Search results
9 records were found.
Contém referências bibliográficas
Contém referências bibliográficas
The paper presents and discusses a framework to promote older adults cognitive and emotional stimulation via Robotic Cognitive Games. The work is based on classic games for older adults, e.g., to place objects in pre-defined positions in an arena, where the authors introduce a robot in the games. The paper not only presents the robotic games, but also the methodology developed to properly introduce them to older adults in a nursing home. As such, the paper proposes three cognitive robotic games, a methodology to assess the success of its introduction to older adults, keeping in mind cognitive and emotional aspects. To validate the proposed robotic solution, experimental tests were performed in a nursing home. A prior cognitive and emotional test was done with older adults to have a ground truth to compare with after a batch of games was completed by each older adult. The results and their discussion validate the robotic games approach, and also the methodology used for its introduction in the nursing home.
The purpose of this work is to develop computational intelligence models based on neural networks (NN), fuzzy models (FM), support vector machines (SVM) and long short-term memory networks (LSTM) to predict human pose and activity from image sequences, based on computer vision approaches to gather the required features. To obtain the human pose semantics (output classes), based on a set of 3D points that describe the human body model (the input variables of the predictive model), prediction models were obtained from the acquired data, for example, video images. In the same way, to predict the semantics of the atomic activities that compose an activity, based again in the human body model extracted at each video frame, prediction models were learned using LSTM networks. In both cases the best learned models were implemented in an application to test the systems. The SVM model obtained 95.97% of correct classification of the six different human poses tackled in this work, during tests in different situations from the training phase. The implemented LSTM learned model achieved an overall accuracy of 88%, during tests in different situations from the training phase. These results demonstrate the validity of both approaches to predict human pose and activity from image sequences. Moreover, the system is capable of obtaining the atomic activities and quantifying the time interval in which each activity takes place.
This paper presents and discusses ethical issues related with the interaction of elderly people with robots while playing cognitive games. Four games were developed from classical nursing homes cognitive games, to incrementally increase its difficulty. The last one does include interaction with a robot. Several ethical issues were tackled to obtain a proper system with robots: those ethical issues were raised during the design phase and the interaction with the elders. As example of ethical issues tackled: beneficience - do not harm, and the deception and infantilisation of elderly. The games are based in objects with different shapes and colours to be positioned by the elder in pre-defined positions of a 3x3 arena, on the floor. The system then interacts with the elder given him/her the respective score, after performing image processing techniques to the image of the arena. The game with the robot consists to knock down objects in the arena, by pushing them outside the area using a car-like tele-operated robot, by the elders. The system was validated in two nursing homes, by performing experiments in the last months, and it is ethically proper.
