A tolerance constrained G2 continuous path smoothing and interpolation method for industrial SCARA robots

Abstract Piecewise-linear robot paths lack G1 and G2 continuity. The discontinuity results in feedrate reduction and fluctuation during real-time execution. State-of-art linear path smoothing methods have issues such as lack of data or chord tolerance constraints, numerical instability, and lack of synchronized position and orientation parameterization. This study proposes a robot path smoothing method for SCARA (Selective Compliance Assembly Robot Arm) robots. The smoothing method simultaneously satisfies requirements of G2 continuity, tolerance constraints (target point interpolation and chord tolerance constrained), shape-preserving (free of self-intersection), unified parameterization, and real-time performance. A global geometric iteration method combined with local corner transition scheme is used to generate smooth path. A convex combination presentation is proposed to simplify the formulation, which also makes it possible to unify various smoothing schemes into one extensible framework. Simulations with data analysis and actual experiments are performed for the verification of quality and efficiency. Comparisons with other solutions are included to evaluate the advantages of the proposed robot path smoothing method.