This study applies a nonlinear receding horizon differential game (NRHDG) to control a multicopter to avoid an adversarial obstacle pursuing the multicopter in three-dimensional space. The multicopter determines its control inputs without prior knowledge of the obstacle’s control law. We propose a method for designing a performance index that takes into account the avoidance performance of the multicopter, and we verify its effectiveness through simulations of avoidance control against an approaching obstacle. For an adversarial obstacle approaching by proportional navigation, we show that the NRHDG, which assumes that the obstacle approaches according to a game optimal solution, has better avoidance performance than the nonlinear model predictive control, which assumes constant velocity linear motion of the obstacle.

Recently, robot swarm systems have been developed and cooperative control under inter-robot communication based on passivity of human operators has been proposed. These studies assume that a human is a linear time invariant system. However, these studies did not evaluate the linear time invariance of the human. To analyze the linear time invariance of the human, we propose two analysis methods. The first method evaluates the errors of impulse responses by the human. However, this method requires a large amount of data. The second method adopts a coherence function to analyze the linear time invariance with a smaller amount of data. Finally, we perform tests for multiple participants and evaluate the linear time invariance of the participants.

We have previously proposed a statistical test rule induction method (STRIM), which induces the causality by if-then rules hiding in the dataset called the decision table in the field of the Rough Sets and confirmed its validity in a simulation model. However, the task of studying its validity and usefulness in a real-world dataset (RWD) was left to future research. Generally, the result of rule induction from an RWD cannot be directly ascertained. Therefore, after the previous STRIM was applied to an RWD, the induced rules were applied to the classification problem, and the result of classification was recognized as the validity of the rule induction method because the result was directly affected by the induced rules. Here, the classification result by Random Forest (RF) was used for an index of validity of that by the previous STRIM.

Specifically, the following were experimentally and statistically studied: (1) The result by the previous STRIM was inferior to that by RF in an RWD, although the former was superior to the latter in the simulation dataset. (2) The previous was improved into ex-STRIM, and the improved showed superior results to that by RF in the simulation and equal to that by RF in the RWD. (3) Trunk rules by tr-STRIM were extracted from the rules by the previous STRIM, and the inclusion relationships between the previous, ex-STRIM, and tr-STRIM were considered to discuss various levels of rule description in the RWD.

The Cockcroft-Walton (CW) circuit is implemented in high-voltage low-current apparatuses such as electron beam irradiation devices and insulation testing devices. This study focuses on the effect of the output voltage waveform on the changes in the input voltage amplitude. The transfer function of the output voltage corresponding to the input voltage amplitude is theoretically derived based on the CW circuit operation modes. The transfer function is evaluated through numerical and experimental analysis performed on a prototype system. In future works, we will develop a transfer function which includes the effect of the junction capacitance of the diodes.