Yulong Xing

Dr. Yulong Xing


Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
Verified
Journal Article
Open access
PDF file
Zhongzheng LiuZhongzheng Liu
Zhongzheng Liu
TZTao Zhang
Tao Zhang
YLYuhang Li
Yuhang Li
XZXin Zhao
Xin Zhao
YXYulong Xing
Yulong Xing
YGYanzhang Geng
Yanzhang Geng
Underwater acoustic energy transmission (UAET) is critical for sustaining long-term operations of underwater platforms, but its efficiency is constrained by the limited aperture of underwater receivers—requiring acoustic energy to be concentrated within a pre-defined target angular domain. Existing array-weighting methods face inherent limitations: traditional window-based techniques optimize mainlobe–sidelobe trade-offs rather than target-specific energy concentration, while intelligent algorithms suffer from high computational cost, quasi-optimality, and poor reproducibility. To address these gaps, this study proposes an array beam energy aggregation optimization method based on Rayleigh quotient eigenvalues for UAET. First, a rigorous mathematical model of the acoustic energy concentration problem was established: by defining a target-domain energy operator matrix RΘ with a Toeplitz–sinc structure (Hermitian positive definite), the energy-focusing problem was transformed into a tractable linear algebra problem. Second, the optimization objective of maximizing target-domain energy was formulated as a generalized Rayleigh quotient maximization problem, where the optimal amplitude weights correspond to the eigenvector of the maximum eigenvalue of RΘ—solved via Cholesky whitening and eigenvalue decomposition to ensure theoretical optimality and low computational complexity. Comprehensive validations were conducted via simulations and underwater physical experiments. Simulations on 1D uniform linear arrays and 2D 4-layer circular ring arrays showed that the proposed method outperformed traditional weighting methods and PSO in target angular energy concentration: for the 16-element linear array, its energy radiation efficiency in the 30° domain was 14% higher than classical methods (Blackman weighting). Underwater physical tests further confirmed its superiority: for the 4-layer circular ring array at 1 m, the acoustic energy efficiency in the 30° target domain reached 21.5% higher than Blackman weighting. Additionally, the method exhibited strong adaptivity (dynamic weight adjustment with target angular width) and scalability (performance improvement with array size), meeting UAET’s real-time and reliability requirements. This work provides a theoretically optimal and engineering-feasible solution for directional acoustic energy transfer in underwater environments, offering valuable insights for UAET system design.

28 July 2026 in Sensors
Verified
Journal Article
Open access
PDF file
KXKaihan Xie
Kaihan Xie
RQRuobing Qin
Ruobing Qin
JXJiang Xu
Jiang Xu
XWXiaofei Wang
Xiaofei Wang
YXYulong Xing
Yulong Xing
For sorting mechanisms that operate on a planar platform, the perception module may need to estimate object category, image-plane position, object extent and principal-axis orientation within the computing budget of a low-power device. This paper reports an oriented object detection workflow for a MaixCAM Pro (Shenzhen Sipeed Technology Co., Ltd., Shenzhen, China) waste-sorting prototype. The sensing output is represented by oriented bounding boxes (OBBs), which provide two-dimensional position, rectangular extent and principal-axis orientation for downstream sorting control. To adapt the detector to a small self-built dataset and INT8 deployment, class- and aspect-ratio-aware re-sampling (CAR-RS) repeats selected training images, while class- and aspect-ratio-aware calibration (CAR-Calib) selects a stratified calibration subset for post-training quantization. The YOLO11-OBB model is exported as a static ONNX graph, adapted by exposing the OBB head output tensors, compiled with TPU-MLIR and executed through native MaixPy OBB inference. On the original PC validation split, the baseline and CAR-RS models both obtain 0.995 mAP50, with mAP50–95 values of 0.935 and 0.932, respectively. CAR-RS reduces the angle mean absolute error from 3.403° to 3.201° for objects with aspect ratio greater than or equal to 1.3. On an additional 80-image cross-background test set that was not used for training or calibration, CAR-RS improves precision from 0.839 to 0.907, recall from 0.882 to 0.893 and mAP50 from 0.916 to 0.927, while mAP50–95 changes from 0.797 to 0.789. On a 100-image MaixCAM Pro validation set, the CAR-RS + CAR-Calib INT8 model achieves 0.961 precision, 0.994 recall, 0.977 F1-score and 0.980 mAP50. The deployed INT8 model occupies 11.85 MB and runs at 8.40 FPS. The results support the technical feasibility of executing an OBB detector on MaixCAM Pro, while also showing that independent-scene validation remains necessary for broader deployment.