CAD-CAE Integration: From Theory and Algorithms to Engineering Practice
Date:
Slides, Photo1, Photo2, Photo3, Photo4, Photo5, Photo6
I gave an invited talk titled “CAD CAE 一体化：从理论、算法到工程实践” (CAD-CAE Integration: From Theory and Algorithms to Engineering Practice), reviewing recent progress in bridging Computer-Aided Design (CAD) and Computer-Aided Engineering (CAE) through isogeometric analysis — from analysis-suitable parameterization theory, to r-adaptive and enriched discretization algorithms, and finally to engineering applications such as thermal simulation in metal additive manufacturing and structured mesh generation for twin-screw machines. As part of this visit, I also joined the “新锐人才沙龙暨校企行业对话” (Young Talent Salon & Industry–University Dialogue) panel discussion.
In addition to the invited talk, our group brought one conference paper and two posters to CSIAM GDC 2026 — a concentrated showcase of our recent work:
Conference paper: “基于B样条滚动时域优化的无人机避障方法” (A B-Spline-Based Receding-Horizon Trajectory Optimization Method for UAV Obstacle Avoidance). To meet the combined requirements of trajectory smoothness, feasibility, and online solution efficiency for real-time UAV obstacle avoidance in complex environments, this paper proposes a B-spline-based receding-horizon trajectory optimization method. Exploiting the local support property and high-order continuity of B-splines, the UAV velocity trajectory is parameterized using a quasi-uniform open knot vector. The control points are partitioned into fixed and free control points, and a linear constraint system is solved analytically so that the trajectory strictly satisfies velocity, acceleration, and higher-order derivative constraints at the start of each receding horizon, guaranteeing continuity and feasibility. Flight performance and obstacle-avoidance constraints are then unified within a penalty-function framework combined with the receding-horizon strategy to construct a multi-objective cost function coordinating local avoidance with the global trajectory. Numerical simulations show that the method stably generates smooth, dynamically feasible, directly executable trajectories under single-obstacle, multi-obstacle, complex-reference-trajectory, and moving-obstacle scenarios, with per-step solution time far below the execution horizon, demonstrating strong real-time performance and robustness. This paper received the Conference Best Paper Award.
Poster: Extended r-adaptive isogeometric analysis for weak-discontinuous problems. This poster received the Conference Best Poster Award.
Keywords: CAD-CAE integration, Isogeometric analysis, Trajectory optimization, UAV obstacle avoidance, Additive manufacturing