Search

Humanoid robots progress everyday closer and closer to a more stable walking suitable for a human environment as the researchers in the Humanoid robotics field focus their effort on the understanding of the human locomotion. Nonetheless for Social Robotics researchers, humanoid robots might have another use, such as being our companions from birth to nursing home. Designing social humanoid robots is one critical step if we want the robots to be active in our society. However, to our knowledge, only a few studies in the area of humanoid robotics have addressed emotion expression with robot gaits. In this paper we propose to assess different emotional gait patterns and the perception of the emotion intensity in those patterns. Actors' emotional movement were captured and then normalized for our robot platform. Several robot simulations were shown to human observers who completed a survey questionnaire in which they indicated their assessment of the portrayed emotion by the robot simulation. The surveyed emotions consist of Sadness, Happiness, Anger, Fear with different intensities (Intermediate, High and Exaggerated). We achieved a high recognition rate of emotions (72.32%). Even if the intensities were less well recognized (33.63%), our study indicates that the intensity might help the recognition of emotional walking.

A common problem among elderly people is the loss of motor ability. Rehabilitation exercises can help these people recover strength and maintain a good level of mobility. However, high costs and the need for special equipment make professional rehabilitation impractical for regular use in daily life, precluding elderly the possibility to perform focalized training at home. The idea of telerehabilitation is becoming more and more concrete with the rapid development of internet technology. Telerehabilitation would allow the user to perform exercises at home with online professional direction from the doctor. However, at the present state, the doctor cannot obtain real-time and quantitative data from the user, and this limits the training effectiveness. To overcome this problem, an extremely miniaturized, portable motion capture system, named WB-4R, has been developed. Calibration and real-time link orientation reconstruction are very important to improve the accuracy in real-time measurement. In this paper, using the positive results of preliminary experiments on lower limbs, the authors will show the feasibility of the method and confirm the effectiveness of the developed system.

In this paper, we describe a human gesture recognition system developed to make a humanoid robot understand non-verbal human social behaviors, and we present the results of preliminary experiments to demonstrate the feasibility of the proposed method. In particular, we have focused on the detection and recognition of laughter, a very peculiar human social signal. In fact, although it is a direct form of social interaction, laughter is classified as semi voluntary action, can be elicited by several different stimuli, and it is strongly associated with positive emotion and physical well-being. The possibility of recognize, and further elicit laughter, will help the humanoid robot to interact in a more natural way with humans, to build positive relationships and thus be more socially integrated in the human society.

Endotracheal Intubation (ETI) is a common airway procedure used to connect the larynx and the lungs through the windpipe of patients under emergency situations. The process is carried out by a laryngoscope inserted into the mouth, and helps doctors in viewing the glottis and inserting the tube. In 2012, we developed a humanoid robot for ETI training, called WKA-5R, which was able to simulate various patients and scenarios and to evaluate the doctor's medical skills according to applied force, position of laryngoscopes and tubes, and procedure time. In 2013, we have focused our study on the development of a system for the biomechanical evaluation during simulated ETI. We conducted an experiment with 5 expert and 6 novice anesthesiologists and measured 3 biomechanical parameters: angular velocity of wrist joint and elbow joint, surface electromyography (sEMG) of the upper limb, and some angles on the upper part of the body, in order to evaluate joint stiffness in the arm and posture during ETI. In this paper, the authors show the experimental results on comparison between experts and novices, followed by discussion.