Multi- constraint enhanced dynamic window approach for robust and smooth local navigation

Abstract

The limitations of the traditional dynamic window approach in local path planning, including poor trajec tory smoothness, unstable control performance and insufficient global navigation guidance, are examined. An improved multi-constraint dynamic window approach is proposed as a solution. This method enhances the trajectory evaluation by introducing several refined constraint metrics: dual-angle heading deviation, global path adherence, curvature va ria - tion control and jerk suppression. They are integrated into a unified multi-objective scoring function that promotes smoo ther, more reasonable and dynamically feasible trajectory generation. To verify the effectiveness of the propo- sed approach, a series of experiments is conducted across various representative map scenarios, including both standard and complex environments. The enhanced algorithm is compared with the classical dynamic window approach and several of its improved variants. It is established that the proposed approach achieves superior performance in terms of path quality, safety and overall planning robustness, demonstrating improved adaptability and reliability in solving complex navigation tasks.