
Trajectory planning in industrial manipulators remains a challenge when attempting to simultaneously balance productivity, motion smoothness, and actuator durability. Traditional approaches often focus on a limited set of metrics, which restricts the ability to comprehensively evaluate robot performance. This work proposes an intelligent multi-objective optimization methodology for trajectory generation, applied to the collaborative robot ABB GoFa5. The approach combines MATLAB-based simulation with a detailed dynamic model of the robot, incorporating the physical properties of its links and enabling collision detection with both the environment and the manipulator itself. Optimization is carried out using a multi-objective particle swarm algorithm, considering three experimental configurations: (i) minimization of execution time and average joint torques, (ii) optimization of execution time together with the maximum velocities reached by each joint, and (iii) an extended formulation that integrates execution time, average joint torques, and maximum velocities, for a total of thirteen objectives. The results demonstrate that the proposed methodology enables the identification of optimal trajectories depending on the prioritized criterion-whether minimizing execution time, reducing maximum joint velocities, or extending manipulator lifespan-while ensuring in all cases smooth, collision-free, and physically efficient solutions.
In this work, a hybrid technique that combines adaptive Differential Evolution and MATPOWER is proposed to solve the economic dispatch problem with flow constraints. The method is evaluated on an 8-bus system under four cases: (i) DC model without power system network constraints; (ii) DC model with a 30 MW restriction on the active power flowing between buses 1 and 2; (iii) AC model without line limits; and (iv) AC model with a 30 MVA restriction on the apparent power between buses 1 and 2. This setup quantifies the impact of including losses, reactive power and line limits on the system cost and losses. In the most realistic case (AC with limits), the operating cost increases by about 4 % with respect to the unconstrained DC model, and active power losses are approximately 17 % higher than in the AC model without line limits. The proposed technique converges in fewer than 200 generations and is competitive against MILP, Genetic Algorithms and classical Differential Evolution.
This paper presents the design, implementation, and validation of a digital twin of the ABB IRB 140 robot, integrating Unity and RobotStudio through TCP/IP communication with the IRC5 controller and programming in RAPID. The proposal combines an immersive virtual reality environment with a simulation in RobotStudio, allowing users to interact safely and intuitively with the robot before executing movements in the actual physical system. The system includes manual control modes, axis manipulation, and trajectory programming, with real-time transmission of joint coordinates to the physical system. Preliminary tests showed stable communication with an average delay of 0.1 s, ensuring synchronisation between the virtual environment and the real robot. From an educational perspective, it is an innovative resource for teaching industrial robotics. Tests with teachers highlighted the ease of use, the fidelity of the simulation, and the motivational impact of the immersive experience. Additionally, it mitigates the typical limitations of physical robots, including availability and operational risks. The digital twin is projected as a valid tool for active and project-based learning, facilitating the understanding of manipulative robotics concepts, programming practice, and safe experimentation in university.
This article presents a teaching innovation tool, specifically an educational video game designed to facilitate the teaching of the fundamental concepts of the discipline of Automatic Control, aimed at engineering students. This tool has been developed using the Matlab programming environment, taking advantage of the capabilities of this specialised software to create an interactive learning experience. In addition, the implementation of this tool in a second-year engineering course is documented, with results showing positive feedback from students regarding the fulfilment of its main objective: to facilitate the understanding of the theoretical and practical concepts of this discipline through an innovative and attractive methodology.
This work addresses the importance of optimizing water use in agriculture, given the increasing water scarcity and the need for sustainable strategies such as regulated deficit irrigation (RDI). It proposes an automated system based on machine learning to estimate water stress levels in 'Arbequina' olive orchards. This method utilizes leaf turgor pressure (LTP) measurements and meteorological data collected between 2014 and 2019, employing techniques such as principal component analysis (PCA) and linear discriminant analysis (LDA) to classify water stress levels. The system aims to facilitate the integration of these tools into commercial irrigation strategies, overcoming the limitations of invasive methods or those dependent on expensive hardware.