Open-source RISC-V CPU architectures provide FPGA developers with fine-grained control over resource utilization and performance. This work presents a case study in throughput maximization and PPA (power, performance, area) optimization for a minimal RISC-V core on FPGA, with an emphasis on structured SystemVerilog design practices. We propose a short, single-cycle pipeline architecture targeting resource-constrained deployments and systematically compare its PPA characteristics against similar performance-class implementations. FPGA-specific optimizations, including tailored Register File and ALU configurations, are employed to improve critical path timing and overall throughput. The resulting design, eduBOS5, achieves a 2× increase in DMIPS/MHz while reducing LUT utilization by 24% compared to PicoRV32 on the Gowin LittleBee FPGA. PPA metric scaling over different FPGAs was addressed by porting the design to Xilinx and Lattice devices.

Motion tracking achieved via conventional video processing and machine vision algorithms is often hindered by challenges such as motion blur and the lack of distinctive visual features, particularly when tracking fast-moving objects. To address these limitations, active visual markers are often used. In this paper, we present the design and prototype implementation of an active marker that is compact, detachable, and self-powered, making it well-suited for real-world tracking applications. Furthermore, the marker is fully configurable through an accompanying software solution and an additional wireless communication controller via an infrared protocol. The applicability of the developed markers is demonstrated using both conventional RGB and event-based cameras, highlighting their versatility and robustness across diverse sensing modalities. Their tracking capabilities are validated in both single- and multi-object scenarios. Overall, the developed multi-functional markers provide a flexible and practical foundation for high-speed motion tracking under challenging visual conditions, paving the way for further research and advanced applications in related fields.

Timely and accurate defect detection is essential in the leather industry, as the quality of raw leather directly impacts both the usability and value of finished products. This paper provides a systematic overview of state-of-the-art solutions and proposes a novel approach for automated detection of leather surface defects using deep neural networks based on the Inception-V3 architecture. Five defect categories are introduced, focusing on their impact on leather quality. In addition, two deep neural network architectures were analyzed and implemented for defect detection and classification: a single-channel model and a multi-channel model with arbitration. The evaluation was carried out using a combination of a custom-developed dataset and publicly available datasets, assessed with standard performance metrics. Moreover, an image annotation tool was developed to facilitate precise defect labeling and the creation of variable-size datasets. Both models demonstrated promising results on the custom dataset, achieving accuracy rates exceeding 93%. The suggested methodology enhances the research domain of leather inspection automation by creating an openly accessible image dataset, performing a comparative analysis of detection models and creating software tools for data preparation. These contributions lay the foundation for further research in leather defect detection and potential industrial implementation.

This paper explores the application of FPGA programmable structures in the field of digital image signal processing (ISP). FPGAs offer high flexibility, speed and parallelism, making them ideal for general digital signal processing (DSP), as well as specific ISP tasks. The paper utilizes standard ISP algorithms such as morphological operations, filtering and edge detection to compare practical implementations of FPGA and CPU-based compute engines. Through illustrative examples and empirical results, we demonstrate the distinct advantages of employing FPGA for these use-cases, and contrast them with traditional CPU approaches, clearly showing FPGA capacity to significantly accelerate execution. The challenges that arise from resource-limited IOT-class hardware configurations are highlighted in the paper, namely resource optimization, memory management and maximal frequency.

This paper focuses on the design and implementation of a discrete digital PID (Proportional - Integral - Derivative) controller utilizing an FPGA (Field Programmable Gate Arrays) platform, which inherently supports parallel implementation of algorithms. Typically, cost-effective FPGA boards lacks peripherals, such as analog inputs and outputs, so they need to be added externally. The main hypothesis is that a DC motor system can be controlled with a low-cost variant of FPGA-based PID controllers. Therefore, an I2C (Inter-Integrated Circuit) based AD (Analog-to-digital) converter is added as input, while PWM (Pulse width modulation) based output signal is used as an output. The effectiveness of the designed regulator is demonstrated on an example of a DC (direct current) motor control. Additionally, for control and monitoring purposes, the FPGA is connected to the PC using the UART (Universal Asynchronous Receiver Transmitter) protocol. Experimental results indicate that the FPGA-based PID implementation offers solid performance.

The paper presents an automated method for solving traditional single side 2D jigsaw puzzles, focusing solely on shape features. Termed as semi-apictorial puzzles, our approach utilizes pictorial content solely for image segmentation, not for puzzle matching. Through enhancements in background separation, corner extraction, and feature matching, our method simplifies and accelerates puzzle reconstruction. A key contribution is the introduction of an edge matching technique that employs approximate triangles to evaluate a possible match, which notably improves computational efficiency and reduces algorithm complexity. Experimental results demonstrate that the proposed method outperforms existing solutions, enabling the handling of a larger puzzles within a reasonable timeframe.

This paper presents the development and implementation of a flexible industrial machine model for automated visual inspection, called ETFCam, designed to improve the learning outcomes of electrical engineering students in the field of machine vision and robotics. Unlike prefabricated didactic models, which are typically “closed” systems with a predefined set of experiments, custom didactic systems for teaching and training built from scratch tend to be more flexible and provide a deeper insight in engineering, machine design and planning, while being more cost-effective. The proposed system is based on a 3DOF stepper motor-based manipulator, a DC motor driven conveyor, a pneumatic actuated gripper and a machine vision system. The paper discusses several applications of such a system in an educational environment, with a special focus on machine vision applications. Due to the fact that the system is versatile, open, modular, and easy to upgrade, it has unlimited potential and possibilities for further development. In addition, it provides students with a perfect testbed for learning new engineering skills in many areas such as schematic drawing and understanding, PLC based control, sensing, and machine vision.

