Abstract Dynamic gesture recognition has become a new type of interaction to meet the needs of daily interaction. It is the most natural, easy to operate, and intuitive, so it has a wide range of applications. The accuracy of gesture recognition depends on the ability to accurately learn the short-term and long-term spatiotemporal features of gestures. Our work is different from improving the performance of a single type of network with convnets-based models and recurrent neural network-based models or serial stacking of two heterogeneous networks, we proposed a fusion architecture that can simultaneously learn short-term and long-term spatiotemporal features of gestures, which combined convnets-based models and recurrent neural network-based models in parallel. At each stage of feature learning, the short-term and long-term spatiotemporal features of gestures are captured simultaneously, and the contribution of two heterogeneous networks to the classification results in spatial and channel axes that can be learned automatically by using the attention mechanism. The sequence and pooling operation of the channel attention module and spatial attention module are compared through experiments. And the proportion of short-term and long-term features of gestures on channel and spatial axes in each stage of feature learning is quantitatively analyzed, and the final model is determined according to the experimental results. The module can be used for end-to-end learning and the proposed method was validated on the EgoGesture, SKIG, and IsoGD datasets and got very competitive performance.
Abstract The fitness function value is a kind of important information in the search process, which can be more targeted according to the guidance of the fitness function value. Most existing meta-heuristic algorithms only use the fitness function value as an indicator to compare the current variables as good or bad but do not use the fitness function value in the search process. To address this problem, the mathematical idea of the fitting is introduced into the meta-heuristic algorithm, and a symmetric projection optimizer (SPO) is proposed to solve numerical optimization and engineering problems more efficiently. The SPO algorithm mainly utilizes a new search mechanism, the symmetric projection search (SP) method. The SP method quickly completes the fitting of the projection plane, which is located through the symmetry of the two points and finds the minima in the projection plane according to the fitting result. Fitting by using the fitness function values allows the SP to find regions where extreme values may exist more quickly. Based on the SP method, exploration and exploitation strategies are constructed, respectively. The exploration strategy is used to find better regions, and the exploitation strategy is used to optimize the discovered regions continuously. The timing of the use of the two strategies is designed so that the SPO algorithm can converge faster while avoiding falling into local optima. The effectiveness of the SPO algorithm is extensively evaluated using seven test suites, including CEC2017, CEC2019, CEC2020, and CEC2022. It is also compared with two sets of 19 recent competitive algorithms. Statistical analyses are performed using five metrics such as the Wilcoxon test, the Friedman test, and variance. Finally, the practicality of the SPO algorithm is verified by four typical engineering problems and a real spacecraft trajectory optimization problem. The results show that the SPO algorithm can find superior results in 94.6% of the comparison tests and is a promising alternative for solving real-world problems.
With the increase in on-orbit maintenance and support requirements, the application of space manipulator is becoming more promising. However, how to control the vibration generated by the space manipulator has been a difficult problem to be solved. The advent of variable stiffness joint (VSJ) has brought about a dawn in solving this problem. But how to achieve coordinated control of joint angle and stiffness is still a problem to be solved, especially when considering system model parameter uncertainty, unknown disturbance and control input saturation. In order to realize the controllable attenuation of the vibration of the space flexible manipulator based on the variable stiffness joint, the dynamic model of the variable stiffness joint was constructed. Then the linear transformation and feedback linearization method are used to transform its complex nonlinear dynamic model system into a pseudo-linear system containing aggregate disturbance and input saturation constraints. This paper constructs a linear extended state observer (LESO) for estimating the state of unknown systems in pseudo-linear systems. Based on the idea of state feedback control, a Neural State Feedback Adaptive Robust (NSFAR) control is constructed by using Radial Basis Function Neural Network. The adaptive input saturation compensation control law is also designed by using Radial Basis Function Neural Network to deal with the input saturation compensation problem. The ultimate uniform bounded stability of the constructed system is proved by the stability analysis based on Lyapunov function. Finally, the effectiveness and superiority of the constructed tracking algorithm are verified by compared simulation and semi-physical experiment.
