GAMES Webinar 2022 – 240期(高效和可微的带碰撞的软体仿真) | 蓝磊(犹他大学),李一飞(麻省理工学院)

【GAMES Webinar 2022-240期】(模拟专题-高效和可微的带碰撞的软体仿真)

报告嘉宾:蓝磊(Utah University)


报告题目:Penetration-free Projective Dynamics on the GPU


Efficient physics-based simulation with a penetration-free guarantee has always been a big challenge. In this talk, we introduce a GPU-based method for deformable simulation. This method integrates projective dynamics (PD) and incremental potential contact (IPC) by reworking the local-global iteration modality. We present an aggregated Jacobi solver to speed up solving the global step of PD. This solver better exploits the computation capacity of the GPU than the Jacobi solver. Together with a faster CCD processing under the IPC mechanism, our method offers good computational efficiency and a penetration-free guarantee for complicated scenes at the same time.


Lei Lan is currently a postdoctoral researcher at the University of Utah. He previously worked as a  postdoctoral researcher at Clemson University and received Ph.D. from Xiamen University. His research interests focus on physics-based simulation, 3D modeling, robotics, and deep learning.




报告题目:DiffCloth: Differentiable Cloth Simulation with Dry Frictional Contact


Cloth simulation has wide applications in computer animation, garment design, and robot-assisted dressing. This work presents a differentiable cloth simulator whose additional gradient information facilitates cloth-related applications. Our differentiable simulator extends a state-of-the-art cloth simulator based on Projective Dynamics (PD) and with dry frictional contact. We draw inspiration from previous work to propose a fast and novel method for deriving gradients in PD-based cloth simulation with dry frictional contact. Furthermore, we conduct a comprehensive analysis and evaluation of the usefulness of gradients in contact-rich cloth simulation. Finally, we demonstrate the efficacy of our simulator in a number of downstream applications, including system identification, trajectory optimization for assisted dressing, closed-loop control, inverse design, and real-to-sim transfer. We observe a substantial speedup obtained from using our gradient information in solving most of these applications.


Yifei Li is a second-year Ph.D. student at MIT CSAIL, advised by Prof. Wojciech Matusik. Her research interests lie in the intersection of physical simulation, computational design, and machine learning. Before coming to MIT, she received her bachelor’s degree in Computer Science from Carnegie Mellon University where she worked with Prof. Jessica Hodgins. She has also spent time working at NVIDIA, Facebook AI Research, Google, and Activision & Blizzard. She is a recipient of the MIT Stata Family Presidential Fellowship.





GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播?”及“如何加入GAMES微信群?”的信息;

You may also like...