In order to solve the problems of high induced drag ratio during low Reynolds number flight, poor flight adaptability of conventional wingtip drag-reduction devices at multi-state, and insufficient aerodynamic performance at high angles of attack, a bionic blended-wing-body UAV was designed taking the slotted primary feather structure of large soaring birds such as goshawks and golden eagles as the bionic prototype. A new bionic wing grid layout and parameter design method which could significant enhance lift and reduce drag was proposed according to the progressive stepwise optimization method. Based on the N-S equations and k-ω SST turbulence model, the effects of the number of grid winglets, aspect ratio, spatial arrangement, geometric shape and camber on the lift-drag characteristics were calculated. The flow field influence mechanism of the bionic wing grid was investigated by combining pressure nephograms, velocity streamlines and vorticity distributions. The research results show that the proposed bionic wing grid design scheme can achieve significant lift enhancement and drag reduction effects simultaneously in the angle of attack range of 2~16°. Compared with the baseline model, the lift coefficient at cruise angle of attack is increased by 5.08%, the drag coefficient is decreased by 5.87%, the lift-to-drag ratio is increased by 11.64%, and the average lift-to-drag ratio is increased by 10.78%. The bionic wing grid can decompose the concentrated large-scale wingtip vortex into multiple discrete weak vortex cores, thereby weakening the wingtip vortex intensity, reducing induced drag and delaying flow separation at high angles of attack. After the wing grid-wing integration optimization, the drag coefficient under cruise condition is further reduced by 1.39%, the lift-to-drag ratio is increased by another 1.78%, and the flow field continuity at the wingtip junction is improved. The research results can provide theoretical support and engineering reference for the optimization design of wingtip drag-reduction devices for small low-Reynolds-number UAVs.