Shunxing Ning

Shunxing Ning

Auto-generated profile, based on publicly available information

Shunxing Ning

Hunan Institute of Engineering, Xiangtan, Hunan, CHINA

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
20 December 2024
Verified
Conference Paper
Open access
QWQin Wan
Qin Wan
PXPin Xu
Pin Xu
SNShunxing Ning
Shunxing Ning
ZLZhi Li
Zhi Li
This paper proposes a vision-based formation control method for multi-robot systems in the absence of inter-robot communication, employing a leader-follower scheme with a single Kinect camera as the sole sensor. By utilizing a pose estimation algorithm based on RGBD images, the relative position and orientation of the leader are captured. An anticipated formation shape aligned in the same movement direction is generated using a dipole vector field, ensuring formation stability. To address the limited field of view of the Kinect camera, an auxiliary system is designed to transform visibility constraints into input constraints, enabling effective control under non-communicative conditions. Simulation and experimental results demonstrate that the proposed method achieves real-time and robust formation control despite visibility limitations, simplifying the system architecture and enhancing practicality.

9 April 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
SNShunxing Ning
Shunxing Ning
QWQin Wan
Qin Wan
TLTong Liu
Tong Liu
YWYaonan Wang
Yaonan Wang
HZHui Zhang
Hui Zhang
LHLingxiang Huang
Lingxiang Huang
Efficient measurement of the distance between blades and the tower is crucial for the safe operation of wind turbines. Aiming at the technical difficulties of real-time detection and measurement of blade clearance distance, this paper proposes a Fan blade clearance measurement method based on the Lightweight Case Segmentation Network (FLCS-Net). Firstly, to effectively balance the accuracy and computational complexity of the detection network, we constructed an instance segmentation network based on the YOLOv8-seg framework, which adopts the lightweight EfficientNetV2 backbone network. Secondly, a new type of parallel hybrid architecture module was developed to enhance the model ability to extract multi-scale features. Thirdly, we collected the operational data of wind turbine blades under sunny, cloudy and foggy conditions, and constructed a dataset Fan_blades dedicated to wind turbine blade clearance measurement. Finally, we verified that FLCS-Net can effectively measure the distance between blades and towers on the Fan_blades dataset and in real-world wind power scenarios, thereby ensuring the safe operation of wind turbines. The dataset and code for measuring the blade clearance of wind turbines will be made public after the paper is accepted.