Pure monocular 3D reconstruction is a complex problem that has attracted the research community’s interest due to the affordability and availability of RGB sensors. SLAM, VO, and SFM are disciplines formulated to solve the 3D reconstruction problem and estimate the camera’s ego-motion; so, many methods have been proposed. However, most of these methods have not been evaluated on large datasets and under various motion patterns, have not been tested under the same metrics, and most of them have not been evaluated following a taxonomy, making their comparison and selection difficult. In this research, we performed a comparison of ten publicly available SLAM and VO methods following a taxonomy, including one method for each category of the primary taxonomy, three machine-learning-based methods, and two updates of the best methods to identify the advantages and limitations of each category of the taxonomy and test whether the addition of machine learning or updates on those methods improved them significantly. Thus, we evaluated each algorithm using the TUM-Mono dataset and benchmark, and we performed an inferential statistical analysis to identify the significant differences through its metrics. The results determined that the sparse-direct methods significantly outperformed the rest of the taxonomy, and fusing them with machine learning techniques significantly enhanced the geometric-based methods’ performance from different perspectives.

This paper proposes a heuristic procedure created to improve the collection routes obtained with the support of the ArcGIS software and its complement, Network Analyst. After a series of experiments, it was found that the software presents several inconsistencies with logistical and operational management concepts, such as the unnecessary realization of U-turns in a dead end and unnecessary access to areas with difficult access to a single customer. These are issues that a collection route planner must consider to make a good decision that considers the cost of visiting areas with difficult access and the benefits of reaching that area. In this sense, our heuristic procedure considers a set of best practices in operational and logistical strategies that could be programmed within the Network Analyst. As it is well known in the science of vehicle routing, U-turns and sub-tours in the routes travelled by vehicles increase distances and operating costs, so in our work, we propose a systematic heuristic procedure to reduce the number of U-turns performed by a municipal waste collection truck, while using the ArcGIS-Network Analyst add-on to reduce the number of sub-tours in the route under the Directed-Capacitated Arc Routing Problem approach. It is then shown how the routes improved using our conceptual heuristic procedure, which provides better quality than the original routes obtained with ArcGIS and Network Analyst. Specifically, reducing the total distances travelled by the vehicle fleet, increasing the coverage of sidewalks visited by the truck in the urban perimeter of a city, and minimizing the time used in municipal solid waste collection operations. The importance of our work lies in the fact that effective MSW management is an essential municipal service whose regulation can drive innovation, sustainability, and efficiency in the MSW sector.

The purpose of this work is to combine the concepts of sliding mode control and internal model control to design a Dynamical Sliding Mode Control (DMSC). The Internal Model Control (IMC) structure is used to obtain a simple and easy way to design the controller. This new approach eliminates the major sliding mode control problem: chattering. The process is reduced to a first-order-plus-deadtime (FOPDT) model, and it is divided in two components an invertible one and other non-invertible. A lead term is added to the invertible term and a lead-lag model is obtained and used to get the desired controller. The approach is tested in a high order system with elevated deadtime. The performance of the controller is tested under tracking and regulation conditions: change in the set point, persistent disturbance applied at a determined time, and addition of noise. Finally, a robustness test is conducted to prove how well the controller works when the plant ages. In every of these cases, the performance of the controller is satisfactory, also neither chattering nor instabilities are presented.

The objective of this study is to develop a kinetic model that incorporates time-delay mechanisms to describe the gradual release behavior of pectin. Three models were used: modified Peleg, logistic, and second-order linear with time delay (SOPDT), to represent the extraction kinetics of grapefruit peels at 60, 70, and 90 °C. Model fitting was performed using a hybrid methodology combining Monte Carlo simulations with genetic algorithms. The SOPDT model provided the best fit, with the lowest squared error values (J = 7.06 at 60 °C, 5.63 at 70 °C, and 8.71 at 90 °C), requiring only two parameters to represent the entire process. In contrast, the other models required six. The maximum pectin yield reached over 120% at 90 °C, compared to <80% at lower temperatures. These results demonstrate the potential of the SOPDT model for efficient process description and control.

Pure monocular 3D reconstruction is an ill-posed problem that has attracted the research community's interest due to the affordability and availability of RGB sensors. SLAM, VO, and SFM are disciplines formulated to solve the 3D reconstruction problem and estimate the camera’s ego-motion, so many methods have been proposed. However, most of these methods were not evaluated in large datasets, under various motion patterns, had not been tested under the same metrics, and most of them had not been evaluated following a taxonomy, making their comparison and selection difficult. In this research, we performed a comparison of ten publicly available SLAM and VO methods following a taxonomy, including one method for each category of the primary taxonomy, three machine learning-based methods, and two updates of the best methods, to identify the advantages and limitations of each category of the taxonomy and test if the addition of machine learning or the updates made on those methods improved them significantly. Thus, we evaluated each algorithm under the TUM-Mono benchmark and performed an inferential statistical analysis to identify significative differences through its metrics. Results determined that sparse-direct methods significantly outperformed the rest of the taxonomy, and fusing them with machine learning techniques significantly improves the performance of geometric-based methods from different perspectives.

A dynamic neck model is proposed for functional assessment or ergonomic studies using data obtained from conventional biomechanical tools such as video photogrammetry and force platforms. The head and neck inertial parameters are obtained through regression equations and refined through a calibration process to improve accuracy. Head movement is recorded through video photogrammetry, where marker coordinates are used to calculate finite displacements, linear and angular velocities, and accelerations. An inverse dynamics approach is used to estimate the forces and moments at the C7 vertebral level and the generated muscle power. The model was validated through an experimental study with 30 participants, where its estimates were compared to the measurements obtained from a force platform. The comparison aimed to assess the accuracy and reliability of the model. The results show excellent agreement, with a correlation of 0.976 or higher and a standard error of less than 1% of the head weight Keywords: Biomechanics, biomechanical model, neck, ergonomics, functional assessment Se presenta un modelo dinámico del cuello para su aplicación en valoración funcional o estudios de ergonomía, a partir de datos obtenidos mediante equipamiento biomecánico convencional: sistema de video fotogrametría y plataformas de fuerzas. El modelo utiliza parámetros inerciales de la cabeza y el cuello obtenidos mediante ecuaciones de regresión que se ajustan posteriormente en un proceso de calibración. El movimiento de la cabeza se registra mediante un sistema de video fotogrametría convencional. A partir de las coordenadas de los marcadores se calculan los desplazamientos finitos, velocidades y aceleraciones lineales y angulares. Las fuerzas externas se miden con una plataforma de fuerzas. Mediante un planteamiento de dinámica inversa se estiman fuerzas y momentos a nivel de C7, así como la potencia desarrollada. El modelo se ha validado mediante un estudio experimental con 30 participantes, comparando las estimaciones del modelo con las medidas en una plataforma de fuerzas. Los resultados muestran una correspondencia excelente entre las estimaciones y las medidas experimentales, con una correlación superior a 0.976 y un error estándar inferior al 1% del peso de la cabeza. Palabras clave: Biomecánica, modelo biomecánico, cuello, ergonomía, valoración funcional