Deformable Object Simulation
Overview
Deformable object simulation studies how soft bodies move, bend, stretch, collide, and sometimes tear under forces, enabling prediction and control in settings where rigid-body models fail. Approaches range from finite element and material point methods to spring-mass and Cosserat-rod models, often accelerated with implicit or GPU-parallel solvers; differentiable simulators and learned graph, transformer, and neural-field models help with parameter fitting and fast prediction. These tools support robot manipulation, surgical planning, tactile sensing, animation, and digital twins, where accurate dynamics can reduce costly real-world trial and error. A central challenge is balancing speed and generalization with physical fidelity: contact, material variation, cutting, and real-world variability remain difficult to calibrate and reproduce reliably.
Latest papers
84 papers17 papers in the last four weeks, up 42% on the four weeks before. 0.1% of all new papers.