Efficient contact dynamics for deformable bodies via novel signed distance field reconstruction method

Efficient and accurate contact detection remains a major bottleneck in deformable-body contact dynamics, especially for problems involving complex geometries, large rigid-body motions, and finite deformations. To address this issue, this paper proposes a rapid signed distance field (SDF) reconstruction method for efficient contact detection of deformable bodies without frequent voxelization-based reconstruction. The method first employs axis-aligned bounding boxes (AABBs) in the broad-phase to identify potential contact pairs, and then uses SDFs in the narrow-phase when contact is imminent. To overcome the loss of validity of precomputed SDFs during deformation, a hybrid reconstruction strategy is developed: a first-order Taylor expansion is used for rapid reconstruction under small displacements and small deformations, whereas an inverse mapping method is adopted for large rigid-body motions and finite deformations. Based on the reconstructed SDF, the contact gap and contact normal can be directly evaluated through trilinear interpolation. The effectiveness of the proposed method is validated through representative examples by comparison with ABAQUS. The results show that the proposed method significantly improves computational efficiency while maintaining accuracy.