Preprint
Sep 2026
SAGE: Safety-Aligned Gradient Enforcement for Human--Robot Collaboration
Safety-aligned gradient enforcement (SAGE) is presented, which uses a differentiable finite-penalty proposal map while retaining the exact CBF quadratic program for execution, preserving constraint-normal sensitivity to internalize repeatedly active safety constraints.
Yi-Sen Li, Hao Zhang, Ru Geng et al.
· 0 citations