Bartek Łukawski

Mr. Bartek Łukawski

Universidad Carlos III de Madrid, Spain
orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
1 April 2020
Verified
Conference Paper
Closed access
Edwin Daniel Oña SimbañaEdwin Daniel Oña Simbaña
Edwin Daniel Oña Simbaña
BŁBartek Łukawski
Bartek Łukawski
Alberto Jardón HueteAlberto Jardón Huete
Alberto Jardón Huete
CBCarlos Balaguer
Carlos Balaguer
This paper presents the implementation of a modular framework that allows the integration of different peripheral components towards facilitating the control of an assistive robotic arm. For that purpose, several components, such as proximity sensors, an RGB camera and a joystick, were integrated into the same control paradigm of the robotic arm. Firstly, the proximity sensors are arranged on the robot casing to sense the nearby workspace. This configuration is helpful to detect obstacles in robot trajectories and ensure the user’s safety. Secondly, an RGB camera was fixed in the robot end-effector to capture visible information in order to assist the object’s grasping via a visual servoing (VS) strategy. Finally, a multimodal approach was considered to increase the input options for moving the robot by off-the-shelf joysticks. All of the above components were integrated into the control strategy through a software-based modular architecture. The proposed framework aims to provide an intuitive, reliable, and modular solution to facilitate the control of a robotic manipulator in task performing, and extendable to people with disabilities.

11 August 2024 in Robotics
Verified
Journal Article
Open access
PDF file
JHJuan Hernández
Juan Hernández
BŁBartek Łukawski
Bartek Łukawski
SCSantiago Martinez de la Casa
Santiago Martinez de la Casa
NTNikos Tsagarakis
Nikos Tsagarakis
CBC. Balaguer
C. Balaguer
Humanoid robots often struggle with tasks such as lifting objects due to the complexity involved in identifying contact points, applying the correct force, and tracking task progress. We propose an integrated solution that leverages the dual-arm capability of humanoids and utilizes sensor fusion from vision and force sensors. Our system employs a computer vision algorithm to detect and characterize object properties (shape, size, position, orientation) and differentiate between parallel and non-parallel bi-manipulation tasks. The controller then identifies optimal contact points for the end effectors, generating trajectories fed into a closed-loop controller using force feedback. For parallel bi-manipulation, momentum cancellation is achieved through sensor fusion. For non-parallel surfaces, a reinforcement learning algorithm determines the appropriate lifting force to prevent slippage using only two contact points. Experimental validation on a real humanoid platform demonstrates the effectiveness of our approach in autonomously lifting objects, regardless of contact surface configuration. This advancement significantly enhances the reliability and versatility of humanoid robots in performing complex manipulation tasks, contributing to their practical deployment in human-oriented environments.

27 August 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
Andrea Gil RuizAndrea Gil Ruiz
Andrea Gil Ruiz
JVJuan G Victores
Juan G Victores
BŁBartek Łukawski
Bartek Łukawski
CBC. Balaguer
C. Balaguer
The inspection of Personal Protective Equipment (PPE) is one of the most necessary measures when treating patients affected by infectious diseases, such as Ebola or COVID-19. Assuring the integrity of health personnel in contact with infected patients has become an important concern in developed countries. This work focuses on the study of Reinforcement Learning (RL) techniques for controlling a scanner prototype in the presence of blood traces on the PPE that could arise after contact with pathological patients. A preliminary study on the design of an agent-environment system able to simulate the required task is presented. The task has been adapted to an environment for the OpenAI Gym toolkit. The evaluation of the agent’s performance has considered the effects of different topological designs and tuning hyperparameters of the Q-Learning model-free algorithm. Results have been evaluated on the basis of average reward and timesteps per episode. The sample-average method applied to the learning rate parameter, as well as a specific epsilon decaying method worked best for the trained agents. The obtained results report promising outcomes of an inspection system able to center and magnify contaminants in the real scanner system.

1 September 2025 in Jornadas de Automática
Verified
Journal Article
Open access
PDF file
BŁBartek Łukawski
Bartek Łukawski
MCMiquel Martín Cervera
Miquel Martín Cervera
CPCarmen Menchén Pascual
Carmen Menchén Pascual
Edwin Daniel Oña SimbañaEdwin Daniel Oña Simbaña
Edwin Daniel Oña Simbaña
JVJuan G Victores
Juan G Victores
Alberto Jardón HueteAlberto Jardón Huete
Alberto Jardón Huete
Presentamos un ejercicio práctico que involucra el brazo colaborativo robótico ABB “GoFa” CRB 15000-5: una estación de café para una cafetera doméstica convencional con cápsulas. El objetivo es proporcionar un caso de uso práctico e intuitivo para su adopción en sesiones prácticas con estudiantes de grado o máster. Esperamos enriquecer de este modo dichas sesiones, mostrando algunos mecanismos que ofrece el entorno de programación de ABB de cara al desarrollo de interfaces gráficas personalizadas. Exploramos AppStudio, la herramienta software de reciente introducción (enero de 2025) para crear aplicaciones gráficas, así como la más consolidada OmniCore App SDK. En este trabajo, describimos dos nuevas aplicaciones que demuestran la validez de ambas herramientas a través de experimentos tanto sobre el simulador RobotStudio como en un montaje real.

28 May 2026 in Simposios del Comité Español de Automática (CEA)
Verified
Journal Article
Open access
PDF file
ÁSÁlvaro Santos-García
Álvaro Santos-García
BŁBartek Łukawski
Bartek Łukawski
DCDorin Sabin Copaci
Dorin Sabin Copaci
JVJuan G Victores
Juan G Victores
CBC. Balaguer
C. Balaguer
Trabajos previos establecieron una arquitectura de teleoperación para brazos robóticos sobre YARP y ROS 2. El presente trabajo extiende dicha arquitectura incorporando retroalimentación háptica para cerrar un bucle de control bilateral. Un dispositivo Phantom Omni transmite comandos de posición al robot y recibe fuerzas de contacto del entorno remoto. La integración comprende tres capas: YARP para la abstracción hardware del dispositivo háptico, ROS 2 como backbone de comunicación, y una capa de control específica por plataforma, que emplea la pila de control cartesiano para el robot humanoide TEO y el protocolo EGM de ABB para el brazo colaborativo GoFa. Para evitar calibración espacial explícita entre marcos de referencia, se adopta una estrategia de control diferencial basada en desplazamientos incrementales. La retroalimentación de fuerzas se implementa mediante un nodo dedicado que evalúa la penetración del extremo del robot en regiones de restricción virtual y genera fuerzas de restitución proporcionales, cerrando el bucle bilateral de forma independiente a la plataforma objetivo.