Flying insects perform active flight control with flapping wings by continuously adjusting their wing kinematics in stabilizing the body posture to stay aloft under complex natural environment. While the Proportional Derivative (PD) / Proportional Integral Derivative (PID)-based algorithms have been applied to examine specific single degree of freedom (DoF) and/or 3 DoF flight control associated with insect flights, a full 6 DoF flight control strategy remains yet poorly studied. Here we propose a novel 6 DoF PD controller specified for flight stabilization in flapping flights, in which proportional and derivative gains are optimized to facilitate a fast while precise flight control by combing Laplace transformation and root locus method. The vertical position, yaw, pitch and roll are directly stabilized by tuning the wing kinematics while the forward/backward position and lateral position are indirectly stabilized by controlling the pitch and roll, respectively. Coupled with a recently developed flight dynamic model informed by high-fidelity CFD simulation (Cai et al. 2021), this methodology is proven to be effective as a versatile and efficient tool to achieve fast flight stabilization under both small and large perturbations for bumblebee hovering. The 6 DoF PD flight control strategy proposed may provide a useful bioinspired flight-controller design for flapping-wing micro air vehicles (FWMAVs).
As visual stimuli for exercise and cognitive rehabilitation of biomechanics, virtual reality (VR) and augmented reality (AR) devices have getting popularity. In the process of developing the relevant content for VR and AR, there has been a problem that only a specific platform must be supported or multiple programs must be used. Recently, the Unity 3D platform has been developed for the convenience of game development for VR or AR environments that potentially solve these problems. Unity 3D’s game engine and animation can easily implement a moving avatar as visual stimuli, and the speed of the avatars can be checked in real-time. Therefore, we developed a moving avatar as the visual stimuli using Unity 3D and conducted pilot experiments with four healthy subjects by performing the knee extension and ankle dorsiflexion tasks with and without visual stimuli. The number of movements was counted to check the feasibility of the effectiveness using visual stimuli using Unity 3D. The results showed that the number of movements was higher when the visual stimuli were presented compared to that without the visual stimuli in both ankle dorsiflexion and knee extension. Our findings and approach can be a basis for further developing rehabilitation training protocols using various visual stimuli with Unity 3D.
Carbon fiber running-specific prostheses (RSPs) are widely used among lower-limb amputee runners. However, which prosthesis provides the best performance for runners remains unknown. For this purpose, a computational model of the human body with a prosthesis was created and the effect of the prosthetic parameters on performance was investigated. First, motion capture systems were used to collect motion data from amputees. Furthermore, marker and force plate data were obtained to create a digital human model. Kinematic data such as limb lengths and joint angles were calculated using marker data. Afterward, the inertial properties were estimated to conduct inverse dynamic analyses. After building a computational model of amputee sprinting, the joint positions and ground reaction forces (GRFs) were compared with the experimental results. The design parameters of the prosthesis were introduced to understand the effects of the prosthesis on motion and performance. The response surface method was used to express motion adaption regarding the geometry and stiffness of the prosthesis. Hip and knee sagittal joint angles were updated based on the response surface method to simulate joint motion adaptations of the worn prosthesis. Additionally, average horizontal velocity, horizontal velocity change over one gait cycle, vertical and horizontal impulses were considered as performance functions. An evaluation parameter was proposed to generalize the idea of performance. The moment of the prosthetic knee and the closest point of the prosthesis to the ground during the swing phase were defined as design constraints to consider knee buckling and prosthetic leg tripping, respectively. The effect of the design parameters on the performance and constraint functions was also investigated and a method to determine and design a suitable prosthesis for an individual was proposed. It was revealed that proper selection and design of prostheses represent an important way to increase performance.
Several training methods have been developed to obtain motion information during real-time walking and feed it back to trainees who adjust their gait to ensure that the measured gait parameters approach target value, which may not always be suitable for every trainee owing to physical differences between individuals. This paper proposes a method of setting this target value considering these physical differences and discusses the usefulness of the gait training method, wherein a multichannel deep convolutional neural network (MC-DCNN) gait classification model constructed by learning ideal or non-ideal gait features beforehand is used for trainee gait classification. Activation maximization is applied to the MC-DCNN model; data wherein the ideal walking features are activated are generated based on trainee gait data. However, the amounts of features to be activated to generate a possible and natural gait are restricted. The original trainee gait, beyond individual physical differences, and gait data generated based on the original gait data seem to yield the target value considering the physical differences among individuals. This study focused on gait related to stumbling. To verify its usefulness, a multivariate gait dataset consisting of kinematic and kinetic indices labeled as “gait rarely associated with stumbling” or “gait frequently associated with stumbling” was divided into a training set, validation set, and test set. The MC-DCNN model learned gait features for multivariate gait data classification in the training set. It classified the gait with 96.04±0.12% accuracy against the validation set. Finally, by applying the proposed method to the multivariate gait data contained in the test set, we generated multivariate gait data classified as “gait rarely associated with stumbling” based on the input data. In addition, the generated multivariate gait data include motion that increases the thumb-to-ground distance and describe possible and natural gait considering the physical differences among individuals.