Yoshifumi Sekine

Development of a pulse control-type MEMS microrobot with a hardware neural network

This article presents the micro-electro-mechanical systems (MEMS) microrobot which demonstrates locomotion controlled by hardware neural networks (HNN). The size of the microrobot fabricated by the MEMS technology is 4 × 4 × 3.5 mm. The frame of the robot is made of silicon wafer, and it is equipped with a rotary-type actuator, a link mechanism, and six legs. The rotary-type actuator generates rotational movement by applying an electrical current to artificial muscle wires. The locomotion of the microrobot is obtained by the rotation of the rotary-type actuator. As in a living organism, the HNN realized robot control without using any software programs, A/D converters, or additional driving circuits. A central pattern generator (CPG) model was implemented as an HNN system to emulate the locomotion pattern. The MEMS microrobot emulated the locomotion method and the neural networks of an insect with the rotary-type actuator, the link mechanism, and the HNN. The microrobot performed forward and backward locomotion, and also changed direction by inputting an external trigger pulse. The locomotion speed was 0.325 mm/s and the step width was 1.3 mm.

Biomimetics Micro Robot with Active Hardware Neural Networks Locomotion Control and Insect-Like Switching Behaviour:

In this paper, we presented the 4.0, 2.7, 2.5 mm, width, length, height size biomimetics micro robot system which was inspired by insects. The micro robot system was made from silicon wafer fabricated by micro electro mechanical systems (MEMS) technology. The mechanical system of the robot was equipped with small size rotary type actuators, link mechanisms and six legs to realize the insect-like switching behaviour. In addition, we constructed the active hardware neural networks (HNN) by analogue CMOS circuits as a locomotion controlling system. The HNN utilized the pulse-type hardware neuron model (P-HNM) as a basic component. The HNN outputs the driving pulses using synchronization phenomena such as biological neural networks. The driving pulses can operate the actuators of the biomimetics micro robot directly. Therefore, the HNN realized the robot control without using any software programs or A/D converters. The micro robot emulated the locomotion method and the neural networks of an insect with rotary type actuators, link mechanisms and HNN. The micro robot performed forward and backward locomotion, and also changed direction by inputting an external trigger pulse. The locomotion speed was 26.4 mm/min when the step width was 0.88 mm.

Synchronization of coupled pulse-type hardware neuron models for CPG model

It is well known that locomotion rhythms of living organisms are generated by CPG (Central Pattern Generator). In this paper, we discuss the synchronization phenomena and oscillation patterns of the coupled oscillators system using pulse-type hardware neuron models (P-HNMs) for the purpose of constructing the CPG model. It is shown that the plural coupled P-HNMs connected by excitatory-inhibitory mutual coupling can generate various oscillation patterns. Therefore, we construct the CPG model by using the coupled P-HNMs to generate several locomotion rhythms. As a result, we show clearly that the IC chip of CPG model, which can generate the quadruped locomotion patterns, can be constructed by CMOS process.

Short-term memory circuit using hardware ring neural networks

A number of studies have recently been made on various neuron models and neural networks. This research is studied for applications to engineering problems and an understanding of the information processing functions of living organisms. We are studying an asynchronous neural network using a pulse-type hardware neuron model (P-HNM). Recently, we have been trying to construct a short-term memory circuit using hardware ring neural networks (RNN) with P-HNM. In this article, we discuss the construction of a short-term memory circuit using the hardware RNN, and conduct experiments that explain the characteristics of the network through circuit simulation using PSpice. As a result, we verify that the RNN which is proposed in this article can be used as the short-term memory circuit.

A Colpitts-Type Crystal Oscillator for Gigahertz Frequency

Recent research shows that stable frequency is required to improve efficiency in the high frequency range. A high frequency oscillator has been required for the development of radio communications, measurement equipment, etc. Generally, the crystal oscillator is excellent in short-term frequency stability, and is being applied to telecommunication equipment, information technology, etc. On the other hand, the quartz resonator, which was developed for high frequency range, is aimed at being used in systems for the next generation. The quartz resonator with the high frequency range requires a steady oscillator to oscillate in high frequency range. Therefore, it is very difficult to use a Colpitts-type crystal oscillator in the high frequency range. It requires more negative resistance on the crystal oscillator circuit. In this paper, we suggest a method to decrease Miller capacitance of the transistor used for the conventional Colpitts-type crystal oscillator. Next, we show a new Colpitts-type crystal oscillator, which can obtain a negative resistance value in a gigahertz frequency band. Analyzing the equivalent circuit clarifies the effects of the proposed circuit compared to the conventional Colpitts-type crystal oscillator

A wide variable range VCXO for IC

We suggest the use of a Miller capacitor as the variable capacitance circuit of the BJT-VCXO (VCXO using a bipolar junction transistor). When the external MOSFETs gate-drain capacitance is 10 pF, we show that the BJT-VCXO using this variable capacitance circuit has a wide frequency variation of about 500 ppm. Also, we suggest the use of the MOSFETs resistance change as the variable capacitance circuit of the CMOS-VCXO (VCXO using a CMOS crystal oscillator). When the external gate-drain capacitance is 30 pF, we show that the CMOS-VCXO using this variable capacitance circuit has a wide frequency variation of over 330 ppm.

