotic arm junction showing that the analogue spiking neural networks represents a very good alternative to control the contraction of SMA actuators in a biological plausible manner.

This paper deals with the problem of damping driveline oscillations, which is crucial to improving driveability and passenger comfort. This problem has received an increased interest due to the introduction of the dual-clutch powershift automatic transmission with dry clutches. This type of transmission improves fuel economy, but it results in a challenging control problem, due to driveline oscillations. These oscillations occur during gear-shift, while traversing backlash or when tip-in and tip-out maneuvers are performed. This paper presents the derivation of an accurate piecewise nonlinear drivetrain model with three inertias. Then, a horizon-1 predictive controller based on flexible Lyapunov functions is designed for the piecewise nonlinear drivetrain model. Several simulations show that the proposed control scheme can handle both the performance and physical constraints and the strict limitations on the computational complexity, while obtaining better performances compared to using PWA models of the vehicle drivetrain.

The goal of this paper is to provide a control design methodology that can assure the closed-loop performances of a physical plant, while compensating the time-varying delays introduced by the communication network that links the controller with the remote process. Firstly, the error caused by the time-varying delays is modeled as a disturbance and a novel method of bounding the disturbances is proposed. Then, a robust one step ahead predictive controller based on flexible control Lyapunov functions is designed, which explicitly takes into account the bounds of the disturbances caused by time-varying delays and guarantees also the input-to-state stability of the system in a non-conservative way. Moreover, it is shown that by choosing an appropriately Lyapunov function, the MPC algorithm amounts solving a single, low-complexity linear program each sampling instant. The modeling method and the control strategy were tested on a vehicle drivetrain controlled through CAN, with the aim of damping driveline oscillations, which is crucial in improving driveability and passenger comfort. Several True-Time simulations based on realistic scenarios show that the proposed control scheme can handle both the performance/physical constraints and the strict limitations on the computational complexity.

This paper deals with the problem of damping driveline oscillations, which is crucial to improving driveability and passenger comfort. Recently, this problem has received an increased interest due to the introduction in several production vehicles of the dual-clutch powershift automatic transmission with dry clutches. This type of transmission improves fuel economy, but it results in a challenging control problem, due to driveline oscillations. These oscillations, also called shuffles, occur during gear-shift, while traversing backlash or when tip-in and tip-out maneuvers are performed. The first contribution of this paper is the derivation of an accurate piecewise affine drivetrain model with three inertias. The second contribution is concerned with the design of a horizon-1 predictive controller based on flexible Lyapunov functions. Several simulations based on realistic scenarios show that the proposed control scheme can handle both the performance and physical constraints, and the strict limitations on the computational complexity.