Explore the vast range of titles available.

Proton Exchange Membrane Fuel Cells (PEMFCs) enable continuous energy production regardless of environmental conditions due to the storability of hydrogen. When examining the current–power (I–P) curve of a PEMFC under steady-state operating conditions, maximum power is observed at a specific current level. To extract this power, Maximum Power Point Tracking (MPPT) algorithms are employed. These algorithms should feature a simple structure and rapidly track the maximum power point. However, intelligent and optimization-based methods in the literature often involve high computational complexity. In this study, a modified Perturb and Observe (P&O)-based MPPT algorithm is developed to achieve a fast steady-state response under varying PEMFC operating conditions. The proposed algorithm also minimizes power oscillations in the steady state. Its performance is evaluated in a MATLAB/Simulink environment under five different scenarios. A comparative analysis is conducted against the conventional P&O and optimization-based MPPT algorithms, including Particle Swarm Optimization (PSO), Cuckoo Search Algorithm (CSA), and Genetic Algorithm (GA). The results, presented graphically, demonstrate the advantages of the proposed approach.

This paper presents an estimation method for determining the position and orientation of a real mobile robot using raw data from an Inertial Measurement Unit (IMU) sensor, alongside linear and angular velocities obtained from simulation. The dataset was collected using a real TurtleBot3 differential drive wheeled mobile robot in the ROS-Gazebo simulation environment, encompassing 2018 routes-2009 from simulation and 9 from real-world experiments-each consisting of five randomly generated waypoints. To improve the accuracy of the estimation models, noise from the real IMU sensor was incorporated into the input data, and velocities derived from the pure pursuit algorithm were also included. Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Gradient Boosting (GB), and Random Forest (RF) models were employed to estimate the robot’s position and orientation, and their performance was compared across both simulated and experimental scenarios. The results indicate that the CNN architecture consistently outperforms other models across all routes. Unlike many existing studies, this work directly utilizes raw sensor data without applying any feature extraction techniques, highlighting its novelty and contribution to the field.

In industrial robotics, it is crucial to ensure the completion of ongoing processes in the event of a power outage. In this study, a robotic system integrated with a solar panel production system was designed using the ABB RobotStudio program. The energy consumed by the robot during a single cycle was calculated within the same software. Additionally, the energy consumption of the motors in the belt and table system was estimated based on real‐world systems. To address power interruptions, a flywheel energy storage system (FESS) was designed to ensure the continuation of operations. The FESS is capable of supplying the required energy even at the initial start of the robotic system's mission. A notable aspect of this setup is that the drive systems of the motors operate at 800 V. When functioning as a generator, the FESS delivers this voltage to the DC link of the robotic system by acting as a boost converter. The FESS utilizes a high‐speed BLDC motor, and an LC filter is placed between the motor and the inverter. When the motor operates in generator mode, the filter components enable its use as a boost converter. During a single process cycle, the system's speed ranges between 4500 and 3700 r/s. The FESS system was simulated in the MATLAB/Simulink environment, and the results are presented in graphical form.

This paper presents a motor design which can operate directly with a low-voltage output photovoltaic panels or batteries. A high-efficiency synchronous reluctance motor which can operate directly at low voltage level without a boost converter was designed in this study. The motor was optimized for maximum torque and minimum torque ripple by using the multi-objective genetic algorithm. A robust, durable and low-cost motor structure was obtained due to the obtained rotor structure. The optimized motor can generate less than 5% torque ripple with rated torque of 2 Nm. The prototype motor efficiency was obtained as 81.2% in experimental study, while the motor efficiency designed was obtained as 87.9% in theoretical study. For this reason, the designed motor was suitable for the IE5 efficiency class. However, the experimentally produced prototype motor was obtained in the IE4 efficiency class. In addition, the motor drive and control algorithm were developed for the designed motor. The details were analyzed for different load conditions in both simulation and experimental environments.

Stereotactic surgery aims to access critical areas of the brain with high accuracy. The classical surgical process requires two separate radiological imaging datasets (MRI-CT) and their precise registration. Additionally, specific anatomical landmarks (AC, PC, TAL) are manually identified by the neurosurgeon in both datasets, and MRI-CT registration is performed using these landmarks. To address the issues of patients' double exposure to radiological imaging and the manual identification of landmarks, this paper proposes a new approach based on the registration of facial landmarks. The proposed approach consists of four stages. The first stage involves creating 2D facial masks (MRHead and DHead) from the MRI and depth camera data obtained from the patient. Each mask, automatically generated using Google Mediapipe software, consists of 468 points. In the second stage, the mask points are transformed from 2D to 3D. In the third stage, precise registration of the 3D mask points is achieved using singular value decomposition (SVD) and random forest (RF) methods. In the final stage, using the registration matrix, the robotic arm is guided to reach the desired target point on a 3D-printed head prototype. Using the RF method for MHead and DHead mask registration, we obtained fiducial registration error (FRE) values of 1.633 mm and 1.523 mm, and target registration error (TRE) values of 2.217 mm and 2.164 mm for each patient, respectively. These promising results will form the basis of further developments in fully autonomous brain-targeting software with robotic assistance.

