Aeromagnetic surveys, renowned for their operational flexibility and high efficiency, serve as a crucial technique for measuring the geomagnetic field. However, aeromagnetic surveys are easily affected by magnetic int...
ISBN:
(数字)9798350352627
ISBN:
(纸本)9798350352634
Aeromagnetic surveys, renowned for their operational flexibility and high efficiency, serve as a crucial technique for measuring the geomagnetic field. However, aeromagnetic surveys are easily affected by magnetic interference from navigation platforms, making the compensation of aeromagnetic interference a crucial step in the measurement process. To address the inadequate consideration of nonlinear magnetic field interference in traditional compensation algorithms, this paper introduces an aeromagnetic compensation approach based on broad learning system. The broad learning system employs an incremental learning mechanism aimed at enhancing the precision of the network alongside the increase in nodes. With each expansion of the network node, computation is streamlined to calculating the pseudo-inverse of the expansion node, eliminating the necessity for retraining the entire network structure. Leveraging the nonlinear fitting characteristics of the broad learning system, this paper improves the accuracy of aeromagnetic interference compensation. Through UAV flight experiments, the broad learning system is compared with methodologies using particle swarm optimization (PSO) and BP neural network. Compared with PSO, training time was reduced by $21.3 \%$ and magnetic interference by $33.6 \%$. Compared with BP neural networks, training time was reduced by $34.9 \%$ and magnetic interference by $28.6 \%$. This paper provides references and ideas for the selection of aeromagnetic interference compensation algorithms.
In the field of radar data processing, track interruption seriously affects target tracking, track fusion, and other *** existing track segment association algorithms have low correlation accuracy in dense distributed...
In the field of radar data processing, track interruption seriously affects target tracking, track fusion, and other *** existing track segment association algorithms have low correlation accuracy in dense distributed or long-time interruption situations. To this purpose, a dense multi-target track segment association(DMTTSA) algorithm is proposed. Firstly, two identical networks based on the multi-head probability sparse(ProbSparse) self-attention are used to capture the long-term dependencies of the tracks. Then, the bidirectional quadruplet hard sample loss(BiQuaHard loss) is constructed to make the tracks belonging to the same targets closer and the tracks belonging to the different targets farther. Finally, DMTTSA takes the closest track pairs in the feature space as the associated tracks and divides the unassociated tracks into the birth and dead tracks in chronological order. Some comparative experiments are carried out to show the anti-noise performance of the DMTTSA, as well as the effectiveness of solving the problem of dense multi-target track interruption.
This paper provides an oscillation trajectory optimization and control method for two-link underactuated manipulators(UMs) in a vertical *** proposed method solves the problem that the UMs cannot always enter the bala...
This paper provides an oscillation trajectory optimization and control method for two-link underactuated manipulators(UMs) in a vertical *** proposed method solves the problem that the UMs cannot always enter the balance region in the partitioning ***,we establish the system dynamic model,and analyze the system couple ***,we program an oscillation trajectory for the active link,and use the intelligent method to obtain the trajectory parameters,so ensuring the system can reach the area adjacent to the target position through tracking ***,we design the controller to realize the stable control at the target ***,the simulation results show the effectiveness and generality of the control strategy.
With the rapid development of deep learning, it has been widely applied in fields such as computer vision, natural language processing, and robotics. Despite the superior performance of deep learning in object detecti...
详细信息
In order to address the problem of current object detection models being too large to be deployed on robot controllers, this paper proposes improvements to YOLOv5 for real-time detection. The YOLOv5s model is pruned a...
详细信息
This paper presents a novel method of distinguishing signal, the way expected to reduce the nuisance alarm rate since the high nuisance alarm rate will restrict the capability of phase-sensitive optical time-domain re...
详细信息
Multi-feature fusion is a useful way to improve the classification of hyperspectral image (HSI). But the multi-feature fusion is usually at the decision level of classifier, which causes less link between features or ...
详细信息
Stable control and active disturbance rejection strategy is proposed for planar 2R underactuated robot via intelligent algorithm in this paper. At first, we build the dynamic model and describe the control characteris...
详细信息
Although skeleton-based gesture recognition based on supervised learning has made promising achievements, the reliance on large amounts of annotation for training poses a significant cost. This paper addresses semi-su...
详细信息
Detecting the lithium battery surface defects is a difficult task due to the illumination reflection from the surface. To overcome the issue related to labeling and training big data by using 2D techniques, a 3D point...
详细信息
暂无评论