Haoxiang Li

Haoxiang Li

Simulation Team Lead at a robotics startup in Shenzhen. Ph.D. in Computer Graphics, Nanyang Technological University (NTU), Singapore. I build GPU-accelerated rigid–soft coupled simulation engines, generate large-scale robot interaction data for world-model training, and work on nonlinear optimization for multi-material 3D printing design.

Optimizing heterogeneous elastic material distributions on 3D models

Published in Computer-Aided Design, vol. 175, 103748 (SPM 2024 special issue), 2024

We propose a solution for optimizing the spatial distribution of heterogeneous elastic materials on a 3D part to achieve desired deformation behaviour. The method formulates the design as a discrete optimization over a finite set of base materials and solves it by combining graph cuts with ADMM, which improves both optimization accuracy and printability. A nonlinear finite-element simulator and physical experiments validate the results, and the pipeline has been applied to the material layout design of 3D-printed customized insoles.

Recommended citation: Haoxiang Li, Wenjing Zhang, Jianmin Zheng, Edward D. Davis, Jun Zeng. (2024). "Optimizing heterogeneous elastic material distributions on 3D models." Computer-Aided Design, 175, 103748.
Download Paper