In photovoltaic systems, PSC occur when PV panels are exposed to nonuniform solar irradiance levels. Extracting the maximum power from PV systems operating under PSC represents a complex and challenging task for MPPT algorithms. Optimization-based MPPT techniques have therefore gained significant attention due to their ability to achieve fast convergence and high efficiency under such conditions. In this study, a novel M-DHO algorithm is proposed by integrating the DHO algorithm, which is inspired by the cooperative hunting behavior of the Asiatic wild dog, with a Levy flight strategy to enhance global search capability. Especially with Levy flight support, the M-DHO algorithm eliminates the problems of fast convergence and getting stuck in a local minimum. Furthermore, while the fast convergence problem is eliminated, the Levy Flight algorithm allows reaching the global maximum value faster with high accuracy in complex optimization problems. Nine distinct PSC scenarios are created across six different voltage regions, and the performance of the proposed M-DHO algorithm is comparatively assessed against GWO, WOA, FPA, and the conventional DHO algorithm. Simulation results demonstrate that the proposed M-DHO algorithm achieves faster convergence to the global maximum power point and higher tracking efficiency compared to the benchmark algorithms. When averaged over all scenarios, M-DHO achieved an average extracted power of 838.58W, tracking speed of 0.15s and an average tracking efficiency of 99.52%, outperforming other algorithms.

Maximum Power Point Tracking (MPPT) algorithms, which are employed to extract the maximum power from photovoltaic (PV) systems, exhibit different performance characteristics under uniform irradiance and partial shading conditions (PSC) arising from nonuniform solar irradiance distribution on PV panels. Under PSC, the performance of conventional MPPT algorithms becomes inadequate, leading to increased interest in optimization-based approaches. In this study, the Grey Wolf Optimization (GWO) algorithm, commonly used in MPPT applications, was modified, and an Improved Grey Wolf Optimization (IGWO) algorithm was proposed. A PV system model consisting of four series-connected PV panels and a boost converter was developed in the MATLAB/Simulink environment to evaluate the performance of the proposed algorithm. The effectiveness of the algorithm was tested under nine distinct and complex PSC scenarios. The results obtained under these nine PSC cases were analyzed through comparisons of the proposed IGWO algorithm with GWO, the Cuckoo Search Algorithm (CSA), and the Flower Pollination Algorithm (FPA). The results demonstrate that the IGWO algorithm achieves the highest mean maximum power and exhibits superior MPPT performance compared to the other algorithms, with a mean tracking efficiency of 98.34%.

This study evaluates the performance of a pulse‐excited printed circuit board (PCB) resolver in comparison with the traditional square‐wave excitation method. Although the pulse‐excited resolver exhibits slightly higher error levels than the square‐wave excitation approach, these differences remain acceptable for applications that prioritize reduced power consumption and improved thermal performance. Experimental results indicate that the average position errors of the resolver are 1.109° at low speeds and 0.338° at high speeds. These error levels are considered acceptable for a wide range of industrial applications. Thermal analysis demonstrates that the proposed pulse excitation method substantially reduces thermal stress on the resolver, yielding a maximum temperature of approximately 32.0°C during testing, compared with nearly 59.7°C under square‐wave excitation. Overall, the findings highlight the advantages of the proposed method, particularly for applications requiring effective heat management. The results suggest that the pulse‐excited resolver represents a viable alternative, combining operational efficiency with enhanced thermal performance.

This study proposes a novel hybrid learning approach for developing a visual path-following algorithm for industrial robots. The process involves three steps: data collection from a simulation environment, network training, and testing on a real robot. The actor network is trained using supervised learning for 500 epochs. A semitrained network is then obtained at the $250^{th}$ epoch. This network is further trained for another 250 epochs using reinforcement learning methods within the simulation environment. Networks trained with supervised learning (500 epochs) and the proposed hybrid learning method (250 epochs each of supervised and reinforcement learning) are compared. The hybrid learning approach achieves a significantly lower average error (30.9 mm) compared with supervised learning (39.3 mm) on real-world images. Additionally, the hybrid approach exhibits faster processing times (31.7 s) compared with supervised learning (35.0 s). The proposed method is implemented on a KUKA Agilus KR6 R900 six-axis robot, demonstrating its effectiveness. Furthermore, the hybrid approach reduces the total power consumption of the robot’s motors compared with the supervised learning method. These results suggest that the hybrid learning approach offers a more effective and efficient solution for visual path following in industrial robots compared with traditional supervised learning.

PurposeBecause of the increased use of robots in the industry, it has become inevitable for humans and robots to be able to work together. Therefore, human security has become the primary noncompromising factor of joint human and robot operations. For this reason, the purpose of this study was to develop a safe human-robot interaction software based on vision and touch.Design/methodology/approachThe software consists of three modules. Firstly, the vision module has two tasks: to determine whether there is a human presence and to measure the distance between the robot and the human within the robot’s working space using convolutional neural networks (CNNs) and depth sensors. Secondly, the touch detection module perceives whether or not a human physically touches the robot within the same work environment using robot axis torques, wavelet packet decomposition algorithm and CNN. Lastly, the robot’s operating speed is adjusted according to hazard levels came from vision and touch module using the robot’s control module.FindingsThe developed software was tested with an industrial robot manipulator and successful results were obtained with minimal error.Practical implicationsThe success of the developed algorithm was demonstrated in the current study and the algorithm can be used in other industrial robots for safety.Originality/valueIn this study, a new and practical safety algorithm is proposed and the health of people working with industrial robots is guaranteed.