Search arXiv⌕ Search

arXiv subjects

Wenxia Xu

Publications and source records attributed to Wenxia Xu.

2 recordsLinked to original sources

Realization of valley-controlled orbital angular momentum via inter-orbital coupling

Orbital angular momentum (OAM) represents a fundamental degree of freedom (DoF) for understanding novel phenomena in solids and synthetic lattices. However, the generation and control of OAM in lattice systems remain challenging. Here, we experimentally observe valley-controlled OAM states in inversion-symmetry-broken photonic honeycomb lattices consisting of s- and higher-orbital sublattices. These OAM states manifest as chiral vortex arrays in real space. By selectively exciting s-orbital modes carrying no initial OAM at either the K or K' valley, we observe higher-orbital OAM modes generated via inter-orbital coupling in neighboring sites. Specifically, s-p orbital coupling generates OAM states with topological charge l=+1 or l=-1, while s-d orbital coupling yields higher-charge OAM states with l=+2 or l=-2. The sign of the topological charge is determined by the valley DoF, enabling deterministic control of vortex chirality across the entire array. Our theory further reveals that a local vortex-chirality symmetry enforces the valley-dependent selection of vortex chirality at the K and K' valleys, establishing a direct connection between the valley DoF and real-space vortex chirality. These results establish a versatile platform for engineering chiral vortex arrays via orbital physics and exploring orbital- and topology-mediated phenomena in photonic systems and beyond.

physics.optics↗

DisGNet: A Distance Graph Neural Network for Forward Kinematics Learning of Gough-Stewart Platform

In this paper, we propose a graph neural network, DisGNet, for learning the graph distance matrix to address the forward kinematics problem of the Gough-Stewart platform. DisGNet employs the k-FWL algorithm for message-passing, providing high expressiveness with a small parameter count, making it suitable for practical deployment. Additionally, we introduce the GPU-friendly Newton-Raphson method, an efficient parallelized optimization method executed on the GPU to refine DisGNet's output poses, achieving ultra-high-precision pose. This novel two-stage approach delivers ultra-high precision output while meeting real-time requirements. Our results indicate that on our dataset, DisGNet can achieves error accuracys below 1mm and 1deg at 79.8\% and 98.2\%, respectively. As executed on a GPU, our two-stage method can ensure the requirement for real-time computation. Codes are released at https://github.com/FLAMEZZ5201/DisGNet.

cs.RO↗