Cubic B-spline-based trajectory planning for robotic welding of curved surfaces

This paper presents a joint-space trajectory planning approach based on cubic B-spline interpolation for robotic welding of autoclave end caps, addressing the trajectory planning challenges arising from complex curved surfaces with spatially varying curvature. Weld seam data are acquired using a line-laser vision system, from which joint-space interpolation points are obtained through inverse kinematic computation. Cubic B-spline curves are then constructed by introducing auxiliary control vertices to ensure trajectory continuity and smoothness during curvature transitions. The proposed method is implemented in a six-degree-of-freedom industrial welding robot system and evaluated through numerical simulation and offline programming environments. Joint displacement, velocity, and acceleration profiles are analyzed to examine motion continuity and stability. Simulation and experimental validation are conducted on scaled autoclave end caps to assess trajectory tracking behavior and positional accuracy along the welding path. The results demonstrate that the proposed trajectory planning framework enables smooth joint-space motion and accurate path tracking for robotic welding of curved surfaces.