We propose a flexible robot gripper with a simple structure that controls grasping by magnetic force. This gripper is constructed with an electromagnet, a permanent magnet, an elastic membrane, and water. Since it is a simple structure without driving elements, this gripper is easy to be downsized. In this paper, we show the structure of the gripper and the holding and releasing principle. In addition, we demonstrate the effectiveness of the proposed gripper through several experimental results by gripping various objects using two sizes of developed grippers attached to an industrial robot.
For construction of overhead power distribution line, it may be difficult to build a rope before wiring power line between utility poles when there are some obstacles between the poles. Multi rotor helicopter (MRH) can fly over the obstacles and carry the rope between the poles. Therefore, it is expected that this work will be facilitated by utilization of MRH. On the other hand, the MRH utilization in the real site causes some risks. If the risk level is unacceptable, the level needs to be reduced to an acceptable range. In this paper, an example of the utilization of MRH in construction of overhead distribution line is introduced. We completed the construction with reducing the risks caused by using MRH to an acceptable range. As the result, the effectiveness and safety of utilization of MRH in the construction are confirmed.
The impacts of robots and ICT in care settings have not been sufficiently evaluated, particularly in relation to workers' workload. The purpose of this paper is to evaluate quantitatively the effects of introducing communicative robots on night shift duties of nursing facility workers. The subjects are five late-night care workers, looking after elderly people to whom communicative robots with an infra-red radiation monitoring system was introduced. We investigated fatigue level using the ‘method for checking subjective symptoms”. Assessments were conducted at the beginning of the night shift (16:30), before dinner (20:00), before nap (00:00), after nap (02:00) and at the end of the night shift (09:50) before and 4 weeks after introduction. The effects of the system on the nighttime work burden and frequency of accidents were compared. During the fourth week, compared with the pre-introduction phase, the total fatigue level was improved before nap (p < 0.05), after nap (p < 0.05), and the end of the night shift (p < 0.001). Along with this improvement, the accidents especially the trauma accidents decreased. The introduction of a communicative robot with a monitoring system significantly improved the total fatigue level of late night nursing care workers and alleviated the nighttime work workload.
We developed a novel robot equipped with 4 crawler units, that can cross over steps as high as 45[cm] (more than 1.5 times the robot's height) and steep slopes more than 45[degrees]. Despite comprising of few driven axes and very simple mechanism, this robot offers high cross-country ability, and is easy to drive by simply remote controlling the right and left speed of the crawlers. The crawler units are independently connected to the body through suspension arms, which are simply spring-loaded at the body side, and have passive swing axis at the crawler side. This configuration gives the robot high terrain adaptability, while effectively generating propulsion and maintaining a stable posture, even though each crawler unit contacts the ground independently at different states. The performance of the 4 crawler robot is validated by many outdoor experiments. Furthermore, this paper also introduces a new type of coupled robot comprising 2(or more) coupled single robots. When 2(or more) single robots are moving separately they can explore very wide area, and if they face the large obstacle which the single robot cannot cross over, they couple together and the coupled robot can greatly extend the capabilities of a single robot unit. Owing to the passively and elastically movable axes of the connecting arm, a coupled robot comprising 2 single robots with the connecting arm can cross over the obstacle as high as 30[cm] although a single robot cannot cross over as high as 20[cm].
Accurate grasping of objects such as industrial parts and everyday necessities is an important task for industrial robots and living-support robots. Many methods have been proposed for grasp point detection for robots, some that utilize machine learning and some that do not. Recently, a grasp point detection method using a 2-stage deep neural network has been proposed. Although the 2-stage deep neural network could detect the grasping point of no-learned objects, the computation cost would be high. In this paper, we propose a method for detecting grasping points using one deep convolutional neural network (DCNN) introducing graspability. Simultaneous detection of grasping points and graspability in one neural network lessens calculation costs. Evaluation experiments confirmed that grasping points could be properly detected using graspability.
Recently, assisting human motion with a wearable robot is studied widely. These assisting systems are proposed from various points of view. Our study has proposed an assisting system that the torque, angle, and the viscoelasticity of the joint are variable because the human muscle has the characteristic of variable viscoelasticity. This system consists of antagonized muscle that the stiffness is variable and magneto-rheological fluid brake that the viscosity is variable. Therefore, the system enables the wearable assistive device such as exoskeletons to be structurally soft and to provide the wearer with a high back-drivability. We have proposed the assisting method by changing the viscoelasticity of the device joint according to the viscoelasticity of wearer's muscle. But suitable viscoelasticity of the joint for the assistive motion is not well known. In this paper, an experiment is conducted to examine how the viscoelasticity of the joint affects to human motion when assisting. Also, based on the results from the experiments, we propose the control method for the joint viscoelasticity of the device that responds to the assistive motion and confirm the effects with the experiments.