In order to represent virtual objects photorealistically in augmented reality (AR), the problem of optical consistency is important. There are several methods to achieve optical consistency using known real objects and special cameras, but they are difficult to use in AR applications. In this research, we propose an end-to-end method to convert an optically inconsistent AR image into an optically consistent AR image using a generative adversarial network (GAN). In addition, we propose a GAN that focuses on the structural edges of virtual objects in order to be able to handle different virtual object shapes. We confirmed that the GAN can generate photorealistic AR images consistent with the real world and that it is possible to generate images with versatility for virtual object shapes.

The initiative of Digital Transformation (DX) for the purpose of productivity improvement and reforming work style has been more and more active in the construction industry. In particular, research and development of autonomous construction machinery has been carried out with the aim of improving productivity through autonomous construction. On the other hand, Internal Model Control (IMC) system based on Database-Driven Modeling for an autonomous excavator is developed by authors. However, the control performance of this control system may deteriorate by the sudden change of the control target property. In addition, the control system can't deal with constraints explicitly in the case of the limitation of the hardware such as actuators. This paper presents a method of Database-Driven Model Predictive Control (DD-MPC) system which has also good control performance during the change of the control target property and deals with constraints explicitly. The effectiveness of the proposed method is verified by the numerical simulations and the experiment using a radio-controlled (RC) excavator.

In this paper, nonlinear controllers are designed for buck converters via the Sum of Squares (SOS) method based on their discretized bilinear model. We first consider a class of control laws in a rational funciton. The condition of Lyapunov stability theory is converted to an SOS condition, and by using the SOS method, a nonlinear stabilizing controller for the discrete-time system is derived to ensure that input constraints on buck converters can also be satisfied. However, with such a control law, the rate of convergence may be poor. To solve this problem, we next obtain through the SOS method a control law satisfying the input constraints and maximizing the Lyapunov function decrement in each sampling time. Finally, simulation and experimental results are shown to confirm the effectiveness of the second control law.

A method for obtaining the colored noise gain function is proposed. The colored noise gain function is the function which gives the auto-covariance of the output signal of SISO (single input and single output system) from the auto-covariance of the input singnal. The input signal is asumed to be white or colored Gausian noise. And ACF(auto-covariance function) for ARMA (autoregressive moving average) model is derived from the proposed function with white noise input in closed form. Some results of the calculation of ACF with white noise input and the simulation of the system with colored noise input show the effectiveness of the proposed method.