Alberto Mendez

Mr. Alberto Mendez

orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 July 2024 in IEEE Access
Verified
Journal Article
Open access
AMAlberto Mendez
Alberto Mendez
Adrian PradosAdrian Prados
Adrian Prados
EMElisabeth Menendez
Elisabeth Menendez
Ramon BarberRamon Barber
Ramon Barber
The rearrangement of objects is an essential task in daily human life. Subconsciously, humans break down such tasks into three components: perception, reasoning, and execution, which are automatically resolved. This process represents a significant challenge for robots, as they must apply complex logic to treat all the information and successfully execute the task. In this research, we propose a solution to perform this task in a human-like manner. For that purpose, we developed a modular framework that provides the capability to observe and understand the scene, imagine the best solution and execute it, following human-like reasoning. This is done by combining a zero-shot deep learning model for perception, a zero-shot large diffusion model to provide an ordered and realistic final scene and a Learning from Demonstration algorithm for execution. To test the performance, we conducted several experiments to check the correct resolution of 2D rearrangement tasks. For that purpose, we have tested the feasibility of the final generated scene, the ability to generate the trajectory by means of human demonstrations and, finally, we have carried out experiments with two different robots in a simulated and a real environment. The results obtained prove the adaptability of our framework to different environments, objects and robots. Moreover, the success rate of solutions provided and the error in the position and orientation demonstrate a significant advance in the accuracy and effectiveness of solving rearrangement tasks.

11 November 2022 in Sensors
Verified
Journal Article
Open access
PDF file
Alicia Mora VelascoAlicia Mora Velasco
Alicia Mora Velasco
Adrian PradosAdrian Prados
Adrian Prados
AMAlberto Mendez
Alberto Mendez
Ramon BarberRamon Barber
Ramon Barber
Santiago GarridoSantiago Garrido
Santiago Garrido
Mobile robot navigation has been studied for a long time, and it is nowadays widely used in multiple applications. However, it is traditionally focused on two-dimensional geometric characteristics of the environments. There are situations in which robots need to share space with people, so additional aspects, such as social distancing, need to be considered. In this work, an approach for social navigation is presented. A multi-layer model of the environment containing geometric and topological characteristics is built based on the fusion of multiple sensor information. This is later used for navigating the environment considering social distancing from individuals and groups of people. The main novelty is combining fast marching square for path planning and navigation with Gaussian models to represent people. This combination allows to create a continuous representation of the environment from which smooth paths can be extracted and modified according to dynamically captured data. Results prove the practical application of the method on an assistive robot for navigating indoor scenarios, including a behavior for crossing narrow passages. People are efficiently detected and modeled to assure their comfort when robots are around.

1 September 2025 in Jornadas de Automática
Verified
Journal Article
Open access
PDF file
Adrian PradosAdrian Prados
Adrian Prados
GEGonzalo Espinoza
Gonzalo Espinoza
AMAlberto Mendez
Alberto Mendez
Alicia Mora VelascoAlicia Mora Velasco
Alicia Mora Velasco
Santiago GarridoSantiago Garrido
Santiago Garrido
Ramon BarberRamon Barber
Ramon Barber
Este artículo presenta ADAMSim, un entorno de simulación basado en PyBullet diseñado específicamente para el Ambidextrous Domestic Autonomous Manipulator (ADAM), desarrollado para apoyar la investigación en navegación, manipulación y aprendizaje en robótica doméstica. El simulador replica con precisión la estructura y el comportamiento del robot físico, lo que permite una transferencia robusta de algoritmos entre simulación y mundo real. ADAMSim sigue un diseño modular, que incluye navegación, cinemática de brazos y manos, percepción y comunicación mediante ROS. Esta arquitectura permite la operación sincronizada entre el robot real y su gemelo digital. Se desarrollaron diversos ejemplos, que abarcan desde tareas de visión y agarre, hasta navegación y teleoperación, incluyendo experimentos ejecutados simultáneamente en el robot simulado y el real. Su diseño de código abierto y flexible convierte a ADAMSim en una herramienta poderosa para el desarrollo seguro y reproducible de algoritmos y experimentos en robótica doméstica. La plataforma también está pensada para apoyar investigaciones en mapeo de interiores, aprendizaje de manipulación avanzada y proyectos educativos, como banco de pruebas.

1 July 2017 in Energy Procedia
Verified
Journal Article
Open access
MAMónica Alonso
Mónica Alonso
BABorja Arias
Borja Arias
AMAlberto Mendez
Alberto Mendez
FFFernando Fuentes
Fernando Fuentes
JAJuan Carlos Abanades
Juan Carlos Abanades

8 January 2025
Verified
Book Chapter
Open access
PDF file
NFNoelia Fernandez
Noelia Fernandez
GEGonzalo Espinoza
Gonzalo Espinoza
AMAlberto Mendez
Alberto Mendez
Adrian PradosAdrian Prados
Adrian Prados
Alicia Mora VelascoAlicia Mora Velasco
Alicia Mora Velasco
Ramon BarberRamon Barber
Ramon Barber
In recent years, the capabilities of mobile robots to interact with humans and their environment have been enhanced by emerging technological advances. The improvement in the quantity and quality of information from robot sensors, along with increased computational capacities, has enabled the development of new algorithms to model the human environment in which the robot moves and interacts. This chapter will describe modelling techniques for scenes and objects that compose the environment, with the aim of generating an internal representation of the robot’s surroundings so it can navigate and perform manipulation tasks in shared environments with humans. Recent advances in environment modelling driven by deep learning and machine learning techniques, using sensor information, will be presented. Additionally, new trends in the generation of simulated environments, their elements, and synthetic databases will be studied. These trends aim to leverage advances in artificial intelligence to expand the quantity and variety of available data, facilitating the process of learning and understanding the environment.

28 May 2026 in Simposios del Comité Español de Automática (CEA)
Verified
Journal Article
Open access
PDF file
GEGonzalo Espinoza
Gonzalo Espinoza
LLLucía Lishan
Lucía Lishan
AMAlberto Mendez
Alberto Mendez
Adrian PradosAdrian Prados
Adrian Prados
Ramon BarberRamon Barber
Ramon Barber
LMLuis Moreno
Luis Moreno
El agarre de objetos desconocidos en entornos domésticos supone uno de los principales retos de los robots asistenciales, donde los manipuladores deben ser capaces de adaptarse a la gran variedad de formas y tamaños de los objetos y la incertidumbre de un entorno desestructurado. En este artículo se presenta una implementación de un método end-to-end de agarre no basado en aprendizaje aplicado al robot asistencial bibrazo ADAM en este tipo de condiciones. La implementación cubre el proceso de agarre completo, desde la detección y reconstrucción 3D de los objetos a partir de formas supercuádricas, la planificación del agarre mediante métodos analíticos y finalmente la selección y planificación del brazo y la ejecución. El método ha sido evaluado en un simulador basado en PyBullet a partir de la reconstrucción de una escena real, sobre objetos de uso cotidiano doméstico y de formas y tamaños variados, demostrando la viabilidad del flujo completo, también transferido a la plataforma robótica real.