We propose the generation method of large-scale image database for pre-training in 3D object recognition. The method is inspired from the principles of nature law. We adopt fractal geometry to represent the principles and build the Fractal Data Base random search (FractalDBrs). In contrast to traditional image database such as ImageNet, Iterated Function System (IFS) automatically generates large amount of image data to build the proposed FractalDBrs in short time without menial labors such as collecting and annotating images. In the experiments, we utilized the FractalDBrs and traditional databases; ImageNet, CIFAR100, Caltech256, or Places365, for pre-training in 3D object recognition with ModelNet40. The model pre-trained with FractalDBrs achieved the highest discrimination accuracy of 97.12% against the second highest accuracy of 96.43% with ImageNet. For reference, the model trained from scratch achieved 91.53% discrimination accuracy. We have verified the effectiveness of the proposed fractal geometry-based image database for pre-training in 3D object recognition.

Master-slave teleoperation robots are useful for tasks that require human judgments and skills. Haptic sensation can be transmitted between a master and a slave by implementing bilateral control. Besides, visual information of the slave side that cannot be directly seen from the master side is presented by devices such as cameras, endoscopes, and displays. Therefore, advanced support of the teleoperation robots for operators can be expected by utilizing extended visual and haptic information. In this paper, a master-slave teleoperation system with ability to extend information transmitted to operators is proposed. The teleoperation system consists of one degree of freedom master-slave system implementing bilateral control based on robust acceleration control and an image processing system. To support operators, visual and haptic augmented reality is generated by image processing, micro-macro bilateral control, and compliance control. In addition, a design method of position and force scaling for micro-macro bilateral control according to magnification ratio of image is presented. The utility of the proposed bilateral teleoperation system is verified by experiments.

Elastic modulus of the grinding wheel is closely related to the configuration to three elements of grinding wheel, and it is an important factor that affects the grinding performance. This study proposes a method for deriving elastic modulus from the surface wave velocity propagating on the abrasive layer of a super abrasive wheel. As a result of experiment, surface waves propagation velocity changes depending on the specifications of the abrasive layer. And, surface waves propagation velocity is possible to evaluate the quality of the abrasive layer. Moreover, although there are some problems left unsolved, I found the possibility to derive the elastic modulus of the abrasive layer.