Kenichi Abe

Quantitative Evaluation of Effects of Visually-Induced Motion Sickness Based on Causal Coherence Functions between Blood Pressure and Heart Rate

To evaluate the effect of VIMS, the present study has analyzed the linearity of the baroreflex system and that of the mechanical hemodynamic system by using causal coherence functions. The causal coherence functions have a capability of calculating linear correlation between two systems independently even if the systems are connected with each other to compose a closed-loop system. In the experiment, 56 healthy human subjects heart rate and continuous blood pressure variability were measured to obtain the causal coherence functions when they were watching an unstable video image. The results showed that there were significant differences in the causal coherence functions as well as the traditional coherence function between the subjects who felt VIMS and those who did not, and that the hemodynamic system was mainly disturbed by VIMS rather than the baroreflex system. These findings suggest that the causal coherence functions of the two systems and the traditional coherence function of the whole system gave different information from one another. This fact implies that the causal coherence functions will be useful and can be objective means to quantify VIMS.

Aerodynamic characteristics of an iced cup-shaped body

The negative effects of ice accretion on wind speed measurements by a cup anemometer have been indicated by analyses of field test results and already been studied numerically and experimentally by the authors of this paper. Implementation of calculations of the dynamic behavior of a cup anemometer with or without ice was carried out employing the fictitiously altered aerodynamic characteristics of an iced or clean cup based on NACA test results for a conical cup because no information covering the aerodynamic characteristics of an iced cup had been provided up to that point. In light of the absence of solid data, wind tunnel tests of the aerodynamic characteristics of iced cup-shaped bodies were performed. Imitated ice models were built and used for the test based on test results from an icing wind tunnel test conducted separately. We found that, depending on its amount, ice accretion in dry-growth conditions results in big penalties to the aerodynamics of cup-shaped bodies, while ice deposits of wet growth cause little effect, regardless of the thickness of ice.

A phased reinforcement learning algorithm for complex control problems

In this article, a phased reinforcement learning algorithm for controlling complex systems is proposed. The key element of the proposed algorithm is a shaping function defined on a novel position–direction space. The shaping function is autonomously constructed once the goal is reached, and constrains the exploration area to optimize the policy. The efficiency of the proposed shaping function was demonstrated by using a complex control problem of positioning a 2-link planar underactuated manipulator.

Analysis of Human Learning Process on Manual Control of Complex Systems

In this paper, a novel analysis technique is applied for investigating human operators trial and error learning process to control a nonholonomic system, 2-link planer underactuated manipulator (2PUAM). An essential core of the technique is to use a value function of the reinforcement learning scheme for revealing how the operators can find a control strategy. It is an advantage of the proposed technique compared to the others that a transition of the value function may explain the changes of the operators strategies during the learning process. According to the results of the analysis, the operators tended to explore an objective trajectory first, and then shift to the tracking control of the trajectory. The tracking was accompanied with acceleration to achieve the goal faster. Interestingly, the acceleration disturbs the objective trajectory due to the complex dynamics of the target, and induces another exploration to get better trajectories. The fact that this phase transition structure under unsupervised learning environment is consistent with previously reported results for a supervised case implies that the structure can be a general nature of human learning process.

Natural intelligence: noise-resistance of neural spike communication

In this article, we analyze the neural spike dynamics of a double feedback neural unit (DFNU). The essential emphasis of the analysis is on the use of the DFNU’s simple formulations that can provide quantitative analytical results. Comparing the dynamics of the Hodgkin-Huxley model to that of the DFNU, it is shown that the dynamics of the DFNU are also physiologically plausible under certain conditions. The results suggest that high-frequency firings are relatively appropriate for a neural informational carrier due to their reliability and robustness to noisy inputs.

Assessment of Biological Effect of Minimal Acupuncture Using Causal Coherence Function between Blood Pressure and Heart Rate

To assess the effect of acupuncture on the human, 17 healthy volunteers participated in a double-blind randomized crossover clinical trial using real-needle and no-needle seals. In this study, by using systolic blood pressure variability (SBP) and heart rate variability (HR), not only coherence function (K²) between SBP and HR but also two causal coherence functions: K² _SBPrarrHR from SBP to HR and K² _HRrarrSBP from HR to SBP were introduced to divide causal linearity of the cardiovascular system regarded as a closed-loop system. K² represents total linearity of the system. K² _SBPrarrHR and K² _HRrarrSBP correspond to the baroreflex system and the mechanical hemodynamics, respectively. The results showed that K² _SBPrarrHR over the low frequency band (0.04-0.15 Hz) underwent a lot of change in real-needle acupuncture but did not in no-needle acupuncture. And there was difference in K² _HRrarrSBP over the high frequency band (0.15-0.45 Hz) immediately after the stimulation of acupuncture. These results have never been obtained from conventional methods dealing with a cardiovascular system as an open-loop system

Development of Flight Simulator for Human-Powered Aircraft - The Road towards a World Record

This is a preliminary report about our plan for a special-purpose flight simulator. The human-powered aircraft developed in Nihon University has achieved the Japanese flight distance record of 49.2 km in 2005. However, this Japanese record is still half the distance of the world record of 115.1 km that Massachusetts Institute of Technology achieved in 1988. Thus, the present goal for the development group of our university students is to break this world record. For this purpose, a research group for a human-powered flight support system was established to develop a special-purpose flight simulator for human-powered flight. The purpose of designing such a simulator is to provide the student pilot with a scientific and practical training facility and to feed the flight data and design factors back into an optimal design of the human-powered aircraft. That is, the simulator would be used to search for an optimal flight path planning, optimal power time-distribution program for the pilot, and the optimal configuration of the aircraft. To find the optimal flight path and optimal power distribution, an appropriate mathematical model for the pilot as a man-powered engine must also be analyzed and determined as well as a conventional pilot model in handling the control stick. The developed simulator would also be used to compare an ordinary student pilot with an athlete pilot. More details would be presented at the conference

Self-Organizing Neural Networks using Discontinuous Teacher Data for Incremental Category Learning

In this paper, we develop a neural model that forms categories of inputs for some practical applications such as pattern recognition, learning, image processing, and trend analysis. The developed model is based on natural mechanisms of biological behavior instead of artificial one such as clustering algorithms. The essential point of the model is to regard the teacher information as a first priority for an accurate learning. Then, the model can carry the accurate classification of complex and imbalanced categories by using discontinuous teacher data under an incremental learning environment. Simulation results demonstrate the usefulness and the weakness of the model on practical category formation tasks

Phased Learning with Hierarchical Reinforcement Learning in Nonholonomic Motion Control

In this paper, a hierarchical reinforcement learning algorithm for controlling nonholonomic systems is proposed. When applying reinforcement learning to the nonholonomic systems, acquiring adequate policies is difficult because of an increase of learning steps and a convergence of local optimal policies. The proposed algorithm is inspired by such human learning behavior. Human can learn to control such systems sufficiently even if they initially have little knowledge about the systems dynamics and the way to control. This human capability is suggested to be caused by their exploration strategies for acquiring the adequate policies. The key element of the proposed algorithm is a shaping function defined on a novel position-direction space. The shaping function is autonomously constructed once the goal is reached and constrains the exploration area to optimize the policy. The efficiency of the proposed shaping function was demonstrated by using a nonholonomic control problem of positioning the 2-link planer underactuated manipulator