This paper proposes a novel object handling device for a robot, in which a conventional vacuum suction cup is equipped with a force sensing ability by coating its surface with a conductive thin-film polymer, poly (3,4-ethylenedioxythiophene) (PEDOT). It can detect the applied force by measuring the resistance change of the coated conductive thin film. The feature of our study is that the suction cup can work, not only as a grasping tool, but also as a force sensor. PEDOT thin film was prepared using the dip-coating method, which can coat thin film directly on a three dimensional surface. We fabricated the sensing suction cup and evaluated its sensing ability by measuring the signal-to-noise ratio, sensitivity, reliability, and frequency response. The resistance changed linearly with respect to the applied force, showing the device's applicability as a force sensor for a vacuum-type robotic hand. Grasping tests for an inclined object, a misaligned object, and an undulated surface were conducted using two or four sensing suction cups attached to the end of the robot. All suction cups showed positive results and the robot successfully picked up the object in each scenario.

A novel capacitive tactile sensor using a polyurethane thin film is proposed in this paper. In previous studies, capacitive tactile sensors generally had an air gap between two electrodes in order to enhance the sensitivity. In this study, there is only polyurethane thin film and no air gap between the electrodes. The sensitivity of this sensor is higher than the previous capacitive tactile sensors because the polyurethane is a fairly flexible elastomer and the film is very thin (about 1 µm). The polyurethane film is formed by spin-coating and etched back from 6 µm to 1 µm using 48% sulfuric acid. As a result of evaluation, the sensitivity of the developed sensor (diameter is 1 mm) is 1.3 pF/Pa (800 pF/N considering the sensing area). Young’s modulus of the thin polyurethane film was estimated to be 20 kPa.

We propose a concept of a flexible sensor array using a novel capacitive force sensor not having a vulnerable electrode on the force applied site. It has a polymer domed structure inside which silicone oil is contained. When the force is applied, the oil is pushed into the surrounding thin channels, where the change in capacitance due to the inflowing dielectric oil is measured between two electrodes on the top and bottom surfaces of the channel. Since the channel does not have a directly applied external force to it, the electrodes do not suffer from damage problems. The change in capacitance was simulated using a simplified flow model. The first trial device of the sensing element has been fabricated. A sensitivity of 0.05 pF/gf was achieved.

Object detectors using deep learning are currently used in various situations, including robot demonstration experiments, owing to their high accuracy. However, there are some problems in the creation of training data, such as the fact that a lot of labor is required for human annotations, and the method of providing training data needs to be carefully considered because the recognition accuracy decreases due to environmental changes such as lighting. In the Nakanoshima Challenge, an autonomous mobile robot competition, it is challenging to detect three types of garbage with red labels. In this study, we developed a garbage detector by semi-automating the annotation process through detection of labels using colors and by preparing training data by changing the lighting conditions in three ways depending on the brightness. We evaluated the recognition accuracy on the university campus and addressed the challenge of using the discriminator in the competition. In this paper, we report these results.

Autonomous mobile robots need to acquire surrounding environmental information based on which they perform their self-localizations. Current autonomous mobile robots often use point cloud data acquired by laser range finders (LRFs) instead of image data. In the virtual robot autonomous traveling tests we have conducted in this study, we have evaluated the robot’s self-localization performance on Normal Distributions Transform (NDT) scan matching. This was achieved using 2D and 3D point cloud data to assess whether they perform better self-localizations in case of using 3D or 2D point cloud data.