In this paper, we propose a work description by elemental motion and a work identification method for creating detailed care records. This method describes tasks as a time-series of elemental motion and identifies care works. In order to achieve robust work identification against irregular motion and annotation of incorrect motion labels during a work, some elemental motion are integrated into one class as complementary event motion. Using the time-series data of elemental motion information, our proposed work identification method utilizing Hidden Semi-Markov Models can recognize performed care works based on the frequency distribution of state transitions. In the experiment, three simulated care works were measured and identified. In addition, simulations using artificial data were conducted to verify the robustness of the proposed method. Furthermore, we measured actual caregiving tasks at a nursing home to confirm whether the task detection was feasible. The results demonstrated the effectiveness of the proposed system.

The paper presents a new SLAM algorithm with two main contributions. Firstly, it proposes a new fast and accurate scan matching algorithm. It uses a multi-layer 3D voxel map and effectively updates features in each voxel, allowing it to choose the best voxel based on resolution and features for the scan matching. Secondly, it introduces a flexible tightly-coupled SLAM framework. This framework does not require a specific sensor setup and automatically finds the best tight coupling. Our experiments demonstrate that our method is capable of real-time processing on Intel and ARM-based processors and is more accurate than current advanced SLAM methods.

This paper proposes a method for grasping a corner of a suspended and transporting towel-like object with a robot arm. In the past, manipulation of flexible objects has been performed when the towel-like object is in a static/quasi-static state or when the deformation of the object is predictable. However, there have been few studies on grasping under the condition that the toweling object is moving randomly. This study aims to enable grasping of a randomly moving towel-like object by introducing a cooperative vision system. As the cooperative vision system, we develop an environmental vision system using a stereo camera and a robot-mounted vision system using a camera and a laser with DOE (Diffractive Optical Element). Also, these vision systems operate cooperatively to recognize a state of the corner of a randomly moving towel-like object in real time. Moreover, we propose a method for tracking and grasping of the corner of the towel-like object using the information obtained by the vision systems. As a result, the grasp execution time in a series of operations was about 70% faster than in the previous study, and the expected value of execution time until success was halved.

It is widely known that the aerodynamic characteristics of a rotor change near the ground. This phenomenon is called ground effect. The ground effect affects flight stability at low altitudes. This study aims to realize a stable flight for rotorcraft near the ground. Toward this goal, we have proposed a ground effect model of a single rotor in hovering near the ground. This paper presents a detailed derivation of the proposed model. The proposed model is the same as Joseph's model in that two vortices represent the rotor. The difference is the process of deriving the representative velocity. The model by Joseph et al. makes assumptions that are not physically or mathematically reasonable. The proposed method derives a ground effect model from the induced velocity distribution induced by the two vortices. The effective blade length physically corresponds to the fact that blade tip losses are considered in the model. We validated the proposed model by measuring the thrust of a single rotor. Furthermore, experimental results show that blade shape affects the strength of the ground effect. The proposed method can represent the strength of the ground effect by using the effective blade length. In the experiment, the ground effect was smaller for blades with smaller blade tips, and this phenomenon corresponds to the proposed model. On the other hand, a phenomenon that may be due to the combined effects of approaching the ground was also evident. Investigation of this phenomenon is a future issue.

Various man-machine interfaces controlled by electromyogram (EMG) signals such as the myoelectric prosthetic hand have been proposed. General classifiers can not consider unintended motions in the training phase and require learning all the motions. In this paper, the authors propose a motion estimation system with unlearned classes and combined motions based on a muscle synergy model. The proposed method can identify unlearned five-finger combined motions by learning a single motion only. Furthermore, this method utilizes the history of muscle synergy and unlearned motion detection, using a state transition model. In the experiments, it was shown that the discrimination accuracy was sufficient for simple combined motions.