Yi Zhang

Ms. Yi Zhang

Nanjing University of Aeronautics and Astronautics (NUAA)
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 August 2021 in Sensors
Verified
Journal Article
Open access
PDF file
KZKedong Zhao
Kedong Zhao
YSYongrong Sun
Yongrong Sun
YZYi Zhang
Yi Zhang
HLHua Li
Hua Li
In aerial refueling, there exists deformation of the circular feature on the drogue’s stabilizing umbrella to a certain extent, which causes the problem of duality of position estimation by a single circular feature. In this paper, a monocular visual position and attitude estimation method of a drogue is proposed based on the coaxial constraints. Firstly, a procedure for scene recovery from one single circle is introduced. The coaxial constraints of the drogue are proposed and proved to be useful for the duality’s elimination by analyzing the matrix of the spatial structure. Furthermore, we came up with our method, which is composed of fitting the parameters of the spatial circles by restoring the 3D points on it, using the two-level coaxial constraints to eliminate the duality, and optimizing the normal vector of the plane where the inner circle is located. Finally, the effectiveness and robustness of the method proposed in this paper are verified, and the influence of the coaxial circle’s spatial structure on the method is explored through simulations of and experiments on a drogue model. Under the interference of a large amount of noise, the duality elimination success rate of our method can also be maintained at a level that is more than 10% higher than others. In addition, the accuracy of the normal vector obtained by the fusion algorithm is improved, and the mean angle error is reduced by more than 26.7%.

7 November 2023
Verified
Conference Paper
Closed access
JMJing Ma
Jing Ma
Peng LiuPeng Liu
Peng Liu
HLHao Liu
Hao Liu
YTYonglin Tian
Yonglin Tian
FZFengyi Zhang
Fengyi Zhang
ZWZhaojie Wang
Zhaojie Wang
In view of the problems existing in the field of marine equipment test and training, such as complex construction of large-scale system test scenes, difficulty in equipment coordination, difficulty in training close to realistic scene, and lack of immersion in daily training, this paper proposes the concept of parallel ships based on ACP theory, aiming to make use of advanced technologies and concepts such as metaverse and digital twins. The test and training environment of "digital model + principle prototype" of parallel ships is constructed to support the realization of large-scale joint test and face-to-face confrontation training. Considering the construction of parallel ship environment, this paper focuses on the key issues of parallel ship system framework, proposes the intelligent marine equipment test and training framework, and further designs the parallel ship description intelligence, parallel ship prediction intelligence and parallel ship guidance intelligence respectively. The professional description architecture for test verification, the agent cultivation architecture for adversarial game and the strong real-time guidance architecture for behavior control are proposed, which provide a new way of thinking for carrying out the marine equipment test and training that is close to reality, high immersion and high credibility.