This paper proposes a flying humanoid robot with a bi-rotor flight unit. While walking is a common method of movement for humanoid robots, it is not always sufficient for certain tasks. To enhance mobility, we apply aerial robotics to create a flying humanoid robot that can walk and fly. We focus on using bi-rotor configuration for the flight unit due to its capability to change the position of the center of gravity. We describe the modeling and control of the bi-rotor and present a method for generating takeoff poses. Additionally, we demonstrate the hardware implementations of the bi-rotor flight unit, humanoid robot, and entire embedded software system. Finally, we perform several experiments to verify the feasibility of flight control, extended mobility, and aerial manipulation and transportation.

We have previously developed a rope tether-climbing robot that is capable of spiral climbing. The climber controls its posture by tilting two rollers against the tether. However, we encountered the problem that the climber rotated around the rope. We attempted to control its posture by constructing a control moment gyro mechanism and a reaction wheel mechanism, but it became clear that angular momentum saturation was inevitable in movements over long time periods. Therefore, to achieve attitude control without angular momentum saturation, we developed a climber robot with a two-roller spiral propulsion mechanism equipped with a derailment prevention mechanism that we developed. We also built an optimal synchronous speed control system for the two rollers. Then, for optimal control, we modeled the frictional forces generated by the rollers, the rolling motion of the rollers, and the ascending motion of the climber. Then, using this model, we constructed an optimal control system for the position and posture of the climber. Climbing experiments verified that attitude control without angular momentum saturation can be realized.

We have been intending to realize a quadruped robot controller autonomously generating rhythm and gait while utilizing natural body dynamics under the gravity. For this purpose, we newly proposed a model of the rhythm generator mainly consisting of sensorimotor functions in the spinal cord, and simulated the walking–running transition of a low spinal cat with hindlimbs according to increased belt speed on a treadmill (Forssberg et al. 1980). In this simulation, we constructed a physical model of the hind-legged biped robot with a trunk and fixtures, and used an independent leg controller with such rhythm generator for each leg. When we employed hip flexion/extension and leg unloading as sensor information for the stance–to–swing phase transition of the rhythm generator, the rhythm in the steady walking could be generated mainly based on leg unloading. In the transient state under the perturbation of belt speed change and the steady running, different rhythms could be autonomously generated based on the combination of leg unloading and hip flexion/extension. Moreover, the walking–running transition could emerge by the change of the dynamics structure triggered by the increase of belt speed.

In contact-rich tasks, robots can correct positional errors by generating compliant motion. Methods using compliance control have the advantage of easily adjusting parameters. However, employing an asymmetric matrix as the stiffness matrix extends design freedom while posing a challenge in demonstrating passivity. This study includes a stability analysis in 3-dimensional space based on the root locus method and proposes a parameter design method based on the analysis. The asymmetric part generates a curl force field that may cause instability. Nonetheless, by designing the asymmetric part such that the eigenvalues of the stiffness matrix exclude imaginary components, vibration resulting from the rotational force field can be avoided.

To support Synecoculture^TM, in which a variety of crops are grown in mixed densely vegetated areas, I developed a seed planting mechanism that can handle seed dumplings made of soil and can be used for multiple species and interchangeable operation. I also developed a manufacturing machine and succeeded in planting seed dumplings that automatically compressed and formed.

Mobility analysis of wheeled mobile robots is essential for the design, development, and control of the robot. While many types of wheeled mobile robots have been developed, having different numbers/configurations of wheels, including driving and steering axles, there are few studies that mathematically analyze and correlate the wheel configuration with robotic mobility. In this paper, we propose a maneuverability ellipsoid, an extension of the manipulability ellipsoid for robotic arms, to quantitatively represent the translational and rotational mobility of the wheeled robot. We applied this analytical approach to three different vehicle models and confirmed that the maneuverability ellipsoid systematically represents the mobility performance depending on the wheel configurations and vehicle dimensions.

In this paper, we attempted to improve object recognition from the images with occlusion by image completion. To interpolate the hidden regions of objects from the surrounding data, this paper proposed a U-Net type image generation network, in which Transformer was used in Generator. In this model, U-Net composed of Transformer encodes the contextual relationships between the pixels, and the additional Transformer generate the interpolated image. The subsequent model determines if the generated image is appropriate for the input. The effectiveness of the proposed method was confirmed by image interpolation experiments for several occlusion images.