Safe and Efficient Robot Navigation via Costmap-Guided Multi-Objective A*

Autonomous mobile robots must balance path efficiency and safety when navigating cluttered environments. This paper proposes Costmap-MOA*, a multi-objective A* planner that couples a continuous obstacle-aware costmap with local Pareto front maintenance. The costmap fuses terrain relief (obstacle distance) and goal-directed turbulence to model both safety and directional guidance. The Costmap-MOA* algorithm simultaneously optimizes path length and accumulated safety cost, extending classical A* by maintaining non-dominated solutions at each node via lexicographic ordering of distance and risk. This design preserves shortest-path optimality while systematically exploring distance–safety trade-offs and avoiding redundant trajectories. Experiments on sparse and cluttered 2D environments demonstrate that Costmap-MOA* achieves 2–4× more non-dominated solutions than NAMOA*, improves best-safety cost by 25–40 %, and maintains comparable computation times. The method provides decision-makers with a diverse set of feasible trajectory alternatives, making it a practical tool for safety-critical autonomous navigation.