Lower limb Motor Imagery (MI) of a sit-stand motion using electroencephalography (EEG) has been studied with the aim of rehabilitation. We aim to explore the use of self-perspective stimulus using virtual reality head-mounted displays (VR-HMD), which has promising results in upper limb MI tasks but is yet to be explored in sit-stand. We compare VR-HMD and on-screen stimulus using an action observation (AO), MI, and motor execution (ME) paradigm with EEG and electromyography (EMG) recording. We examined features including time-frequency analysis of event-related synchronization/desynchronization (ERS/D), corticomuscular coherence, and decoding with Common spatial pattern. Results from time-frequency analysis show a significant difference in ERS/D in alpha and beta. Coherence analysis shows higher functional connection between the lower limb sensorimotor cortex region and the tibialis anterior muscle in AO tasks for VR-HMD. The classification performance from VR-HMD sessions is better in most subjects, though not significantly. The before-after effect was examined and found to be not affected. Results indicate that VR-HMD stimulus for the sit-stand task is a feasible MI paradigm with some benefits over a screen-based method.

Variable digital FIR filters (VDFs) are used to reduce noise in automatic weighing instruments. VDFs can reduce delay without degrading measurement accuracy by piecewise high attenuation in the stopband. However, conventional VDFs were difficult to obtain the desired piecewise high attenuation in order to change the attenuation by weight function. In this paper, we propose a design method for equiripple linear-phase variable digital FIR filter having the specified piecewise high attenuation in the stopband. The proposed VDF has high attenuation in part of the stopband, and its attenuation, position, and stopband edge frequency are variable. The VDF's filter coefficients are realized by polynomial using variable parameters. Therefore, when multiple elements are varied, the polynomial orders increase. The proposed method reduces the degree of the polynomial by taking advantage of the fact that the higher order terms of the polynomial have little effect on the filter coefficients. The proposed method adds and a constraint in the minimax design problem, that directly changes piecewise high attenuation band by a variable parameter. Effectiveness of the proposed method is illustrated by design examples.

This paper proposes a new event-triggered robust control system for robot manipulators, based on two event-triggered mechanisms (ETMs)，which lead to intermittent communication not only for the sensor-to-controller channel but also for the controller-to-actuator channel. The proposed control system enjoys the advantage of requiring fewer resources of data communication and less signal updating of the control actuators. The control performance of the proposed control system is analyzed based on the Lyapunov approach. Furthermore, the absence of the Zeno behavior in the triggering sequence is proved rigorously. Finally, experiments are carried out on a two-link robot manipulator system to support the theoretical results.

This study deals with the control problems for automating the operation of an excavator loading soil onto the back of a dump truck. In the loading operation, the bucket should not touch the dump truck and should spill as little soil in the bucket as possible. We have been studying how to apply Model Predictive Control (MPC) to this problem to achieve ideal loading operation. When trying to achieve the desired operation using MPC, it is extremely important to tune the weights of the objective function appropriately. However, since this control problem may depend on the situations, that is, initial posture of the excavator and the position of the truck, optimization for specific conditions would not be desirable. Therefore, we constructed a method to generate suitable weight parameters according to the loading situation using reinforcement learning. The effectiveness of the proposed method was verified by numerical simulations.

In multi-task learning, the goal is to improve the generalization performance of the model by exploiting the information shared across tasks. In this paper, we propose a neural network that simultaneously learns depth estimation and semantic segmentation of the environment from omnidirectional images captured by an omnidirectional camera. Our proposed neural network is developed by modifying UniFuse network, which was originally developed for depth estimation from omnidirectional images, to simultaneously learn depth estimation and semantic segmentation of the environment by exploiting the features shared between depth estimation and semantic segmentation tasks. In the experiments, the proposed method was evaluated with the well-known Stanford 2D3D Dataset. High accuracy for the two tasks was not obtained with a single network. However, if either of the two tasks was prioritized in learning, the synergistic effect of the two tasks with shared feature maps would improve accuracy, resulting in better results than a single-task network. It showed the effectiveness of simultaneously learning semantic segmentation and depth estimation from omnidirectional images.

In the product design, it is important deeply understanding the profiles of target user. For deriving the user profiles, it is required to collect, organize, and analyze data on the target users. However, there are some users from whom it is difficult to directly obtain qualitative data in the initial stage of product design, due to factors such as age. Therefore, for such target users, we have devised a spiral model development method combining human-centred design by Personas and the verification of product candidate by users, starting with Assumption Personas. In this study, we applied this methodology to the development of a "Digital Learning Aid for Writing" for infants and younger children from whom it is difficult to obtain qualitative data directly in the initial stage of product design, and studied the effectiveness of this methodology.

A meat quality evaluation method based on the electrical characteristics has been proposed. In this paper, we developed a measurement system for the electromagnetic response of meat during the maturation process and evaluated its basic performance. Experiments using beef and pork samples demonstrated that the developed system could stably measure the bioimpedance of the meat over a period of 14 days.