This paper is concerned with the design and implementation of a robust position backstepping tracking controller for a permanent magnet synchronous motor (PMSM). The information on the angular position and velocity, provided by a classical resolver-to-digital converter, additionally employs a phase lock loop (PLL) circuit. A backstepping control law is designed from the input-output linearization of the PMSM model, written in d-q coordinates. This controller is adapted via the online estimation of the unknown load torque and friction effects. A linear extended state observer is devised for this purpose, thus ensuring high closed-loop performance of the motor trajectory tracking task. An input-state stability analysis of the entire system is also provided. Cosimulation via the MATLAB/Simulink-PSIM package, including realistic measurement disturbances, is used to investigate the stability and accuracy of the proposed control algorithm. Experimental results are provided.

In this work, the sensorless speed control of a brushless direct current motor utilizing a neural network is presented. This control is done using a two-layer neural network that uses the backpropagation algorithm for training. The values provided by a Proportional, Integral, and Derivative (PID) control to this type of motor are used to train the network. From this PID control, the velocity values and their corresponding signal control (u) are recovered for different values of load pairs. Five different values of load pairs were used to consider the entire working range of the motor to be controlled. After carrying out the training, it was observed that the proposed network could hold constant load pairs, as well as variables. Several tests were carried out at the simulation level, which showed that control based on neural networks is robust. Finally, it is worth mentioning that this control strategy can be realized without the need for a speed sensor.

The article deals with the implementation of an electronic differential system (EDS) for a four-wheel drive (4WD) electric vehicle (EV), using in-wheel brushless DC motors. Applying a graph theory analysis, we design a decentralized control scheme based on active disturbance rejection control (ADRC) for each wheel, with the purpose of all motors work together to follow the angular speed references in a synchronized way when driving in a straight line and at different speeds in a curve. The angular speed of each wheel is calculated from a reference turning angle and speed for the EV using the steering Ackermann model. Simulation results are presented to show the feasibility of the scheme proposed.

In the Yucatan Peninsula, there is a rich diversity of mangroves, notably including Rhizophora mangle, Avicennia germinans, and Laguncularia racemosa. These mangroves contribute to the recovery of degraded natural areas caused by human activities. Additionally, they serve as natural habitats for various animal and plant species. Studies have highlighted the significance of preserving and restoring these species through traditional methods. More recently, the integration of remote sensing and deep learning techniques has allowed for the automated detection and quantification of mangroves. In this study, we explore the application of deep neural network techniques to address computer vision challenges in the field of remote sensing. Specifically, we focus on the detection and quantification of mangroves in remote image sensing, employing transfer learning and fine-tuning with three distinct deep neural network architectures: SegNet-VGG16, U-Net, and Fully Convolutional Network (R-FCN), with the latter two based on the ResNet network. To evaluate the performance of each architecture, we applied key evaluation metrics, including Intersection over Union (IoU), Dice Coefficient, Precision, Sensitivity, and Accuracy. Our results indicate that SegNet-VGG16 exhibited the highest levels of Precision (98.03%) and Accuracy (97.03%), while U-Net outperformed in terms of IoU(96.97%), Dice Coefficient (92.20%), and Sensitivity (96.81%).

La predicción del ángulo de dirección en vehículos inteligentes es un tema ampliamente estudiado en la literatura. Sin embargo, se requiere la intervención de un usuario para operar al vehículo, limitando el grado de autonomía. Este trabajo propone algoritmos basados en redes neuronales artificiales para determinar el ángulo de giro de un vehículo en movimiento, con aplicaciones en tiempo real para vehículos a escala en una pista de dos carriles. Utilizando una cámara frontal centrada en el vehículo se determinan dos áreas de interés: la imagen original y una versión rectificada mediante un cambio de perspectiva. Posteriormente, se emplea la tupla de imágenes para construir dos bases de datos etiquetadas con el ángulo deseado de giro, las cuales, son empleadas para entrenar dos redes neuronales convolucionales. Una vez entrenadas, ambos modelos son implementados en tiempo real para guiar un vehículo a escala en una pista libre de obstáculos, permitiendo su navegación autónoma dentro del carril designado y observando el comportamiento en cada caso.

Este trabajo presenta un enfoque para la navegación autónoma de robots móviles en entornos interiores desconocidos, integrando los algoritmos SLAM y PRM. El método propuesto permite tanto alcanzar un objetivo específico como explorar áreas no mapeadas, gestionando situaciones en las que el robot pueda quedar atrapado y optimizando las trayectorias mediante la replanificación de rutas. Mientras navega, el robot actualiza continuamente el mapa del entorno y ajusta su trayectoria en función de la información obtenida. La efectividad del enfoque fue validada mediante simulaciones en ROS Gazebo y pruebas experimentales con un TurtleBot 2, demostrando su aplicabilidad en tareas de exploración, como en áreas afectadas por desastres naturales.