A lower-limb exoskeleton robot identifies the wearer′s walking intention and assists the walking movement through mechanical force; thus, it is important to be able to identify the wearer′s movement in real-time. Measurement of the angle of the knee and ankle can be difficult in the case of patients who cannot move the lower-limb joint properly. Therefore, in this study, the knee angle as well as the angles of the talocrural and subtalar joints of the ankle were estimated during walking by applying the neural network to two inertial measurement unit (IMU) sensors attached to the thigh and shank. First, for angle estimation, the gyroscope and accelerometer data of the IMU sensor were obtained while walking at a treadmill speed of 1 to 2.5 km/h while wearing an exoskeleton robot. The weights according to each walking speed were calculated using a neural network algorithm programmed in MATLAB software. Second, an appropriate weight was selected according to the walking speed through the IMU data, and the knee angle and the angles of the talocrural and subtalar joints of the ankle were estimated in real-time during walking through a feedforward neural network using the IMU data received in real-time. We confirmed that the angle estimation error was accurately estimated as 1.69° ± 1.43 (mean absolute error (MAE) ± standard deviation (SD)) for the knee joint, 1.29° ± 1.01 for the talocrural joint, and 0.82° ± 0.69 for the subtalar joint. Therefore, the proposed algorithm has potential for gait rehabilitation as it addresses the difficulty of estimating angles of lower extremity patients using torque and EMG sensors.

Lower limb exoskeleton robots help with walking movements through mechanical force, by identifying the wearer’s walking intention. When the exoskeleton robot is lightweight and comfortable to wear, the stability of walking increases, and energy can be used efficiently. However, because it is difficult to implement the complex anatomical movements of the human body, most are designed simply. Due to this, misalignment between the human and robot movement causes the wearer to feel uncomfortable, and the stability of walking is reduced. In this paper, we developed a two degrees of freedom (2DoF) ankle exoskeleton robot with a subtalar joint and a talocrural joint, applying a four-bar linkage to realize the anatomical movement of a simple 1DoF structure mainly used for ankles. However, bidirectional tendon-driven actuators (BTDAs) do not consider the difference in a length change of both cables due to dorsiflexion (DF) and plantar flexion (PF) during walking, causing misalignment. To solve this problem, a BTDA was developed by considering the length change of both cables. Cable-driven actuators and exoskeleton robot systems create uncertainty. Accordingly, adaptive control was performed with a proportional-integral-differential neural network (PIDNN) controller to minimize system uncertainty.

This study addresses the challenges that conventional network models face in detecting small foreign objects on industrial production lines, exemplified by scenarios where a single piece of iron filing occupies approximately 0.002% of the image area. To tackle this, we introduce an enhanced YOLOv8-MeY model for detecting foreign objects on the surface of sugar bags. Firstly, the introduction of a 160 × 160-scale small object detection layer and integration of the Global Attention Mechanism (GAM) attention module into the feature fusion network (Neck) increased the network’s focus on small objects. This enhancement improved the network’s feature extraction and fusion capabilities, which ultimately increased the accuracy of small object detection. Secondly, the model employs the lightweight network GhostNet, replacing YOLOv8’s principal feature extraction network, DarkNet53. This adaptation not only diminishes the quantity of network parameters but also augments feature extraction capabilities. Furthermore, we substituted the Bottleneck in the C2f of the YOLOv8 model with the Spatial and Channel Reconstruction Convolution (SCConv) module, which, by mitigating the spatial and channel redundancy inherent in standard convolutions, reduced computational demands while elevating the performance of the convolutional network model. The model has been effectively applied to the automated sugar dispensing process in food factories, exhibiting exemplary performance. In detecting diverse foreign objects like 2 mm iron filings, 7 mm wires, staples, and cockroaches, the YOLOv8-MeY model surpasses the Faster R-CNN model and the contemporaneous YoloV8n model of equivalent parameter scale across six metrics: precision, recall, mAP@0.5, parameters, GFLOPs, and model size. Through 400 manual placement tests involving four types of foreign objects, our statistical results reveal that the model achieves a recognition rate of up to 92.25%. Ultimately, we have successfully deployed this model in automated sugar bag dispensing scenarios.

In this study, an ankle intention detection algorithm was developed to calculate the torque the user wants to exert from the ankle starting from the user’s EMG signal. Since the subtalar joint axis of the ankle is very important for stability, the intent detection algorithm also calculates the torque of the eversion motion of the subtalar joint axis. A dry EMG sensor was used to measure the EMG signal, and an ankle biaxial torque measurement device was manufactured to measure the ankle torque to perform the experiment. The experiment was conducted on four healthy subjects (mean ± SD: height, 177.6 ± 7.3 cm; weight 70.2 ± 8.9 kg; and age, 27 ± 2 years), and the EMG signals and ankle torque were measured. Using the experimental results and a neural network, we developed an intention detection algorithm. When you input an EMG signal, the algorithm estimates the torque of eversion, dorsiflexion, and plantar flexion. The error of the algorithm is 0.37 Nm (subtalar) and 0.57 Nm (talocrural), which is 0.5% (subtalar) and 1.5% (talocrural) of the torque required for walking. Using the algorithm of this study, more accurate and stable exoskeleton robot control becomes possible.