2021 Outstanding Doctoral Dissertation Award: Minchen Li

Award Winners

Source: ACM SIGGRAPH Citation

ACM SIGGRAPH is pleased to present Minchen Li, PhD, with the 2021 Outstanding Doctoral Dissertation Award. Minchen’s dissertation presents a breakthrough in the notoriously challenging and long-standing problem of robust frictional contact simulation in nonlinear solid dynamics with guarantees of non-intersection.

Contact, including contact with friction, is a classical topic in many scientific areas such as computer graphics, mechanical engineering, computational mechanics, fabrication design, and robotics. A broad range of algorithmic formulations and time discretizations have been proposed over time, offering different trade-offs between physical correctness and restrictions on shapes (e.g., smoothness or convexity), and between algorithmic robustness and computational efficiency. Despite a large body of previous work, numerical methods before Minchen’s doctoral work were either computationally efficient or physically correct, but not both: no existing approach with practical runtimes could offer guaranteed global injectivity (i.e., no part of any material is allowed to overlap with any other portion of a material during the simulation).

Minchen’s dissertation describes a number of contributions centered around his Incremental Potential Contact (IPC) method, a new model and algorithm for the variational solution of implicitly time-stepped nonlinear elastodynamics subject to frictional contact. His approach combines a geometrically exact formulation for collision gap functions, a smoothed friction formulation making it possible to cast it in variational form, and the use of a barrier-based interior point method for optimization. By designing a reliable and unified method that works across very challenging conditions and for volumes, shells, rods and particles alike, Minchen’s work pushes the stability, robustness and expressiveness of deformable body simulations to unprecedented levels. It offers unconditional feasibility and differentiability, maintaining intersection-free and inversion-free trajectories independent of material parameters, time step sizes, impact velocities, severity of deformation, and boundary conditions. This major step forward in simulation outperforms state-of-the-art methods in terms of robustness to the most challenging scenarios. It should thus enable a number of applications in machine learning, fabrication design, or robotics that rely on automated and differentiable simulation output across parameter changes. Dr. Li has released his research code from his dissertation as open-source, with already numerous users in academia and industry.

With his doctoral dissertation, Dr. Li has advanced the scale, robustness, and efficiency of physically based animation tools. The SIGGRAPH community thus recognizes Minchen Li for these extraordinary achievements with the 2021 ACM SIGGRAPH Doctoral Dissertation Award, and looks forward to further developments of his work in the years to come.