WiFi fingerprint-based positioning is a method for indoor localization with the advent of widespread deployment of WiFi and the Internet of Things. However, single WiFi fingerprint positioning has the problems of mismatch, unstable signal strength and limited accuracy. Aiming to address these issues, this paper proposes the fusion algorithm combining WiFi and pedestrian dead reckoning (PDR). Firstly, the particle swarm optimization (PSO) model is utilized to optimize the weighted k-nearest neighbors (WKNN) in the WiFi part. Additionally, the artemisinin optimization (AO) algorithm is used to optimize the particle filter (PF) to improve the fusion effect of the WiFi and PDR. Finally, to thoroughly validate the localization performance of the proposed algorithm, we designed experiments involving two scenarios with four smartphone gestures: calling, dangling, handheld, and pocketed. The experimental results unequivocally indicate that the positioning error of AO-PSO-PF algorithm is lower than that of other algorithms including PDR, WiFi, PF, APF, and FPF. The average positioning errors for the two experiments are 0.95 m and 1.42 m, respectively.

Lithium batteries are pivotal in new energy vehicle (NEV) applications. This study proposes a novel defect detection model, Small-object Detection-YOLO (SD-YOLO), based on YOLOv11, to overcome the limitations of conventional methods in detecting small surface defects in lithium batteries, particularly addressing the challenge of high miss rates for low-contrast targets. The Similarity-Aware Activation Module (SimAM) is integrated into the backbone, enabling the model to prioritize small-scale defects and enhance feature extraction. The C3k2 GhostDynamicConv module is introduced to replace conventional convolution layers, enhancing feature representation while reducing computational complexity through dynamic convolutions and Ghost feature integration. To further refine model performance, the Focaler-Iou loss function is employed, replacing the original CIoU loss function, optimizing bounding box regression, and accelerating model convergence. Experimental results show that the proposed model significantly outperforms the original YOLOv11 in key performance metrics, with a 1.3% increase in precision, a 0.5% increase in recall, and a 2.2% increase in mean average precision (mAP). Furthermore, the model outperforms other state-of-the-art methods in both detection speed (FPS) and small-target detection capabilities.

Lumbar disc herniation is one of the primary causes of lower back pain, and its incidence has significantly increased with the development of industrialization. To assist in rehabilitation therapy, this paper proposes a flexible exoskeleton for active lumbar rehabilitation based on a 4-SPU/SP biomimetic parallel mechanism. By analyzing the anatomical structure and movement mechanisms of the lumbar spine, a four degree of freedom parallel mechanism was designed to mimic the three-axis rotation of the lumbar spine around the coronal, sagittal, and vertical axes, as well as movement along the z-axis. Using a 3D motion capture system, data on the range of motion of the lumbar spine was obtained to guide the structural design of the exoskeleton. Using the vector chain method, the display equations for the drive joints of the mechanism were derived, and forward and inverse kinematic models were established and simulated to verify their accuracy. The dynamic characteristics of the biomimetic parallel mechanism were analyzed and simulated to provide a theoretical basis for the design of the exoskeleton control system. A prototype was fabricated and tested to evaluate its maximum range of motion and workspace. Experimental results showed that after wearing the exoskeleton, the lumbar spine’s range of motion could still reach over 83.5% of the state without the exoskeleton, and its workspace could meet the lumbar spine movement requirements for daily life, verifying the rationality and feasibility of the proposed 4-SPU/SP biomimetic parallel mechanism design.