This paper is part of the "Hand of Hope" research project, which seeks to develop low-cost robotic prostheses for people with motor disabilities in their upper limbs. The objective of this research is to design and implement a fractional order PID embedded controller (FOPID), applied to the speed regulation of the elbow joint based on DC motor, in a robotic prosthesis. The set of parameters for the tuning of a FOPID controller are five: proportional constant, integration constant, derivative constant, and non-integer orders of the integral and derivative terms (K _P , K _I , K _D , λ, μ). In order to fulfill the objective, three phases were executed: first, the several methodologies involved in the design and implementation of FOPID controllers were applied in an autonomous embedded system; then, they were implemented and applied to a mechatronic system; finally, the performance of the fractional controllers was analyzed and compared, the comparison of the controllers was performed based on the time parameters (settling time, overshoot), and the performance indexes (IAE, ITAE, ISE, and ITSE). As results of the present work: 14 types of fractional controllers were designed, of which 9 were implemented in an embedded system based on the multicore FPGA/Processor architecture, and the programming of the mathematical control functions were developed in LabVIEW FPGA and RT; after the experimental evaluation, the embedded controller with higher performance was the FOPID Ziegler-Nichols/Nelder-Mead/ISE, the results obtained were Ts = 0.22s and PO = 7.18%.

This paper shows the implementation and performance analysis of motion control algorithms for a 3D Concrete Printing system based on an industrial robotic arm. This work is part of the project of digital fabrication of low-cost housing. Regarding the technological architecture integrated in the present work, the hardware devices used are the EPSON C12 industrial robot arm and the RC700-A controller, and the software tools are Fusion 360 and RC+. To evaluate the point extraction and sequencing algorithms of a 3D structure and the motion control algorithms of the robot arm, three test wall models were designed: a semi-circle with horizontal undulations, a semi-circle with vertical undulations and an orthohedron without undulations. For the performance analysis, 140 trajectory times were extracted for each test model. The extracted values are the trajectory time intervals of the wall model envelope for each layer and of the internal trajectory of the wall model infill. Due to the increasing and decreasing trends of time for specific cases, it was concluded that in the working area of the robot arm, there are parts in which the robot is more efficient and therefore the joints offer less inertia for certain types of movements, for example straight or curved movements and short or long movements.

Industrial robots are used in a variety of industrial process tasks, and due to the complexity of the environment in which these systems are deployed, more robust and accurate control methods are required. Deep reinforcement learning emerges as a comprehensive approach that directly allows for the mapping of sensor data and the setting of motion actions to the robot. In this work, we propose a robotic system implemented in a semi-photorealistic simulator whose motion control is based on the A2C algorithm in a DRL agent; the task to be performed is to reach a goal within a work area. The evaluation is executed in a simulation scenario where a fixed position of a target is maintained while the agent (robotic manipulator) tries to reach it with the end-effector from an initial position. Finally, the trained agent fulfills the established task; this is demonstrated by the results obtained in the training and evaluation processes, and the reward value increases when the measured distance decreases between the end-effector and the target.

The manufacturing, assembly, and packaging processes of industrial hinges must improve their productivity. To address this challenge, a technological system based on a robot arm and vision machine was developed, integrated, and evaluated. The devices that make up the hardware architecture are the following: the Epson VT6L six-degrees-of-freedom industrial robot arm, the Compact Vision CV2-HA image-processing device, the Basler acA1600-20gc camera, and the end effector based on 3 kg holding force electromagnets. The software elements used for the development of this automated system are the SPEL+ programming language and VisionGuide tool, both of which are integrated into the Epson RC+ robotic system development environment. For the performance evaluation of the hinges and packaging identification algorithms, 80 hinges of each type were used as input to the robotic system workstation. As a result, the following results were obtained: (1) the hinge identification algorithms were 100% correct, (2) the primary packaging identification algorithm was 100% correct, and (3) the hinge assembly algorithm was 92.5% correct because it had six errors. These results provide evidence of the effectiveness of the developed system. In addition, the motion times of the robot arm were analyzed in detail to identify opportunities for improving the automated production process.

In the industrial field, it is necessary to digitize the operation of machines. It is also necessary for a machine operator to remotely monitor and control some of its operating parameters. There are devices and systems based on the Internet of Things (loT) that would contribute to solving this requirement; however, both its mobile application and cloud platform involve costs associated with license maintenance. Based on the above, the present work addresses the design of a native Android mobile application to remotely monitor and control a three-phase AC motor through a low-cost loT hardware system because the hardware and software elements used for the development are open-source. Agile Scrum methodology was used to develop a native mobile application. The hardware architecture of the loT system consists of an INVT Goodrive 10 variable speed drive, a Serial TTL-RS485 electronic converter card, an ESP-12E NodeMCU programmable device, and a smartphone. Six types of tests were executed to evaluate the overall performance of the loT system software: unit, validation, load, performance, black box, and white box tests. The tools used in these tests were App ControlMotor, Android Studio, Visual Studio Code, LoadFocus, JUnit 4, and use-case cards. The results obtained were used to verify the user's compliance with functional requirements and specifications.

The poor management of urban solid waste (USW) causes a series of direct problems in cities. For example, it produces unpleasant odors, favors the presence of rodents and insects that are disease vectors, and produces dirt in the urban landscape. The urban solid waste collection and transportation system in Loja-Ecuador has some problems such as non-compliance with schedules and coverage of collection routes. This demonstrates the need to improve the management of USW in cities. Based on previous information, this project aims to develop an automatic localization system for urban solid waste collection and transportation vehicles. The designed system is structured into three modules: the processing module, the data acquisition and sending module, and the remote server module, which consists of a remote connection unit, database, and web application for visualization. To evaluate the performance of the proposed system, field tests were conducted by selecting two solid-waste collection and transportation routes. The data were compared with a Garmin Monterra reference GPS, and the errors were calculated using the Haversine function. In Test 1 (12.7 km), an average total error of 3.35 m was obtained, and in Test 2 (15.6 km), an average total error of 3.37 m was obtained; these error levels are acceptable for the type of system implemented.