Effect of Segment Types on Characterization of Soft Sensing Textile Actuators for Soft Wearable Robots.

The use of textiles in soft robotics is gaining popularity because of the advantages textiles offer over other materials in terms of weight, conformability, and ease of manufacture. The purpose of this research is to examine the stitching process used to construct fabric-based pneumatic bending actuators as well as the effect of segment types on the actuators' properties when used in soft robotic glove applications. To impart bending motion to actuators, two techniques have been used: asymmetry between weave and weft knit fabric layers and mechanical anisotropy between these two textiles. The impacts of various segment types on the actuators' grip force and bending angle were investigated further. According to experiments, segmenting the actuator with a sewing technique increases the bending angle. It was discovered that actuators with high anisotropy differences in their fabric combinations have high gripping forces. Textile-based capacitive strain sensors are also added to selected segmented actuator types, which possess desirable properties such as increased grip force, increased bending angle, and reduced radial expansion. The sensors were used to demonstrate the controllability of a soft robotic glove using a closed-loop system. Finally, we demonstrated that actuators integrated into a soft wearable glove are capable of grasping a variety of items and performing various grasp types.

An Incremental Class-Learning Approach with Acoustic Novelty Detection for Acoustic Event Recognition.

Acoustic scene analysis (ASA) relies on the dynamic sensing and understanding of stationary and non-stationary sounds from various events, background noises and human actions with objects. However, the spatio-temporal nature of the sound signals may not be stationary, and novel events may exist that eventually deteriorate the performance of the analysis. In this study, a self-learning-based ASA for acoustic event recognition (AER) is presented to detect and incrementally learn novel acoustic events by tackling catastrophic forgetting. The proposed ASA framework comprises six elements: (1) raw acoustic signal pre-processing, (2) low-level and deep audio feature extraction, (3) acoustic novelty detection (AND), (4) acoustic signal augmentations, (5) incremental class-learning (ICL) (of the audio features of the novel events) and (6) AER. The self-learning on different types of audio features extracted from the acoustic signals of various events occurs without human supervision. For the extraction of deep audio representations, in addition to visual geometry group (VGG) and residual neural network (ResNet), time-delay neural network (TDNN) and TDNN based long short-term memory (TDNN-LSTM) networks are pre-trained using a large-scale audio dataset, Google AudioSet. The performances of ICL with AND using Mel-spectrograms, and deep features with TDNNs, VGG, and ResNet from the Mel-spectrograms are validated on benchmark audio datasets such as ESC-10, ESC-50, UrbanSound8K (US8K), and an audio dataset collected by the authors in a real domestic environment.

Robotic approach to the reduction of dental anxiety in children.

OBJECTIVE: We introduced a humanoid robot for the use of techno-psychological distraction techniques in children aged 4-10 to reduce their anxiety and improve their behaviour during dental treatment. MATERIALS AND METHODS: Two hundred children (98 boys, 102 girls; mean age: 6.5 ± 1.66 years) appointed for first time for dental caries were included and randomly divided into two groups [N = 100 for each group; RG: Robot Group (accompanied by the robot), CG: Control Group (without robot accompaniment)]. Half of the children were treated under local anaesthesia (infiltration anaesthesia) (n = 50 within each group) and half of the children were treated without any local anaesthesia (n = 50 within each group) within each group. The success rate of the new robotic distraction technique was evaluated by using Parental Corah Dental Anxiety Scale, Facial Image Scale (FIS), physiological pulse rate and Frankl Behaviour Rating Scale (FBRS). RESULT: Pulse rates, which measured during treatment and after treatment, were statistically higher in CG than in RG (p < .05). After dental treatment, the FIS score was significantly higher in CG than RG (p < .05). 88.3% of the children in RG stated that they wanted the robot to be with them at the next treatment session. CONCLUSIONS: Robotic technology can successfully help in coping with dental anxiety and stress, and helps children to behave better in dental office.

Emotional State Estimation using Sensor Fusion of EEG and EDA.

Emotions potentially have a significant impact on human actions and recognizing affective states is an effective way of implementing Brain-Computer Interface (BCI) systems which process brain signals to allow direct communication and interaction with the environment. In this paper, a real-time emotion recognition model was developed on the basis of physiological signals. A sensor fusion method is developed to detect human emotion by using data acquired from ElectroEncephaloGraphy (EEG) and ElectroDermal Activity (EDA) sensors. The proposed physiology-based emotion recognition system using a neural network was implemented and tested on human subjects, and a classification accuracy of 94% on three different emotions was achieved.