The lower limb exoskeleton technology is designed to facilitate the movement of human lower limbs. Significant progress has been made in this technology, which has important implications for rehabilitation patients and individuals who are eager to enhance their mobility. Electromyogram (EMG) signals, which encompass the complexity of human physiology, are integrated into lower limb exoskeletons due to their deep connection to movement and predictability before movement begins, and this integration is expected to enable intelligent control and improved human-computer interaction. This review explores a pattern based on EMG signals for identifying human motor intent in lower limb exoskeletons. Firstly, the development of lower limb exoskeleton and the existing lower limb exoskeleton products are systematically described. Combined with the intelligent control system of wearable device, the main methods and research progress of recognizing the motion intention of lower limb exoskeleton by surface EMG are discussed. It shows that the use of surface EMG can effectively improve the human-machine interaction of lower limb exoskeleton. Together, the study provides insight into the challenges that are hindering the commercialization of the market and provides a perspective on the future development of EMG signals.

With an aging population and the high incidence of neurological diseases, rehabilitative lower limb exoskeleton robots, as a wearable assistance device, present important application prospects in gait training and human function recovery. As the core of human–computer interaction, control strategy directly determines the exoskeleton’s ability to perceive and respond to human movement intentions. This paper focuses on the control strategies of rehabilitative lower limb exoskeleton robots. Based on the typical hierarchical control architecture of “perception–decision–execution,” it systematically reviews recent research progress centered around four typical control tasks: trajectory reproduction, motion following, Assist-As-Needed (AAN), and motion intention prediction. It emphasizes analyzing the core mechanisms, applicable scenarios, and technical characteristics of different control strategies. Furthermore, from the perspectives of drive system and control coupling, multi-source perception, and the universality and individual adaptability of control algorithms, it summarizes the key challenges and common technical constraints currently faced by control strategies. This article innovatively separates the end-effector control strategy from the hardware implementation to provide support for a universal control framework for exoskeletons.

With the extension of road service life, cracks are the most significant type of pavement distress. To monitor road conditions and avoid excessive damage, pavement crack detection is absolutely necessary and an indispensable part of road periodic maintenance and performance assessment. The development and application of computer vision have provided modern methods for crack detection, which are low in cost, less labor-intensive, continuous, and timely. In this paper, an intelligent model based on a target detection algorithm in computer vision was proposed to accurately detect and classify four classes of cracks. Firstly, by vehicle-mounted camera capture, a dataset of pavement cracks with complicated backgrounds that are the most similar to actual scenarios was built, containing 4007 images and 7882 crack samples. Secondly, the YOLOv5 framework was improved from the four aspects of the detection layer, anchor box, neck structure, and cross-layer connection, and thereby the network’s feature extraction capability and small-sized-target detection performance were enhanced. Finally, the experimental results indicated that the proposed model attained an AP of the four classes of 81.75%, 83.81%, 98.20%, and 92.83%, respectively, and a mAP of 89.15%. In addition, the proposed model achieved a 2.20% missed detection rate, representing a 6.75% decrease over the original YOLOv5. These results demonstrated the effectiveness and practicality of our proposed model in addressing the issues of low accuracy and missed detection for small targets in the original network. Overall, the implementation of computer vision-based models in crack detection can promote the intellectualization of road maintenance.

The characteristics of TiO₂ nanotube arrays (TNTs) prepared on Ti foil in sulfuric acid solution that contains Cl⁻ under different temperatures are investigated by field emission scanning electron microscopy (FESEM), electrochemical impedance spectroscopy (EIS), Mott–Schottky measurement and Raman spectra. The solution temperature significantly affects the morphologies of TNTs, i.e., when solution temperature rises from −10 °C to 90 °C, the inner diameter of the nanotube increases and the barrier layer thickness decreases, and, as TNTs display n-type semiconductive properties, the donor density (N_D) and corrosion protection decrease. Two types (types I and II) of pulse temperature are used to fabricate TNTs, in which type I is firstly anodized at a low temperature for time t, and then increases to a high temperature. While for type II, the solution temperature order is opposite to that of type I. The N_D of TNTs in the case of type I is lower than N_D of TNTs in the case of type II. N_D decreases with the increased pulse step time for type I, while N_D increases with the increased pulse step time for type II.

The inherent heterogeneity of coal significantly influences cutting efficiency, directly impacting energy consumption and dust generation in mining operations. To investigate this effect, this study established a heterogeneous coal model using PFC 2D 5.0, assigning strength parameters based on the Weibull distribution. The influence of the heterogeneity index (λ) on macroscopic strength, brittleness, and micro-crack propagation during coal cutting was systematically analyzed, and comparisons were made with homogeneous models of varying uniaxial compressive strength (UCS). The results show that as λ increases, both UCS and Brazilian tensile strength (BTS) increase exponentially, approaching the values of the homogeneous models, with BTS exhibiting greater sensitivity to λ than UCS. The brittleness index also increases with λ. During cutting, a higher λ leads to more concentrated crack propagation and stress distribution, as well as a reduced proportion of shear cracks, indicating a shift toward a more controllable fragmentation mode. Correspondingly, the specific energy (SE) for cutting decreases monotonically with λ, reflecting enhanced cutting efficiency, a trend attributed to the reduced energy dissipation from shear friction and the homogenization of internal stress distribution. Compared with homogeneous models, the heterogeneous models produce a more complex crack network and a greater number of cracks at lower strength levels, though these differences diminish as λ increases. These findings provide theoretical insights for optimizing cutting parameters to reduce energy consumption and minimize uncontrolled fracturing in efficient coal resource exploitation.