Pulse-Type Hardware Neural Network with Two Time Windows in STDP

In recent years, synaptic plasticity, which is dependent on the order and time interval of pre- and post-synaptic spikes, has been observed by physiological experiments. There are two types of STDP which are characterized by an asymmetric time window and a symmetric time window (mexican hat type window). A symmetric time window especially depends on the influence of an inhibitory neuron. In this paper, we investigate the synaptic circuit and the synaptic weight control circuit using STDP by inhibitory interneuron input or no input. As a result, we show that the synaptic circuit using STDP with the time windows of these two types could be constructed with a simple circuit configuration considering an inhibitory interneuron by using the circuit simulator PSpice. Furthermore, we show the characteristic of reinforcement and suppression.

1 [GHz] high frequency Colpitts oscillator

As progress in the information and communication infrastructure toward the broadband network, rapid and accurate transmission of vast amounts of data is required. In accordance with it, the crystal oscillator as a frequency generator is required to improve the frequency short-term stability as well as to achieve higher frequency. The frequency short-term stability can be specified with two domains, one is time-domain and the other is frequency domain. Each domain is used according to application and condition as below; time domain: Allan variance, TDEV, jitter; frequency domain: phase noise. In the past, the crystal oscillator using PLL was popular for the high frequency oscillator. But since the oscillator using PLL multiplies frequency, it deteriorates jitter characteristics by the influence of spurious at the fundamental oscillation and it could result in causing bit error in the equipment. In order to improve the frequency short-term stability, we think that the best way is a fundamental oscillation by the Colpitts circuit. However, a conventional Colpitts circuit could not get a sufficient negative resistance in higher frequency range more than 600 [MHz] bandwidth. In this paper, we suggest that such the high frequency Colpitts oscillator can get a sufficient negative resistance even in [GHz] bandwidth. In addition, we report the efficiency of our high frequency Colpitts oscillator, which was confirmed by measurement with prototype of VCSO at 1 [GHz] using the SAW resonator.

Extraction of Phase Information Buried in Fluctuation of a Pulse-type Hardware Neuron Model Using STDP

Since neural networks have superior information processing functions, many investigators attempt to model biological neurons and their networks. Furthermore, a number of studies of neural networks have recently been made with the purpose of applying engineering to the brain. In this study, we investigate the effect of STDP on the ability to extract phase information buried in fluctuation. We focus on spike timing dependent synaptic plasticity (STDP), and we construct neural networks from a pulse-type hardware neuron model using STDP. We show that phase information buried in fluctuation is revealed by the effect of STDP, making it possible to decode the synaptic weight. Moreover, we show that it is possible to extract the phase difference buried in fluctuation representing the reinforcement part of the synaptic weight, using neural networks with STDP.

High precision TCXO for rapid environmental temperature change

Recently, the increased high speed of the transmission of mobile communication devices has necessitated higher stabilization of oscillators. Also, mobile communication devices are becoming more miniaturized. It is estimated the downsizing of a high precision frequency source will become necessary, too. Therefore, the method to get a precise frequency more easily is necessary. We study TCXOs to produce a small-sized high precision frequency source. When making small-sized TCXOs, TCXOs are strongly influenced by the environmental temperature. Currently, high stability must be kept in rapid environmental temperature change. But, because the temperature sensor and the crystal resonator have different thermal time constants, the frequency stability of TCXOs in rapid environmental temperature change is generally poor. In this paper, we propose that the temperature estimate method of the change in rapid and complicated environmental temperature can be used, as a way of compensating for the temperature characteristic of TCXOs. We focus on using the temperature change of a sensor and a crystal resonator as the lag function. This method is the way of compensating for the temperature characteristic using the temperature estimate function. The temperature estimate function can support nth derivative of temperature T with respect to time t. As a result, we have shown that it is possible to compensate frequency estimate with fewer errors in rapid environmental temperature change for which conventional TCXOs cannot compensate. Next, we confirm that our proposed TCXO is useful by using simulation with actual circuit parameters and experiments. As an example, we applied this method to conventional TCXOs that have the estimated error of more than /spl plusmn/10[ppm]. As a result, we have shown that it is possible to decrease errors to less than /spl plusmn/0.01[ppm].