Although animal-assisted activity is attracting attention as one of mental health measures, it has many problems in its implementation. We compared stress relief effects in short-term interaction of robot-assisted activity and stuffed animal-assisted activity, which are complementing activities for animal-assisted activity. We confirmed that psychological stress relief effects, especially reducing tension, depression, and confusion and improving vigor, obtained by robot-assisted activity are significantly higher than that obtained by stuffed animal-assisted activity. On the other hand, no significant difference could be confirmed in physiological stress relief effects. Although negative feelings of unsatisfactory due to lack of reactions decreased psychological stress relief effects, some kind of interaction caused by softness were suppressed decreasing physiological stress relief effects combination with physiological stress relief effects by contact with soft things. These are assumed to be factors that results of psychological and physiological evaluations are different.

This paper presents a hybrid position/force control for environment by using a nonlinear disturbance observer. In general, force control is used with a force sensor. However, force sensors are relatively expensive and easily broken. Moreover, their signal is noisy, and limits system response. Disturbance observers are an attractive approach to improving control performance without additional measurement devices in robot controls. We have been proposed a nonlinear disturbance observer for robot manipulator. Experiments of a hybrid position/force control with the nonlinear disturbance observer are presented to show the effectiveness of the proposed disturbance observer.

This paper addresses a position correction using an original absolute difference of difference (ADD) correlation that is robust to texture degradation such as occlusion. The correlation is calculated by summing absolute differences of the difference between two successive intensities in a template image. The quality of the position correction is guaranteed by cross-checking the search results of four image groups. Using the proposed method, our experimental vehicle equipped with a downward facing camera is most successful in correcting position error compared with other correlation methods including zero-mean normalized cross-correlation (ZNCC) on an indoor floor with the texture 67% occluded. For outdoor environment on a paved road, the proposed correlation is more robust against texture degradation by sand than conventional sum of absolute differences (SAD) and increment sign correlation (ISC) calculations, and moreover, the processing time for ADD is practically short to search every template image recorded for the position correction.

Tactile sensors typically have elastic covers for reducing the impact of objects that make contact with the sensor. However, elastic covers decrease the spatial resolution of the sensor because of their low-pass spatial filtering characteristics, which vary based on the stiffness of each cover. Therefore, we propose a tactile sensor that uses shape-memory polymer (SMP) with stiffness that varies based on temperature. SMPs can be deformed above their glass transition temperature (T_g) upon application of a small load. The reversible change in the elastic modulus between the glassy and rubbery states of an SMP can be on the order of several hundred-fold. These characteristics can be exploited to alter the spatial resolution and the rigidity of the sensor surface by adjusting the temperature of the SMP. In this study, a prototype sensor was fabricated using an SMP sheet with embedded electrical heating wires, and its fundamental performance was evaluated. The experimental results of a preliminary proof-of-concept investigation conducted using the prototype confirm the feasibility of the proposed sensor. Moreover, using a 4×4 array of pressure sensors, it was possible to distinguish the shape of the object placed on the sensor above and below T_g. The center of pressure applied on the prototype sensor could be measured by a developed program. In addition, we evaluated the control of the position of an object on the robot arm utilizing the prototype sensor attached on the arm. We could hold the constant position of the object on the sensor, although it took a longer time to stabilize the object above T_g because of the deformation of the SMP.

In order to realize automated picking robot, it is an important task to determine the grasping parameters (position/direction/angle) of the object. In this paper, we propose a method for approximating an object with primitive shape to determine the grasping parameters. Our method applies “object primitive” (for example, hexahedrons, cylinders, and spheres) to the object by using a 3D-deep neural network (DNN) on the surface of the object. Then, we estimate the grasping parameters based on preset grasping rules. The success rate of approximating the object primitive with our method was 94.7%. This result is 6.7% higher than the 3D ShapeNets method using 3D-DNN. Also, as an experimental result of grasping simulation using Gazebo, the success rate of grasping with our method was 85.6%. This result is 17.8% higher than the GPD method using DNN.

Measuring the distance of a specific object accurately is a challenging task. Recently, in mobile robot community, there are several works combining an image-based object detection method and the distance measurement using laser scanner to achieve such challenge. However, the performance of existing methods tend to degrade due to the influence of occlusion and walls behind the target object. To tackle the vulnerability in the existing methods, in this paper, we propose Belief Weighted Max Cluster Average Detection (BWMCAD) method which utilizes the result of 3D Faster R-CNN as the belief of laser scan data and clusters the belief weighted data to extract the distance to the target object. We demonstrate its effectiveness through distance measurement tasks using Tsukuba challenge 2017 data and in-house data in comparison with existing methods.

This paper proposes an adaptive walking method for a six-legged robot controlled by a Follow-the-Contact-Point gait control. In the previous research, the range of allowable duration time at each control mode was derived mathematically by using formal verification, and, by mapping the allowable duration time to the configuration space of a leg, the allowable space for each leg to contact was derived. However, because of simplification in the mapping, the improvement of movability of a robot was limited. To overcome the issue, Timekeeper control that directly keeps staying time at each control mode within allowable range is proposed. First, a control method of six-legged robot, Follow-the-Contact-Point gait control, is introduced. Next, the allowable duration time at each control mode which is derived in the previous research and the proposed Timekeeper control is explained. Additionally, an operating method of the robot by indicating contact point and a posture control of the robot are proposed. Finally, improvement of robot's movability using the proposed method is verified in a physical simulation.

Robots are expected to substitute for humans for work performed in locations at a height, such as the inspection of an airplane surface. The authors propose a traveling-wave-type wall-climbing robot simulating a snail direct wave movement. In this robot, several adsorption units are respectively connected by a universal joint, and the robot progresses by the propagation of expansion and contraction between the units. At this time, the moving unit slides on the wall surface, and the unit that does not move is fixed with a strong force, such that each unit switches the frictional force. Consequently, the robot has a wide ground contact surface, thus maintaining a high adsorption force and stable traveling. Furthermore, since it can be bent by a universal joint, it can cross a curved wall surface. To this end, in this study, we developed a negative pressure adsorption type traveling - wave wall - climbing robot, run it in an environment simulating an aircraft about the robot, and evaluated its characteristics.

In this paper, we propose a control method which combine collision avoidance method and RISE (Robust Integral of the Sign of Error) for multiple Quad-rotor formation control methods, and confirm effectiveness of this method that is resistant to disturbance by lowering the risk of collision. First, we modeled quad-rotor as a linear system. Next, we described a method to suppress nonlinear disturbance using RISE which is a type of sliding mode control. In addition, we introduced collision avoidance method, and derived the conditions to achieve accurate formation by using Lyapunov stability theory. Finally, we implemented an algorithm in the simulation, and confirmed that the proposed method is functioning correctly.

Adjustability for grasping force (AGF), which is one of motor function of fingers, is an ability to grasp an object with appropriate force. Since a person with a disable hand can not adjust grasping force, he/she grasps an object with extra force or can not grasp it due to lack of grasping force. In our previous study, we developed a training and testing device of the AGF, which is called iWakka. In this paper the usefulness of training and testing with iWakka was confirmed by applying CI+i therapy to ten hemiplegic patients after stroke. The AGF of all patients was improved after training with CI+i therapy. CI+i therapy was developed by the fourth and fifth authors by adding the training task with iWakka to CI (Constraint Induced Movement) therapy.