Gestures are an important way to conduct human-computer interaction. The key problem of gesture recognition depending on sEMG (surface electromyography) is how to achieve high recognition accuracy when there are many types of gestures to classify. To solve this problem, first, two basic models were constructed. One is the ConvEMG model based on dense connectivity, the Inception module and depthwise separable convolution; and the other is the LSTMEMG model based on a bidirectional LSTM (Long Short-Term Memory). Then, the basic models were improved with a multistream fusion strategy which utilizes the correlation between gestures and muscles and the complementary advantages of models. To facilitate comparison with others’ models, the models proposed in this paper were tested on the public dataset NinaPro DB5, and the improved model named MultiConvEMG achieves an accuracy of 92.83% for 41 gestures, which is superior to its counterparts in the literature on the same dataset. In addition, experiments containing signal acquisition and gesture recognition were carried out for further testing and evaluation. Experimental results show that all models can achieve an accuracy of more than 95% for 31 gestures, and these models have their own strengths in accuracy, immediacy or training cost. All models built in the paper support using sEMG for end-to-end recognition, which means that artificial features are not needed in the processes and data augmentation or IMU devices are not relied on. In other words, our models outperform and have lower application costs than many known models.

The finite element model (FE) of temperature field of straight thin-walled samples in laser cladding IN718 was established, and the growth of microstructure was simulated by cellular automata (CA) method through macro-micro coupling (CA-FE). The effects of different cooling conditions on microstructure, hardness, and properties of laser-cladding layer were studied by designing cooling device. The results show that the simulation results are in good agreement with the microstructure of the cladding layer observed by the experiment. With the scanning strategy of reducing laser power layer-by-layer, the addition of water cooling device and the processing condition of 0.7 mm Z-axis lift, excellent thin-walled parts can be obtained. With the increase of cladding layers, the pool volume increases, the temperature value increases, the temperature gradient, cooling rate, solidification rate, K value gradually decrease, and eventually tend to be stable, in addition, the hardness shows a fluctuating downward trend. Under the processing conditions of layer-by-layer power reduction and water cooling device, the primary dendrite arm spacing reduced to about 8.3 μm, and the average hardness at the bottom of cladding layer increased from 260 HV to 288 HV. The yield strength and tensile strength of the tensile parts prepared under forced water cooling increased to a certain extent, while the elongation slightly decreased.

Among the numerous indoor localization methods, Light-Detection-and-Ranging (LiDAR)-based probabilistic algorithms have been extensively applied to indoor localization due to their real-time performance and high accuracy. Nevertheless, these methods are challenged in symmetrical environments when tackling global localization and the robot kidnapping problem. In this paper, a novel hybrid method that combines visual and probabilistic localization results is proposed. Augmented Monte Carlo Localization (AMCL) is improved for position tracking continually. LiDAR-based measurements’ uncertainty is evaluated to incorporate discrete visual-based results; therefore, a better diversity of the particle can be maintained. The robot kidnapping problem can be detected and solved by preventing premature convergence of the particle filter. Extensive experiments were implemented to validate the robustness and accuracy performance. Meanwhile, the localization error was reduced from 30 mm to 9 mm during a 600 m tour.

In the current study, our research group proposed an asymmetric lower extremity exoskeleton to enable above-knee amputees to walk with a load. Due to the absence of shank and foot, the knee and ankle joint at the amputation side of the exoskeleton lack tracking targets, so it is difficult to realize the function of assisted walking when going up and downstairs. Currently, the use of lower-limb electromyography to predict the angles of lower limb joints has achieved remarkable results. However, the prediction effect was poor when only using electromyography from the thigh. Therefore, this paper introduces hip-angle and plantar pressure signals for improving prediction effect and puts forward a joint prediction method of knee- and ankle-joint angles by electromyography of the thigh, hip-joint angle, and plantar pressure signals. The generalized regression neural network optimized by the golden section method is used to predict the joint angles. Finally, the parameters (the maximum error, the Root-Mean-Square error (RMSE), and correlation coefficient ($\gamma$)) were calculated to verify the feasibility of the prediction method.

According to the structural characteristics of thin-walled parts, a model slicing method is proposed, and its mathematical process is established. The three-dimensional transient temperature field in the process of synchronous powder feeding laser cladding is studied and verified by numerical simulation method, and the thin-walled parts formed by later experimental processing are processed by the results of numerical simulation. Using the simulation results of temperature field as the basis for optimizing the processing parameters, the forming path of thin-walled parts is programmed and optimized, and the experimental verification shows the reliability of this method.

Three-dimensional printing has great advantages in manufacturing parts with complex internal structures. At present, there are still significant challenges in the unsupported 3D printing of thick-walled parts with inner runners. In this paper, a partitioning method for support-free-fabricating workparts with a built-in inner runner structure is proposed and discussed in detail. With the method, partitioning planes are firstly created according to the direction changes of the inner runner, a “top to bottom” method is used for avoiding the interference of the generated planes, and thirdly the outer surface of the workpiece is considered for a second partition for the support-free purpose. A key algorithm for calculating partitioning planes for inner runner structures is also proposed and introduced in detail, including the iteration method, the calculation for intersectional profile mass centers, and the discussion of the convergence. Algorithm analysis is also performed with a simple model for assessing the influence from the defined parameters, including the proximity Φ_s and the increment coefficient σ, on the iteration results as well as on the iteration process. Also, an application test is carried out on a column model with a complex inner runner structure built-in. The result from all the tests indicates that the proposed algorithm is successful in partitioning inner runner structures for support-free fabrication.