Multi-Constraint Enhanced DWA for Robust and Smooth Local Navigation

Abstract

To address the issues of path discontinuity, control instability, and insufficient global guidance in the traditional D ynamic W indow A pproach (DWA) for local path planning, this paper proposes a M ulti-C onstraint D ynamic W indow A pproach (MC-DWA). The proposed m ethod incorporates multiple constraints in trajectory evaluation, including heading deviation, path adherence, curvature variation, and velocity jerk, thereby im proving the smoothness and stability o f the planned paths. Extensive sim ulation experim ents dem onstrate th at M C-DW A achieves b etter path quality and safety than existing methods, exhibiting enhanced adaptability and robustness.