ALBERT

Description: Thermal conductivity of porous thermal barrier coatings was evaluated using a newly developed five-phase model. It was demonstrated that porosities distributed in coating strongly affect thermal conductivity. The decisive reason for this change in thermal conductivity can be traced back to defect morphology and its orientation, depending on the coating deposition technique and process parameters used during deposition. In this paper, the Bruggeman’s two-phase model was used as a reference, and a five-phase model was developed to evaluate the thermal conductivity of porous coatings. This approach uses microstructural details of the shape, size, orientation and volumetric fraction of defects of coatings as input parameters. The proposed model can predict thermal conductivity values better than the previous two-phase model.

Description: In social situations, people who use a powered wheelchair must divide their attention between navigating the chair and conversing with people. These conversations could lead to increased mental stress when navigating and distraction from maneuvering the chair. As a solution that maintains a good conversation distance between the wheelchair and the accompanying person (Social Following), a wheelchair control system was developed to provide automated side-by-side following by wirelessly connecting the wheelchair to the person. Two ultrasonic range sensors and three piezoelectric ultrasonic transducers were used to identify the accompanying person and determine their position and heading. Identification involved an ultrasonic beacon worn on the person’s side, at hip level, and receivers on the wheelchair. A drive control algorithm maintained a constant conversation distance along the person’s trajectory. A plug-and-play prototype was developed and connected to a Permobil F5 Corpus wheelchair with a modified Eightfold Technologies SmartChair Remote. Results demonstrated that the system can navigate a wheelchair based on the accompanying person’s trajectory, which is advantageous for users who require hands-free wheelchair control during social activities.

Description: Recognizing aggressive movements is a challenging task in human activity recognition. Wearable smartwatch technology with machine learning may be a viable approach for human aggressive behavior classification. This research identified a viable classification model and feature selector (CM-FS) combination for separating aggressive from non-aggressive movements using smartwatch data and determined if only one smartwatch is sufficient for this task. A ranking method was used to select relevant CM-FS models across accuracy, sensitivity, specificity, precision, F-score, and Matthews correlation coefficient (MCC). The Waikato environment for knowledge analysis (WEKA) was used to run 6 machine learning classifiers (random forest, k-nearest neighbors (kNN), multilayer perceptron neural network (MP), support vector machine, naïve Bayes, decision tree) coupled with three feature selectors (ReliefF, InfoGain, Correlation). Microsoft Band 2 accelerometer and gyroscope data were collected during an activity circuit that included aggressive (punching, shoving, slapping, shaking) and non-aggressive (clapping hands, waving, handshaking, opening/closing a door, typing on a keyboard) tasks. A combination of kNN and ReliefF was the best CM-FS model for separating aggressive actions from non-aggressive actions, with 99.6% accuracy, 98.4% sensitivity, 99.8% specificity, 98.9% precision, 0.987 F-score, and 0.984 MCC. kNN and random forest classifiers, combined with any of the feature selectors, generated the top models. Models with naïve Bayes or support vector machines had poor performance for sensitivity, F-score, and MCC. Wearing the smartwatch on the dominant wrist produced the best single-watch results. The kNN and ReliefF combination demonstrated that this smartwatch-based approach is a viable solution for identifying aggressive behavior. This wrist-based wearable sensor approach could be used by care providers in settings where people suffer from dementia or mental health disorders, where random aggressive behaviors often occur.

Description: Residual stresses play an essential role in determining the failure mechanisms and life of an electron beam-physical vapour deposition thermal barrier coating (EB-PVD TBC) system. In this paper, a new transitional roughness model was proposed and applied to describe the interfacial roughness profile during thermal cycles. Finite element models were implemented to calculate residual stresses at specific positions close to the interface of TBCs using temperature process-dependent model parameters. Combining stresses evaluated at valleys of the topcoat (TC) and excessive sharp tip roughness profiles, positions where the maximum out-of-plane residual stresses occur were identified and used to explain possible cracking routes of EB-PVD TBCs as interfacial roughness evolves during thermal cycling.

Description: Foot strike detection is important when evaluating a person’s gait characteristics. Accelerometer and gyroscope signals from smartphones have been used to train artificial intelligence (AI) models for automated foot strike detection in able-bodied and elderly populations. However, there is limited research on foot strike detection in lower limb amputees, who have a more variable and asymmetric gait. A novel method for automated foot strike detection in lower limb amputees was developed using raw accelerometer and gyroscope signals collected from a smartphone positioned at the posterior pelvis. Raw signals were used to train a decision tree model and long short-term memory (LSTM) model for automated foot strike detection. These models were developed using retrospective data (n = 72) collected with the TOHRC Walk Test app during a 6-min walk test (6MWT). An Android smartphone was placed on a posterior belt for each participant during the 6MWT to collect accelerometer and gyroscope signals at 50 Hz. The best model for foot strike identification was the LSTM with 100 hidden nodes in the LSTM layer, 50 hidden nodes in the dense layer, and a batch size of 64 (99.0% accuracy, 86.4% sensitivity, 99.4% specificity, and 83.7% precision). This research created a novel method for automated foot strike identification in lower extremity amputee populations that is equivalent to manual labelling and accessible for clinical use. Automated foot strike detection is required for stride analysis and to enable other AI applications, such as fall detection.