Constraint-aware multi objective ant colony optimization for precision grazing route planning

Overgrazing and uneven spatial use of forage continue to constrain pasture management in the grasslands of northern China. In the context of a family run pasture, livestock repeatedly graze easily accessible areas, while high quality forage in more distant areas remains underutilized. This makes grazing route planning a practical need for improving forage use distribution. To address this issue, we propose a multi objective Improved Grazing Ant Colony Optimizer (IGACO) for constraint aware grazing route planning. IGACO extends classical Ant Colony Optimization (ACO) with grazing specific improvements, including progress based loop guidance, herd aware obstacle avoidance, a nonlinear walking efficiency index and a multi objective pheromone update rule. The algorithm combines UAV multispectral imagery with GIS data to build a high resolution, constraint annotated pasture graph, in which forage patches, watering points and danger points are encoded as route planning elements. In experiments on a representative pasture in northern China, IGACO achieved the lowest composite objective value and shortest mean path length among ACO, GA and PSO under AHP derived weights. Additional 5 x 6 and 7 x 7 grid-based scenarios showed that the method can also support coarse grained rotational grazing unit sequencing. Ablation analysis further confirmed the contribution of the proposed grazing specific components. These results suggest that IGACO can provide a practical spatial decision support tool for precision grazing route planning.