Accurately localizing the pupil is an essential requirement of some new human–computer interaction methods. In the past, a lot of work has been done to solve the pupil localization problem based on the appearance characteristics of the eye, but these methods are often specific to the scenario. In this paper, we propose an improved U-net network to solve the pupil location problem. This network uses the attention mechanism to automatically select the contribution of coded and uncoded features in the model during the skip connection stage of the U-net network in the channel and spatial axis. It can make full use of the two features of the model in the decoding stage, which is beneficial for improving the performance of the model. By comparing the sequential channel attention module and spatial attention module, average pooling and maximum pooling operations, and different attention mechanisms, the model was finally determined and validated on two public data sets, which proves the validity of the proposed model.
Abstract Unmanned aerial vehicle (UAV) relay communication is an important means to realize long-range wireless communication, where the lowest power loss could prolong the air residence time and increase the duration of the relay communication of UAV. In this paper, we present a UAV relay communication path planning based on the minimum encircling circle algorithm for the UAV working at the lowest power loss. We acquired the minimum radius of UAV relay communication based on for moving platform positions, and then determined the 3D position and flight path of UAV. Compared with the centroid enclosing circle algorithm, our simulation can reduce the signal transmitting power, realize the path planning of UAV relay communication under the lowest power loss.
Abstract Hierarchical titanate nanostructures were hydrothermally synthesized in concentrated base solutions using commercial titania powders as starting materials. By varying the base concentration, nanowire arrays, flowers of nanosheets and nanotubes, and urchin‐like nanostructures of nanowires and nanotubes were sequentially fabricated. If the NaOH concentration was higher than 6 M , hydrated Na 2 Ti 6 O 13 nanowire arrays, with nanowire diameters of 20–90 nm and an aspect ratio of 1100–5000, were produced at suitable reaction temperatures over a large area. In 10 M KOH solutions, aligned nanowires with a diameter of 30 nm and a lenght of 80 μm formed. In 4 M NaOH solutions, micrometer‐sized flowers of nanotubes and nanosheets formed. Reactions in 2 M NaOH solutions produced urchin‐like materials with a size of ca. 10 μm that were composed of nanotubes and nanowires. The adsorption behavior of the urchin‐like materials resembled macroporous materials with micropores. Since both base concentration and reaction temperature affected the reaction rate, the formation of various titanate nanostructures was proposed as a growth speed controlled process.
In order to solve the problem of harsh operating condition of the space manipulator, and the problem of not being able to afford great collision Momentum, this paper proposes a new technical method, namely soft-contact technology. According to the proposed method, the dynamic model of space flexible manipulator is built up based on Kane's equations. Then this paper makes a comparison simulation between dynamic models designed by ADAMS using same dynamic parameters. Simulation show that the curves obtained by Kane's equations and ADAMS are basically the same, so the correctness of the dynamic model of the space flexible manipulator based on Kane's equation has been proved.
As programing became more and more important, people are taking a large amount of work to help students to learn programming skills effectively. This paper applies a programming learning game called May's Journey to fit 5 debugging types including syntax, logical, structure, reasoning, and undefined debugging errors into programming levels. Then we can find out the reason why students make mistakes, and which debugging type would cause the mistakes of other debugging types. And we have 6 findings, (1) This paper proposes a student debugging model to describe how students make debugging errors, which is used for further analysis on student debugging behaviors. (2) This paper proposes to use group mean and with-in group variance based on student debugging model, which finds out the common debugging errors and personal debugging errors. (3) This paper proposes to extract student debugging patterns using Random forest, which identifies student debugging behaviors, so that students who have the same debugging pattern can be trained together. (4) This paper also proposes to use student debugging model-based SVM to extract student performance patterns, which identifies student performance changing over programming levels in terms of a specific debugging type. (5) This paper proposes to apply mean decrease accuracy and mean decrease Gini to identify the effectiveness of debugging types; and (6) this paper proposes to use a classification-based LSTM algorithm to predict debugging errors, which improves the predication accuracy a lot. Experiments and results are also provided to prove that our methods are valid and better.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
