Search

The main objective in this proposal is to orchestrate an ecosystem of manipulation of reliable and safe data, applied to the field of health, specifically lung cancer, by introducing the creation of digital twins for personalised healthcare about the behaviour of this disease on patients. Digital twins is a very popular and novel approach in digitisation units in industry which will be used by both kind of experts: (i) data analysts, who will design expert recommender systems and extract knowledge – explainable Artificial Intelligence (AI); and (ii) professionals in medicine, who will consume that knowledge generated with their research for better diagnosis. This knowledge generation/extraction process will work in the form of a lifelong learning system by iterative and continuous use. The produced software platform will be abstracted so it can be applied like a general purpose service tool in other domains of knowledge, specially health and industry. Furthermore, a rule extraction module will be made available for explainability issues.

The increment of dependant people for the realization of Activities of Daily Living is a fact that comes implicit with the increase of population and their lifespan given by the modern technological progress. For this reason, Assistive Robotics is a field that is currently researched and gradually implemented. In this project, two different algorithms to train robots to perform those tasks are being studied and compared, Reinforcement Learning and Imitation Learning. The main objective is to learn the benefits, drawbacks and suitability of both for a given task. To do the experimentation, the Assistive Gym software will be used, to train the nets with Reinforcement Learning and to perform the movements with Dynamic Movement Primitives learned from a demonstration. The environment will consist in a Sawyer robot cleaning the right arm of a human lying down on a bed. The potential of each methodology will be tested in different scenarios, having the user’s arm straight or partially flexed. Furthermore, small implementations to improve the execution of the task will be implemented for the Dynamic Movement Primitives to test the performance and the similarity to the demonstration. Both methodologies are capable to perform the desired task successfully, the model trained with Reinforcement Learning is more suitable for an individual user that will use the robot in a long term, due to the training time. The Imitation Learning algorithm is faster to teach and its trajectory and velocities are easily adaptable to the preferences of the user, it is more suitable in places where there is a high rotation of users that require urgent and short attention, such as clinics or hospitals

Support Vector Machine (SVM) learning from imbalanced datasets, as well as most learning machines, can show poor performance on the minority class because SVMs were designed to induce a model based on the overall error. To improve their performance in these kind of problems, a low-cost post-processing strategy is proposed based on calculating a new bias to adjust the function learned by the SVM. The proposed bias will consider the proportional size between classes in order to improve performance on the minority class. This solution avoids not only introducing and tuning new parameters, but also modifying the standard optimization problem for SVM training. Experimental results on 34 datasets, with different degrees of imbalance, show that the proposed method actually improves the classification on imbalanced datasets, by using standardized error measures based on sensitivity and g-means. Furthermore, its performance is comparable to well-known cost-sensitive and Synthetic Minority Over-sampling Technique (SMOTE) schemes, without adding complexity or computational costs.

The problem of finding a suitable qualitative representation for robots to reason about activity spaces where they carry out tasks such as leading or interacting with a group of people is tackled in this paper. For that, a Qualitative Spatial model for Group Robot Interaction (QS-GRI) is proposed to define Kendon’s F-formations [Kendon, 2010] depending on: (i) the relative location of the robot with respect to other individuals involved in that interaction; (ii) the individuals’ orientation; (iii) the shared peri-personal distance; and (iv) the role of the individuals (observer, main character or interactive). An iconic representation is provided and Kendon’s formations are defined logically. The conceptual neighborhood of the evolution of Kendon formations is studied, that is, how one formation is transformed into another. These transformations can depend on the role that the robot have, and on the amount of people involved.

The sensorimotor approach to cognition states, that the key to bring semantics to the world of a robot, requires making the robot learn the relation between the actions that the robot performs and the change it experiences in its sensed data because of those actions. Those relations are called sensorimotor contingencies (SMCs). This special issue presents a variety of recent developments in SMCs with a particular focus on cognitive robotics applications.

Natural and intuitive human interaction with robotic systems is a key point to develop robots assisting people in an easy and effective way. In this paper, a Human Robot Interaction (HRI) system able to recognize gestures usually employed in human non-verbal communication is introduced, and an in-depth study of its usability is performed. The system deals with dynamic gestures such as waving or nodding which are recognized using a Dynamic Time Warping approach based on gesture specific features computed from depth maps. A static gesture consisting in pointing at an object is also recognized. The pointed location is then estimated in order to detect candidate objects the user may refer to. When the pointed object is unclear for the robot, a disambiguation procedure by means of either a verbal or gestural dialogue is performed. This skill would lead to the robot picking an object in behalf of the user, which could present difficulties to do it by itself. The overall system — which is composed by a NAO and Wifibot robots, a KinectTM v2 sensor and two laptops — is firstly evaluated in a structured lab setup. Then, a broad set of user tests has been completed, which allows to assess correct performance in terms of recognition rates, easiness of use and response times.

We introduce robots as a tools to enhance Ambient Assisted Living (AAL) services. Robots are a unique opportunity to create new systems to cooperate in reaching better living conditions. Robots offer the possibility of richer interaction with humans, and can perform actions to actively change the environment. The current state-of-art includes skills in various areas, including advanced interaction (natural language, visual attention, object recognition, intention learning), navigation (map learning, obstacle avoidance), manipulation (grasping, use of tools), and cognitive architectures to handle highly unpredictable environments. From our experience in several robotics projects and principally in the RoboCup@Home competition, a new set of evaluation methods is proposed to assess the maturity of the required skills. Such comparison should ideally enable the abstraction from the particular robotic platform and concentrate on the easy comparison of skills. The validity of that low-level skills can be then scaled to more complex tasks, that are composed by several skills. Our conclusion is that effective evaluation methods can be designed with the objective of enabling robots to enlarge AAL services.