A5050654737- Robot Manipulation and Learning
- Social Robot Interaction and HRI
- Distributed Control Multi-Agent Systems
- Metaheuristic Optimization Algorithms Research
- Modular Robots and Swarm Intelligence
- Robotic Path Planning Algorithms
- Medical Image Segmentation Techniques
- Hand Gesture Recognition Systems
- Neural Networks and Applications
- AI in cancer detection
- Video Surveillance and Tracking Methods
- Energy Efficient Wireless Sensor Networks
- Speech and dialogue systems
- Image Retrieval and Classification Techniques
- Robotics and Automated Systems
- Remote-Sensing Image Classification
- Image and Signal Denoising Methods
- UAV Applications and Optimization
- Robotics and Sensor-Based Localization
- IoT and Edge/Fog Computing
- Image Enhancement Techniques
- Infrared Thermography in Medicine
- Gaze Tracking and Assistive Technology
- Reinforcement Learning in Robotics
- Autonomous Vehicle Technology and Safety
Ficosa International (Spain)
2022-2025
Military Medical Academy
2020
Universitat Politècnica de Catalunya
2015-2018
Institut de Robòtica i Informàtica Industrial
2015-2018
École Supérieure des Technologies Industrielles Avancées
2014-2015
Robosoft (France)
2012-2014
Universidad Politécnica de Madrid
2007-2013
Universidad Complutense de Madrid
2011
Centre for Automation and Robotics
2009
In this paper, we propose the distributed bees algorithm (DBA) for task allocation in a swarm of robots. proposed scenario, consists assigning robots to found targets 2-D arena. The expected distribution is obtained from targets' qualities that are represented as scalar values. Decision-making mechanism and autonomously choose their assignments taking into account distances. We tested scalability DBA terms number targets. For that, experiments were performed simulator various sets...
Three advanced natural interaction modalities for mobile robot guidance in an indoor environment were developed and compared using two tasks quantitative metrics to measure performance workload. The first modality is based on direct physical requiring the human user push order displace it. second third exploit a 3-D vision-based human-skeleton tracking allowing guide by either walking front of it or pointing toward desired location. In task, participants asked between different rooms...
Robot-assisted dressing is performed in close physical interaction with users who may have a wide range of characteristics and abilities. Design user adaptive personalized robots this context still indicating limited, or no consideration, specific user-related issues. This paper describes the development multimodal robotic system for scenario-putting on shoe, where users' inputs contribute to much improved task success rate. We developed: 1) tracking, gesture recognition, posture recognition...
Multi-sensor systems can play an important role in monitoring tasks and detecting targets. However, real-time allocation of heterogeneous sensors to dynamic targets/tasks that are unknown a priori their locations priorities is challenge. This paper presents Modified Distributed Bees Algorithm (MDBA) developed allocate stationary upcoming using decentralized, swarm intelligence approach minimize the task detection times. Sensors allocated based on sensors' performance, tasks' priorities,...
Conventional image edge detectors always result in missing parts of the edges. Broken linking is an improvement technique that complementary to detection, where broken edges are connected form closed contours order separate regions image. In this paper, Ant System (AS) algorithm modified for problem. As input, a binary obtained after applying Sobel operator used. The proposed method defines novel fitness function dependent on two variables: grayscale visibility pixels and length connecting...
In many complex robotics systems, interaction takes place in all directions between human, robot, and environment.Performance of such a system depends on this interaction, proper evaluation must build modeling relevant set performance metrics, methodology to combine metrics into single value.In paper, existing models human-robot are adapted fit scenarios with one or several humans robots.The the process is formalized, general method fuse values over time for presented.The resulting value,...
Assistive robots have a great potential to address issues related an ageing population and increased demand for caregiving. Successful deployment of working in close proximity with people requires consideration both safety human-robot interaction. One the established activities daily living where could play assistive role is dressing. Using correct force profile robot control will be essential this application interaction requiring careful exploration factors user's pose type garments...
In this paper, Ant Colony System (ACS) algorithm is applied for edge detection in grayscale images. The novelty of the proposed method to extract a set images from original image using Multiscale Adaptive Gain contrast enhancement and then apply ACS detect edges on each extracted resulting represents pheromone trails matrices which are summed produce output image. makes more effective when accumulating true pixels. results experiments presented confirm effectiveness method.
Unmanned Aerial Vehicle (UAV) is defined as aircraft without the onboard presence of pilots. UAVs have been used to perform intelligence, surveillance, and reconnaissance missions. The are not limited military operations, they can also be in commercial applications such telecommunications, ground traffic control, search rescue crop monitoring, etc. In this paper, we propose a swarm intelligence-based method for UAVs' route optimization. team area coverage with set waypoints. problem...
Abstract Metaplasticity property of biological synapses is interpreted in this paper as the concept placing greater emphasis on training patterns that are less frequent. A novel implementation proposed which, during network learning phase, a priority given to weight updating frequent activations over more ones. Modeling interpretation hypothesis an improved tested Multilayer Perceptron type with Backpropagation training. The results obtained for chosen application show much efficient...
Robots that can assist in the activities of daily living, such as dressing, may support older adults, addressing needs an aging population face a growing shortage care professionals. Using depth cameras during robot-assisted dressing lead to occlusions and loss user tracking, which result unsafe trajectory planning or prevent task proceeding altogether. For putting on jacket, is addressed this letter, tracking arm lost when user's hand enters situations for poor interaction experience....
Swarms of robots can use their sensing abilities to explore unknown environments and deploy on sites interest. In this task, a large number is more effective than single unit because ability quickly cover the area. However, coordination teams not an easy problem, especially when resources for deployment are limited. paper, Distributed Bees Algorithm (DBA), previously proposed by authors, optimized applied distributed target allocation in swarms robots. Improved terms cost efficiency achieved...
Ant Colony Optimization (ACO) is a group of algorithms inspired by the foraging behavior ant colonies in nature. Like their biological counterparts, colony artificial ants able to adapt changes environment, such as exhaustion food source and discovery new one. In this paper, one basic ACO algorithms, System algorithm, was applied for edge detection where pixels represent ants. A set grayscale images obtained nonlinear contrast enhancement technique called Multiscale Adaptive Gain used create...
In this paper, a novel edge detection method that computes image gradient using the concept of Center Mass (COM) is presented. The algorithm runs with constant number operations per pixel independently from its scale by integral image. Compared conventional convolutional detector such as Sobel detector, proposed performs faster when region size larger than 9×9. can be used framework for multi-scale detectors goal to achieve fast performance. Experimental results show COM competent Canny detection.
In this paper, we propose a Modified Distributed Bees Algorithm (MDBA) for multi-sensor task allocation in supply chain security scenario. The MDBA assigns sensors to the upcoming tasks using decentralized, probabilistic approach maximize information gain while minimizing costs. Tasks are allocated based on sensors' performance, tasks' priorities and mutual sensor-task distances. Simulation analysis compared different algorithms indicated improved performance of 15% by with respect...
The goal of the Project group created by U.P.M. in collaboration with Foreign Universities, Research Institutions and Companies is development an intelligent mechatronic system for use precision sustainable agriculture. project as a whole includes following components: photographing decoding soil surface; fertility determination formation map; generation controlling signal dosing device; fertilizers; simulation, prototype testing; human-machine interaction training preparation.