Non-invasive blood glucose monitoring using PPG signals with various deep learning models and implementation using TinyML.

Accurate and continuous blood glucose monitoring is essential for effective diabetes management, yet traditional finger pricking methods are often inconvenient and painful. To address this issue, photoplethysmography (PPG) presents a promising non-invasive alternative for estimating blood glucose levels. In this study, we propose an innovative 1-second signal segmentation method and evaluate the performance of three advanced deep learning models using a novel dataset to estimate blood glucose levels from PPG signals.

A Novel Ungrounded Haptic Device for Generation and Orientation of Force and Torque Feedbacks.

To provide deeper immersion for the user in the virtual environments, both force and torque feedbacks are required rather than the mere use of visual and auditory ones. In this paper, we develop a novel propeller-based Ungrounded Handheld Haptic Device (UHHD) that delivers both force and torque feedbacks in one device to help the user experience a realistic sensation of immersion in a three-dimensional (3D) space. The proposed UHHD uses only a pair of propellers and a set of sliders to continuously generate the desired force and torque feedbacks up to 15N and 1N.

Deep learning approaches for assessing pediatric sleep apnea severity through SpO2 signals.

Pediatric Sleep Apnea-Hypopnea (SAH) presents a significant health challenge, particularly in diagnostic contexts, where conventional Polysomnography (PSG) testing, although effective, can be distressing for children. Addressing this, our research proposes a less invasive method to assess pediatric SAH severity by analyzing blood oxygen saturation (SpO2) signals. We adopted two advanced deep learning architectures, namely ResNet-based and attention-augmented hybrid CNN-BiGRU models, to process SpO2 signals in a one-dimensional (1D) format for Apnea-Hypopnea Index (AHI) estimation in pediatric subjects.

Dynamics modeling and path following controller of tractor-trailer-wheeled robots considering wheels slip.

To improve the payload capacity and maneuverability of a Differentially-Driven Wheeled Robot (DDWR), a wheeled vehicle, which is called trailer, is connected to the DDWR. In all of the previous studies of DDWRs with a trailer, the robot wheels are subject to pure rolling constraints. However, when these multibody systems move with high velocities/accelerations, transfer a heavy payload, or travel on a slippery surface, they experience slipping and/or skidding.

Dataset for classifying and estimating the position, orientation, and dimensions of a list of primitive objects.

Objectives: Robotic systems are moving toward more interaction with the environment, which requires improving environmental perception methods. The concept of primitive objects simplified the perception of the environment and is frequently used in various fields of robotics, significantly in the grasping challenge. After reviewing the related resources and datasets, we could not find a suitable dataset for our purpose, so we decided to create a dataset to train deep neural networks to classify a primitive object and estimate its position, orientation, and dimensions described in this report.

Design and control of a lower limb rehabilitation robot considering undesirable torques of the patient's limb.

This research introduces a new exoskeleton-type rehabilitation robot, which can be used in lower limb rehabilitation therapy for post-stroke patients. A novel design of a typical knee and ankle rehabilitation robot is proposed. The kinematic and dynamic models of the knee and ankle rehabilitation robot are derived.