中田 敏是

Abstract Bio-hybrid drones, which combine biological odor sensors with small drones, introduce an innovative navigation method that compensates for traditional image-based navigation, enhancing the capabilities of aerial robots. Inspired by the odor-source search behavior observed in biological organisms, we identified two key elements for improving odor source direction estimation accuracy for bio-hybrid drones: (1) increasing the anisotropy of the odor sensor using a sensor enclosure, and (2) implementing a stepped rotation algorithm that strategically incorporates pauses during scanning. This integration resulted in a doubling of both search accuracy and range, achieving a search distance of up to 5 m, significantly exceeding the performance of a previous algorithm that sequentially combined rotational and linear motions. Although these elements are commonly observed in various arthropods, they are underapplied in robotics applications. This study provides a novel perspective to robotic olfactory navigation techniques by leveraging these biological behaviors to enhance robotic functionality.

The robust exoskeleton of beetles, Coleoptera, is believed to have protective advantages, contributing to their evolutionary success. However, empirical evidence of the defensive capabilities of this exoskeleton remains surprisingly scarce. Here, we demonstrate the critical role of the robust beetle exoskeleton in protecting against avian predation. We found that flower chafers (Scarabaeidae, Cetoniinae) have more robust bodies than other scarab species. Laboratory experiments with naïve Japanese quail, Coturnix japonica, revealed that some individuals of intact Cetoniinae survived attacks without serious injury, whereas all individuals of soft scarab species or elytra-removed Cetoniinae were consumed. The survival rate of intact Cetoniinae increased in complex environments because the combination of their stiffness and elliptical shape made it difficult for quail to handle the prey. Field experiments with wild white-cheeked starlings, Sturnus cineraceus, and Eurasian tree sparrows, Passer montanus, demonstrated that most individuals of Cetoniinae species were ignored, whereas soft species were readily preyed upon. Further, when we presented the starlings with the Cetoniinae species Protaetia orientalis that had artificially softened bodies and altered appearance, the starlings readily preyed upon them. This observation suggests that P. orientalis is not chemically defended. Moreover, wild birds can visually discriminate hard species because of prior experience with the unprofitable prey. These results collectively provide evidence that the robust exoskeleton of beetles protects them from predatory attack.

In recent years, the application of bio-inspired structures has garnered attention for enhancing the performance of fluid machinery. In this study, we experimentally investigated the effects of introducing a bio-inspired cutout structure to the propellers of drones, aiming to improve thrust efficiency and reduce noise levels. Our results demonstrated reductions in noise levels compared to conventional propellers. Parametric studies revealed that the roundness of the structure significantly influenced both flight efficiency and noise levels, suggesting its importance for replicating the inherent fluid characteristics found in nature. Additionally, optimal parameters for noise reduction, such as the length of the cutout, angle of incision relative to the flow direction, and the distance between the gap were identified. Although no improvements in flight efficiency were observed, most of the models investigated exhibited only around a 5% reduction in efficiency compared to the standard propellers, suggesting practical applicability for scenarios such as nighttime drone operations in urban areas. The noteworthy reduction in sound pressure levels in the mid- to high-frequency range achieved by the bio-inspired propellers in this study holds the potential to address the issue of drone noise pollution and encourage drone operations in urban areas. Moreover, the confirmed decrease in sound pressure at specific frequencies and the suggested controllability hint at the possibility of enhancing sound source localization performance using drones.

Flying animals such as insects and birds use wing flapping for flight, occasionally pausing wing motion and transitioning into gliding to conserve energy for propulsion and achieve high flying efficiency. In this study, we have investigated the gliding performance of a gliding model based on a flapping-wing robot developed in a previous study, with the aim of developing a highly efficient flying robot that utilizes bio-inspired intermittent flight. Wind tunnel experiments with a gliding model have shown that the attitude of the wings has a strong influence on gliding performance and that a tail is effective in improving gliding performance. The results of this study provide important insights into the development of flying robots that can travel long distances with high efficiency.

Odours used by insects for foraging and mating are carried by the air. Insects induce airflows around them by flapping their wings, and the distribution of these airflows may strongly influence odour source localisation. The flightless silkworm moth, Bombyx mori, has been a prominent insect model for olfactory research. However, although there have been numerous studies on antenna morphology and its fluid dynamics, neurophysiology, and localisation algorithms, the airflow manipulation of the B. mori by fanning has not been thoroughly investigated. In this study, we performed computational fluid dynamics (CFD) analyses of flapping B. mori to analyse this mechanism in depth. A three-dimensional simulation using reconstructed wing kinematics was used to investigate the effects of B. mori fanning on locomotion and pheromone capture. The fanning of the B. mori was found to generate an aerodynamic force on the scale of its weight through an aerodynamic mechanism similar to that of flying insects. Our simulations further indicate that the B. mori guides particles from its anterior direction within the ~ 60° horizontally by wing fanning. Hence, if it detects pheromones during fanning, the pheromone can be concluded to originate from the direction the head is pointing. The anisotropy in the sampling volume enables the B. mori to orient to the pheromone plume direction. These results provide new insights into insect behaviour and offer design guidelines for robots for odour source localisation.