This paper introduces twisting sliding mode control method (TWSMC) to track 3D trajectories of a quadrotor unmanned aerial vehicle (UAV) exposed to bounded disturbances and perturbations. The key idea behind TWSMC is to introduce a nonlinear twisting term into the sliding surface design, which enables the system to switch between different sliding modes (SMs) smoothly, thereby reducing the chattering phenomenon and improving control performance. Moreover, a high-gain adaptation (HGA) algorithm is adopted in the TWSMC scheme to additionally attenuate the chattering effect, where the switching control gain increases during the convergent phase and decreases in the sliding phase. Through the comprehensive simulation study, it is shown that the proposed approach exhibits improved robustness and performance in tracking a reference under disturbances and perturbations.

This paper presents a full degrees-of-freedom (DOFs) robust control design for a nonlinear quadrotor unmanned aerial vehicle (UAV) operating under bounded disturbances. Second-order sliding modes controllers (SOSMCs) are designed so that the quadrotor UAV can follow a 3D trajectory in the presence of model uncertainties, underactuation, as well as external disturbances that may be matched or mismatched, and vanishing or nonvanishing. The stability analysis of the closed-loop system is presented via the Lyapunov method, showing the finite-time convergence of the system trajectories to the sliding surfaces, as well as the finite-time convergence of the quadrotor position and attitude to their reference values. The high-gain adaptation (HGA) method is adopted in the SOSMC technique, called SOSMC-HGA, to alleviate the chattering phenomenon. Simulation studies in different scenarios demonstrate that the SOSMC technique exhibits superior tracking performance and robustness properties compared to concurrent control methods for tracking reference trajectories of quadrotor UAVs. The simulation results confirm that SOSMC-HGA significantly attenuates the chattering phenomenon in control signals and system states, which is an important improvement, as it increases the safety of UAVs and reduces power consumption.

Control design for multi-rotor aerial vehicles (MAVs) is quite challenging problem due to their nonlinearitles, unknown dynamics, parametric uncertainties, an underactuated property, a nonlinear coupling dynamics and external disturbances. This paper introduces a first order sliding mode control (FOSMC) for robust stabilization of an under-actuated quad-rotor unmanned aerial vehicle (UAV) operating in the presence of external disturbances. The proposed FOSMC guarantees a finite time convergence of the system trajectories to the sliding surface. Obtained simulations show that the FOSM based approach improves robustness properties compared with the concurrent techniques, and enhance tracking performance of the quad-rotor UAV exposed to external disturbances.

Control design for trajectory tracking of multi-rotor aerial vehicles (MAVs) represents a challenging task due to the under-actuated property, highly nonlinear and cross-coupled dynamics, modeling errors, parametric uncertainties and external disturbances. This paper presents the design of the first order sliding mode control (FOSMC) algorithm for trajectory tracking of the octo-rotor unmanned aerial vehicle (UAV) in the presence of various disturbances. The highly nonlinear octo-rotor UAV dynamics is considered via the generalized framework for MAVs modeling. The stability analysis of the closed-loop system is presented using the Lyapunov based approach. The developed FOSMC exhibits finite-time convergence of the octo-rotor trajec-tories to the sliding manifold and the asymptotic stability of the equilibrium in the presence of vanishing disturbances. Simulation studies show a superior tracking performance and robustness properties of the FOSMC in comparison with the concurrent techniques for trajectory tracking of the octo-rotor UAV in the presence of internal and external disturbances.

The paper presents a simultaneous numerical analysis of the geometric and material nonlinearity of the beams. It describes a process of determining the bearing capacity of a stratified cross-section of a beam made of homogeneous and isotropic material in linear and nonlinear domains of material behaviour. Material nonlinearity is analysed by the variation of the cross-sectional stiffness of the beam on bending EI in the stiffness matrix of the system obtained according to the first-order theory. Geometric nonlinearity is introduced into the calculation using the geometric stiffness matrix of the system. Numerical examples present an application of the procedure for solving problems of nonlinear structure analysis. The calculation results obtained in accordance with the procedure described in the paper are compared with the results of the SCIA software package.

This book provides a solution to the control and motion planning design for an octocopter system. It includes a particular choice of control and motion planning algorithms which is based on the authors' previous research work, so it can be used as a reference design guidance for students, researchers as well as autonomous vehicles hobbyists. The control is constructed based on a fault tolerant approach aiming to increase the chances of the system to detect and isolate a potential failure in order to produce feasible control signals to the remaining active motors. The used motion planning algorithm is risk-aware by means that it takes into account the constraints related to the fault-dependant and mission-related maneuverability analysis of the octocopter system during the planning stage. Such a planner generates only those reference trajectories along which the octocopter system would be safe and capable of good tracking in case of a single motor fault and of majority of double motor fault scenarios. The control and motion planning algorithms presented in the book aim to increase the overall reliability of the system for completing the mission.

The paper addresses the problem of detecting pedestrians using three dimensional data acquired by an autonomous mobile robot equipped with an on-board 3D laser scanner. Previous works in this field have dealt with various approaches for combining 2D and 3D range data features for the use in pedestrian classification. In this paper we propose an image processing pipeline for generating a depth image from point clouds data and then localizing object candidates from the depth image. It involves the image segmentation, feature extraction and human classification processes within unstructured dynamic environments. Three different approaches for the detection of pedestrians, vehicles and cyclists using only 3D range data were employed as a part of this system. We train and test the classifiers in an open environment, with presence of multiple pedestrians, cyclists and vehicles, using only point cloud data. The effectiveness and robustness of the proposed system are verified through experiments with real data. This system is also capable to deal with a real-time framerate (10Hz) with high accuracy.