Sheng-lun Yi

Dr. Sheng-lun Yi


Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

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

Representative Publications

View All Publications
24 August 2021 in IEEE Transactions on Automatic Control
Verified
Journal Article
Closed access
SYSheng-lun Yi
Sheng-lun Yi
MZMattia Zorzi
Mattia Zorzi
In this article, we consider a robust state-space filtering problem in the case that the transition probability density is unknown and possibly degenerate. The resulting robust filter has a Kalman-like structure and solves a minimax game: the nature selects the least favorable model in a prescribed ambiguity set, which also contains non-Gaussian probability densities, while the other player designs the optimum filter for the least favorable model. It turns out that the resulting robust filter is characterized by a Riccati-like iteration evolving on the cone of the positive-semidefinite matrices. Moreover, we study the convergence of such iteration in the case that the nominal model is with constant parameters on the basis of the contraction analysis in the same spirit of Bougerol. Finally, some numerical examples show that the proposed filter outperforms the standard Kalman filter.

11 November 2022 in Journal of the Franklin Institute
Verified
Journal Article
Closed access
SYSheng-lun Yi
Sheng-lun Yi
MZMattia Zorzi
Mattia Zorzi

25 October 2024 in ICCK Transactions on Sensing, Communication, and Control
Verified
Journal Article
Open access
PDF file
FWFafa Wang
Fafa Wang
SYSheng-lun Yi
Sheng-lun Yi
The efficient extraction and fusion of video features to accurately identify complex and similar actions has consistently remained a significant research endeavor in the field of video action recognition. While adept at feature extraction, prevailing methodologies for video action recognition frequently exhibit suboptimal performance in the context of complex scenes and similar actions. This shortcoming arises primarily from their reliance on uni-dimensional feature extraction, thereby overlooking the interrelations among features and the significance of multi-dimensional fusion. To address this issue, this paper introduces an innovative framework predicated upon a soft correlation strategy aimed at augmenting the representational capacity of features by implementing multi-level, multi-dimensional feature aggregation and concatenating the temporal features produced by the network. Our end-to-end multi-feature encoding soft correlation concatenation aggregation layer, situated at the temporal feature output terminal of the Video Action Recognition network, proficiently aggregates and integrates the output temporal features. This approach culminates in producing a composite feature that cohesively unifies multi-dimensional information, markedly enhancing the network's competency in differentiating analogous video actions. Empirical findings demonstrate that the approach delineated in this paper bolsters the efficacy of video action recognition networks, achieving a more thorough depiction of images, and yielding superior accuracy and robustness.

14 December 2006 in IEE Proceedings - Radar, Sonar and Navigation
Verified
Journal Article
Closed access
WDW. Dangwei
W. Dangwei
MXM. Xinyi
M. Xinyi
SYSheng-lun Yi
Sheng-lun Yi
On the basis of likelihood ratio test (LRT) theory, a new method for radar target identification with matching pursuits, including the estimation method for target scattering response and the extraction technique for the reference atom dictionary to characterise the target scattering, is introduced. When compared with the generalised LRT (GLRT) target identification method, there is a very important contribution in our method that the whole scattering waveform, not only a few main poles, is used to identify targets without any prior parameterisation model hypothesis. Furthermore, the identification to candidate targets can be performed in an almost aspect-independent manner over a full aspect angle in white Gaussian noise through a finite size reference atom dictionary. Simulation results using scattered responses synthesised from weighted sums of some exponentially damped sinusoids and calculated scattering signatures of three thin wires through a time-domain electric field integral equation are presented to contrast the performance of our method to the GLRT and the best LRT target identification method as a function of signal-to-noise ratio.

9 March 2024 in International Journal of Robust and Nonlinear Control
Verified
Journal Article
Closed access
SYSheng-lun Yi
Sheng-lun Yi
TSTingli Su
Tingli Su
ZTZhenYun Tang
ZhenYun Tang

18 December 2024 in ICCK Transactions on Sensing, Communication, and Control
Verified
Journal Article
Open access
PDF file
HSHuihui Shi
Huihui Shi
SXShuzong Xie
Shuzong Xie
QCQiang Chen
Qiang Chen
SHShuangyi Hu
Shuangyi Hu
SYSheng-lun Yi
Sheng-lun Yi
In engineering applications, high-precision tracking control is crucial for robotic manipulators to successfully complete complex operational tasks. To achieve this goal, this study proposes an adaptive tunable predefined-time backstepping control strategy for uncertain robotic manipulators with external disturbances and model uncertainties. By establishing a novel practical predefined-time stability criterion, a tunable predefined-time backstepping controller is systematically presented, allowing the upper bound of tracking error settling time to be precisely determined by adjusting only one control parameter. To accurately address lumped uncertainty, two updating laws are designed: a fuzzy weight updating law and a boundary adaptive updating law, which together reduce dependence on system model knowledge. In addition, the singularity problem in the predefined-time design process is effectively avoided by constructing the hyperbolic tangent function. The efficacy of the proposed control strategy is verified through numerical simulations.