Owing to the highly nonlinear dynamics and high number of nonlinear constraints, it is exceedingly difficult and computationally expensive to solve the cooperative trajectory planning problem of multimissile formation using existing approaches. To address this issue and improve the convergence property and computational efficiency, a bi-level sequential convex programming (SCP) method consisting of a system coordination level and an individual optimization level is proposed to solve the cooperative trajectory planning of missile formation. At the system level, the time consensus constraints are determined, the cooperative constraints that should be considered in the next iteration of the individual trajectory optimization level are identified, and the members that have converged are removed from the optimization sequence. As the number of members in the optimization sequence and the number of cooperative constraints considered in the individual SCP are clearly decreased, the convergence property and the computational efficiency of cooperative trajectory planning are evidently improved. At the individual level, the proposed method creatively proposes the innovative idea: based on the updated information of the system level, each member solves its individual trajectory optimization sub-problem independently and sequentially by gradually adding and tightening the cooperative constraints with the evolution of optimization iteration of SCP, which can further enhance the convergence property. Numerical simulations show that the proposed bi-level SCP method can effectively solve the multistage cooperative trajectory planning of multimissile formation with good convergence property, exhibiting the excellent scalability to the number of members and higher effectiveness. The comparison with the generation optimal control software (GPOPS) method further demonstrates the high efficiency of the proposed method.
Recently, building an efficient and robust model for object detection has attracted the attention of the vision community. Although binary networks have a fast inference speed, they cannot be used directly on mobile devices such as unmanned aerial vehicles (UAVs) because of their low detection accuracy. Different from improving the detection accuracy of a binary network by adjusting the network structure or adjusting the update gradient, we propose an improved binary neural network based on the block scaling factor XNOR (BSF-XNOR) convolutional layer. In addition, we propose a two-level densely connected network structure, which further enhances the network layer's feature representation capabilities. Experiments using the TensorFlow framework prove the effectiveness of our algorithm in improving accuracy. Compared with the original standard XNOR network, the mean average precision (mAP) detected by our algorithm on the PASCAL VOC dataset was improved. The experimental results on the VisDrone2019 UAV dataset confirm that our method achieves a better balance between speed and accuracy than previous methods. Our algorithm aims to guide and deploy high-precision binary networks on the embedded device and solves the problem of low-precision binary networks.
1 January 2025 in IEEE Robotics and Automation Letters
The Perspective-n-Point (PnP) problem has been widely studied in the literature and applied in various vision-based pose estimation scenarios. However, most existing methods ignore the anisotropy uncertainty of observations, as demonstrated in several real-world datasets in this letter. This oversight may lead to suboptimal and inaccurate estimation, particularly in the presence of noisy observations. To this end, we propose a generalized maximum likelihood PnP solver, named GMLPnP, that minimizes the determinant criterion by iterating the generalized least squares procedure to estimate the pose and uncertainty simultaneously. Further, the proposed method is decoupled from the camera model. Results of synthetic and real experiments show that our method achieves better accuracy in common pose estimation scenarios, GMLPnP improves rotation/translation accuracy by 4.7%/2.0% on TUM-RGBD and 18.6%/18.4% on KITTI-360 dataset compared to the best baseline. It is more accurate under very noisy observations in a vision-based UAV localization task, outperforming the best baseline by 34.4% in translation estimation accuracy.
20 March 2023 in Multimedia Tools and Applications
Drone-based geo-localization is an image retrieval task which is the foundation of many drone-based multimedia applications, such as object detection, drone navigation and mapping. It is challenging due to the large visual appearance changes caused by viewpoint variation and time misalignment. Existing methods primarily focus on global representation embedding while disregarding the local features. We propose a CNN-based model containing a global and a local branch to extract features in these two perspectives and then features are subsequently aggregated by orthogonal fusion. We achieve competitive results on University-1652/160k datasets among the ViT-based state-of-the-art models. Experimental and qualitative results that validate the effectiveness of our solution are also shown.
1 April 2020 in Journal of Physics: Conference Series
We propose an end-to-end rotational motion deblurring method based on conditional generation adversarial networks. The proposed method calculates the blur path value of each pixel on the rotational motion blurred image to provide a priori information of its blur degree, and then connects it to the blurred image as the input of the network. In addition, a rotational motion blurred image dataset is produced, which contains different degrees of rotational motion blurred images, as an evaluation dataset for the method to the effect of rotational motion deblurring. Experiments show that the proposed method is superior to existing end-to-end deblurring methods in both qualitative and quantitative analysis when dealing with different degrees of rotational motion blur.