The upper limb motion intention recognition method based on electroencephalography (EEG) and surface electromyography (sEMG) fusion has achieved significant results in fields such as prosthetic control. However, most ...
The upper limb motion intention recognition method based on electroencephalography (EEG) and surface electromyography (sEMG) fusion has achieved significant results in fields such as prosthetic control. However, most existing fusion methods use manual means to select features, which cannot capture temporal and spatial features at different scales, and ignore the correlation features between the two types of signals. To address these issues, this article proposes a fusion recognition method for upper limb motion intention EEG and sEMG based on Multi-scale Convolution, Polarized Self-Attention, and Cross Intelligence Integration Module. This article extracts multidimensional temporal and spatial features of EEG and sEMG through multi-scale convolution, and introduce polarized self-attention mechanism to filter the extracted multi-scale features. Simultaneously using cross enhancement strategy to extract correlation features between EEG and sEMG. Finally, the features are input into the classification network for recognition. This method was validated on the Jeong database, and the results showed that compared with CNN-LSTM and EEGNet, the recognition accuracy of this method increased by 2.63% and 3.15%, respectively.
Breast cancer is one of the most common cancers among female diseases. Since the classification accuracy of pathological images is crucial to the diagnosis of breast cancer, in order to reduce the error of manual diag...
详细信息
In the intelligent transportation system, vehicle detection is one of the essential technologies in obstacle avoidance and navigation, however the existing vehicle detection methods cannot meet the actual needs. This ...
详细信息
Ascent trajectory optimization problem of air-breathing hypersonic vehicles is a highly nonlinear and nonconvex problems. Most of the early works focus on the traditional indirect method, which needs to derive the com...
Ascent trajectory optimization problem of air-breathing hypersonic vehicles is a highly nonlinear and nonconvex problems. Most of the early works focus on the traditional indirect method, which needs to derive the complete first-order necessary conditions of the trajectory optimization problem. The derivation process is too complicated and error-prone. Additionally, indirect method has a high demand on the initial guess, and it needs to give the initial guess of covariant variables without physical significance. In this paper, we solve the ascent trajectory optimization problem directly using Radau Pseudospectral Method (RPM). Firstly, the complex three-dimensional ascent trajectory optimization problem is established in detail. Conmmon inequality path constraints including those on dynamic pressure and aerodynamic bending moment are taken into account. The performance index is given as maximizing the final mass considering minimizing the fuel consumption. Subsequently, the ascent trajectory optimization problem is transformed into a nonlinear programming problem (NLP) by RPM. Finally, the ascent trajectory optimization for Generic Hypersonic Aerodynamic Model Example (GHAME) is solved by RPM and the optimal results demonstrate the rapidity, effectiveness and high precision of RPM. The comparison between optimal trajectories with and without path constraints shows that path constraints increase fuel consumption.
This paper analyzed the structure and characteristics of cooperative guidance system of aircraft formation, puts forward a multi-level simulation system architecture of centralized/distributed driving to solve the coo...
详细信息
In this paper, the output synchronization in large-scale discrete-time networks is examined by utilizing the novel phase tool, where the agent dynamics are allowed to be significantly heterogeneous. The synchronizatio...
详细信息
ISBN:
(数字)9783907144107
ISBN:
(纸本)9798331540920
In this paper, the output synchronization in large-scale discrete-time networks is examined by utilizing the novel phase tool, where the agent dynamics are allowed to be significantly heterogeneous. The synchronization synthesis problem is formulated and thoroughly investigated, with the goal of characterizing the allowable heterogeneity among the agents to ensure synchronization under a uniform controller. The solvability condition is provided in terms of the phases of the residue matrices of the agents at the persistent modes. When the condition is satisfied, a design procedure is given, producing a low-gain synchronizing controller. Numerical examples are given to illustrate the results.
In the engineering field, there is a requirement for re recognition, including the identity recognition of cooperative and non-cooperative targets, that is, different sensors shoot targets in the same scene, and the i...
In the engineering field, there is a requirement for re recognition, including the identity recognition of cooperative and non-cooperative targets, that is, different sensors shoot targets in the same scene, and the identity recognition of targets from two perspectives is required, also known as target accurate retrieval. This method is widely used in fields such as automatic driving and military strikes. In the actual application scenario, deploying the re-recognition algorithm on the embedded processor needs to ensure the real-time computing speed and accuracy, so the complexity of the deep learning algorithm must be reduced, and pruning and lightweight processing are required. In the engineering field, when the image acquired by the sensor is subject to natural interference such as cloud and fog weather or fire and smoke, the difficulty of re-recognition increases. This paper adopts the lightweight improvement of ConvNeXt network model, and fine-tuned it on the basis of its original feature extraction function, so that it can successfully migrate to infrared image classification and also complete the extraction of infrared image target detail distinguishing features, and at the same time introduce the alignment of similar targets in infrared cross-view recognition. In the original similarity calculation, the average value is directly obtained from the spatial dimension of the output feature layer, which is replaced by the weight of the position in each space for estimation. The feature information of the multi-view marine ship target is extracted from the multi-sensor marine image, and its similarity is calculated, so as to determine the identity, thus reducing the support conditions and reconnaissance costs, and improving the accurate detection and tracking ability under the diverse dynamic reconnaissance conditions.
Driver attention prediction implies the intention understanding of where the driver intends to go and what object the driver concerned about, which commonly provides a driving task-guided traffic scene understanding. ...
详细信息
Driver attention prediction implies the intention understanding of where the driver intends to go and what object the driver concerned about, which commonly provides a driving task-guided traffic scene understanding. ...
Driver attention prediction implies the intention understanding of where the driver intends to go and what object the driver concerned about, which commonly provides a driving task-guided traffic scene understanding. Some recent works explore driver attention prediction in critical or accident scenarios and find a positive role in helping accident prediction, while the promotion ability is constrained by the prediction accuracy of driver attention maps. In this work, we explore the network connection gating mechanism for driver attention prediction (Gate-DAP). Gate-DAP aims to learn the importance of different spatial, temporal, and modality information in driving scenarios with various road types, occasions, and light and weather conditions. The network connection gating in Gate-DAP consists of a spatial encoding network gating, long-short-term memory network gating, and information type gating modules. Each connection gating operation is plug-and-play and can be flexibly assembled, which makes the architecture of Gate-DAP transparent for evaluating different spatial, temporal, and information types for driver attention prediction. Evaluations on DADA-2000 and BDDA datasets verify the superiority of the proposed method with the comparison with state-of-the-art approaches.
In non-cooperative scenarios, wideband frequency hopping (FH) communication reconnaissance including FH signal detection, parameter estimation, and network sorting under single-channel reception is challenging. The FH...
In non-cooperative scenarios, wideband frequency hopping (FH) communication reconnaissance including FH signal detection, parameter estimation, and network sorting under single-channel reception is challenging. The FH pattern contains the most information about FH signals. So it is the core of FH parameter estimation. Based on the task analysis, this paper proposes a blind prediction framework combining deep learning (DL) and clustering methods to improve the accuracy and efficiency of FH pattern estimation for mixed signals in the wideband spectrum. The unique advantage of this framework is that it requires no prior signal information and anchors but exploits the inherent time-frequency (TF) properties of asynchronous FH signals. It has a strong generalization ability and can adapt to signals of any shape. Moreover, a new comprehensive evaluation metric, normalized root mean square error including missed detection (NRMSE-MD), suitable for sequence data prediction is proposed to evaluate the level of missed detection and false detection in signal monitoring. Experimental results demonstrate the superiority of the proposed framework and the effectiveness of the proposed evaluation metric.
暂无评论