Body-worn sensor data is used in monitoring patient activity during rehabilitation and also can be extended to controlling rehabilitation devices based on the activity of the person. The primary focus of research has been on effectively capturing the spatiotemporal dependencies in the data collected by these sensors and efficiently classifying human activities. With the increasing complexity and size of models, there is a growing emphasis on optimizing their efficiency in terms of memory usage and inference time for real-time usage and mobile computers. While hybrid models combining convolutional and recurrent neural networks have shown strong performance compared to traditional approaches, self-attention-based networks have demonstrated even superior results. However, instead of relying on the same transformer architecture, there is an opportunity to develop a novel framework that incorporates recent advancements to enhance speed and memory efficiency, specifically tailored for human activity recognition (HAR) tasks. In line with this approach, we present GLULA, a unique architecture for HAR. GLULA combines gated convolutional networks, branched convolutions, and linear self-attention to achieve efficient and powerful solutions. To enhance the performance of our proposed architecture, we employed manifold mixup as an augmentation variant which proved beneficial in limited data settings. Extensive experiments were conducted on five benchmark datasets: PAMAP2, SKODA, OPPORTUNITY, DAPHNET, and USC-HAD. Our findings demonstrate that GLULA outperforms recent models in the literature on the latter four datasets but also exhibits the lowest parameter count and close to the fastest inference time among state-of-the-art models.

Human gait data can be collected using inertial measurement units (IMUs). An IMU is an electronic device that uses an accelerometer and gyroscope to capture three-axial linear acceleration and three-axial angular velocity. The data so collected are time series in nature. The major challenge associated with these data is the segmentation of signal samples into stride-specific information, that is, individual gait cycles. One empirical approach for stride segmentation is based on timestamps. However, timestamping is a manual technique, and it requires a timing device and a fixed laboratory set-up which usually restricts its applicability outside of the laboratory. In this study, we have proposed an automatic technique for stride segmentation of accelerometry data for three different walking activities. The autocorrelation function (ACF) is utilized for the identification of stride boundaries. Identification and extraction of stride-specific data are done by devising a concept of tuning parameter ( $t_{p}$ ) which is based on minimum standard deviation ( $\sigma$ ). Rigorous experimentation is done on human activities and postural transition and Osaka University – Institute of Scientific and Industrial Research gait inertial sensor datasets. Obtained mean stride duration for level walking, walking upstairs, and walking downstairs is 1.1, 1.19, and 1.02 s with 95% confidence interval [1.08, 1.12], [1.15, 1.22], and [0.97, 1.07], respectively, which is on par with standard findings reported in the literature. Limitations of accelerometry and ACF are also discussed. stride segmentation; human activity recognition; accelerometry; gait parameter estimation; gait cycle; inertial measurement unit; autocorrelation function; wearable sensors; IoT; edge computing; tinyML.

Assistive strategies for occupational back-support exoskeletons have focused, mostly, on lifting tasks. However, in occupational scenarios, it is important to account not only for lifting but also for other activities. This can be done exploiting human activity recognition algorithms that can identify which task the user is performing and trigger the appropriate assistive strategy. We refer to this ability as exoskeleton versatility. To evaluate versatility, we propose to focus both on the ability of the device to reduce muscle activation (efficacy) and on its interaction with the user (dynamic fit). To this end, we performed an experimental study involving $ 10 $ healthy subjects replicating the working activities of a manufacturing plant. To compare versatile and non-versatile exoskeletons, our device, XoTrunk, was controlled with two different strategies. Correspondingly, we collected muscle activity, kinematic variables and users’ subjective feedbacks. Also, we evaluated the task recognition performance of the device. The results show that XoTrunk is capable of reducing muscle activation by up to $ 40\% $ in lifting and $ 30\% $ in carrying. However, the non-versatile control strategy hindered the users’ natural gait (e.g., $ -24\% $ reduction of hip flexion), which could potentially lower the exoskeleton acceptance. Detecting carrying activities and adapting the control strategy, resulted in a more natural gait (e.g., $ +9\% $ increase of hip flexion). The classifier analyzed in this work, showed promising performance (online accuracy > 91%). Finally, we conducted 9 hours of field testing, involving four users. Initial subjective feedbacks on the exoskeleton versatility, are presented at the end of this work.

Our aim is to develop a smartphone-based life-logging system. Human activity recognition (HAR) is one of the core techniques to realize it. Recent studies reported the effectiveness of feed-forward neural network (FF-NN) and recurrent neural network (RNN) as a classifier for HAR task. However, there are still unresolved problems in those studies: (1) a life-logging system using only a smartphone for recording device has not been developed, (2) only indoor activities have been utilized for evaluation, (3) insufficient investigations/evaluations of RNN. In this study, we address these unresolved problems as follows: (1) we build a prototype system for life-logging and conduct data recording experiment on this system to include both indoor and outdoor activities. The experimental results of HAR on this new dataset showed that RNN-based classifier was still effective. (2) From the results of a HAR experiment, it was demonstrated that a multi-layered Simple Recurrent Unit with a non-linear transform at the bottom layer and a highway-connection was the most effective. (3) We could grasp the reason for the improvement of RNN from FF-NN by observing the posterior probabilities over test data.