Locomotion Control of MEMS Microrobot Using Pulse‐Type Hardware Neural Networks

This paper presents the locomotion control of a microelectromechanical system (MEMS) microrobot. The MEMS microrobot demonstrates locomotion control by pulse‐type hardware neural networks (P‐HNN). P‐HNN generate oscillatory patterns of electrical activity like those of living organisms. The basic component of P‐HNN is a pulse‐type hardware neuron model (P‐HNM). The P‐HNM has the same basic features as biological neurons, such as the threshold, the refractory period, and spatiotemporal summation characteristics, and allows the generation of continuous action potentials. P‐HNN has been constructed with MOSFETs and can be integrated by CMOS technology. Like living organisms, P‐HNN has realized robot control without using software programs or A/D converters. The size of the microrobot fabricated by MEMS technology was 4 × 4 × 3.5 mm. The frame of the robot was made of a silicon wafer, equipped with rotary actuators, link mechanisms, and six legs. The MEMS microrobot emulated the locomotion method and the neural networks of an insect by rotary actuators, link mechanisms, and the P‐HNN. We show that the P‐HNN can control the forward and backward locomotion of the fabricated MEMS microrobot, and that it is possible to switch its direction by inputting an external trigger pulse. The locomotion speed was 19.5 mm/min and the step size was 1.3 mm. © 2013 Wiley Periodicals, Inc. Electr Eng Jpn, 186(3): 43–50, 2014; Published online in Wiley Online Library ( wileyonlinelibrary.com ). DOI 10.1002/eej.22473     

一种量子衍生神经网络模型算法及应用

为提高神经网络的逼近和预测能力，提出一种各维输入为离散序列的量子衍生神经网络模型及算法。该模型为三层结构，隐层为量子衍生神经元，输出层为普通神经元。量子衍生神经元由量子旋转门和多位受控旋转门组成，利用多位受控旋转门中目标量子位的输出向输入端的反馈，实现对输入序列的整体记忆，利用受控旋转门输出中多位量子比特的纠缠获得量子衍生神经元的输出。基于量子计算理论设计了该模型的学习算法。该模型可从宽度和深度两方面获取输入序列的特征。仿真结果表明，当输入节点数和序列长度满足一定关系时，该模型明显优于普通神经网络。     

Assessing the Effects of Acute Amyloid β Oligomer Exposure in the Rat

Alzheimer’s disease (AD) is the most common form of dementia, yet there are no therapeutic treatments that can either cure or delay its onset. Currently, the pathogenesis of AD is still uncertain, especially with respect to how the disease develops from a normal healthy brain. Amyloid β oligomers (AβO) are highly neurotoxic proteins and are considered potential initiators to the pathogenesis of AD. Rat brains were exposed to AβO via bilateral intracerebroventricular injections. Rats were then euthanized at either 1, 3, 7 or 21-days post surgery. Rat behavioural testing was performed using the Morris water maze and open field tests. Post-mortem brain tissue was immunolabelled for Aβ, microglia, and cholinergic neurons. Rats exposed to AβO showed deficits in spatial learning and anxiety-like behaviour. Acute positive staining for Aβ was only observed in the corpus callosum surrounding the lateral ventricles. AβO exposed rat brains also showed a delayed increase in activated microglia within the corpus callosum and a decreased number of cholinergic neurons within the basal forebrain. Acute exposure to AβO resulted in mild learning and memory impairments with co-concomitant white matter pathology within the corpus callosum and cholinergic cell loss within the basal forebrain. Results suggest that acute exposure to AβO in the rat may be a useful tool in assessing the early phases for the pathogenesis of AD.     

基于单神经元和前馈补偿的电动加载复合控制

针对飞机舵机电动加载系统的力矩跟踪和干扰抑制的优化问题,提出一种基于单神经元和扰动前馈抑制的电动加载复合控制方法。首先利用改进的粒子群算法调整单神经元PID控制器的参数和系数K,然后对舵机角位移进行前馈补偿来抑制多余力矩,以减小舵机位置干扰对系统的影响;最后对电动加载复合控制系统进行仿真。仿真结果表明,系统的抗干扰能力及加载精度得到进一步提高,证明了该复合控制方法在理论上的可行性。     

人工神经网络概述

近年来,随着人工神经网络在信息、自动化、医学、经济等领域的广泛应用和杰出表现,使得人工神经网络又开始得到广泛关注和重新重视。基于需求,本文对人工神经网络做了简单概述。     

Single‐layer perceptron and dynamic neuron implementing linearly non‐separable Boolean functions

This paper presents a single‐layer perceptron (SLP) scheme with an impulse activation function (IAF) and a dynamic neuron (DN) with a trapezoidal activation function (TAF). Combining with some interesting properties of the offset levels, it is shown that many linearly non‐separable Boolean functions can be realized by using only one SLPwIAF or one DNwTAF. In the present work, a few appropriate IAF and TAF are adopted, and the inverse offset level method is used for the design of the SLPwIAF synaptic weights and the DNwTAF templates. The XOR and NXOR Boolean operations with two inputs and all 152 non‐separable Boolean functions with three inputs can be easily implemented by one SLPwIAF or one DNwTAF. Finally, the entire set of 152 DNwTAF templates associated with 152 non‐separable Boolean functions of three inputs is completely listed. Copyright © 2008 John Wiley & Sons, Ltd.     

Odor discrimination by G protein-coupled olfactory receptors

The vertebrate olfactory system possesses a remarkable capacity to recognize and discriminate a variety of odorants by sending the coding information from peripheral olfactory sensory neurons in the olfactory epithelium to the olfactory bulb of the brain. The recognition of odorants appear to be mediated by a G protein-coupled receptor superfamily that consists of approximately 1% of total genes in vertebrates. Since the first discovery of the olfactory receptor gene superfamily in the rat, similar chemosensory receptors have been found in various species across different phyla. The functions of these receptors, however, had been uncharacterized until the recently successful functional expression and ligand screening of some olfactory receptors in various cell expression systems. The functional cloning of odorant receptors from single olfactory neurons allowed for the identification of multiple receptors that recognized a particular odorant of interest. Reconstitution of the odorant responses demonstrated that odorant receptors recognized various structurally-related odorant molecules with a specific molecular receptive range, and that odor discrimination is established based on a combinatorial receptor code model in which the identities of different odorants are encoded by a combination of odorant receptors. The receptor code for an odorant changes at different odorant concentrations, consistent with our experience that perceived quality of an odorant changes at different concentrations. The molecular bases of odor discrimination at the level of olfactory receptors appear to correlate well with the receptive field in the olfactory bulb where the input signal is further processed to create the specific odor maps.     

基于单神经元自适应PID控制的温控系统研究

采用Bang—Bang控制和单神经元自适应PID控制相结合的控制方法，并且对Bang—Bang控制进行改进，即引入逐级递减Bang—Bang控制的概念，并且使Bang—Bang控制的控制初始值以一定的规律按照温度设定值的变化而变化。实验结果表明，该控制算法对于电加热炉温度具有较好的控制效果。     

一种新型的自适应模糊单神经元控制算法

提出了一种新型的模糊单神经元控制算法，它不需要对象的数学模型，同时又具有模糊控制和单神经元控制的优点，并且参数在动态过程可自适应调整．仿真实验表明这种算法具有较好的动态过程性能指标和较强的鲁棒性．     

基于RBF神经网络的开关磁阻电机单神经元PID控制

论文提出了一种基于径向基函数（radial basis function）神经网络在线辨识的开关磁阻电机（SRM）单神经元PID自适应控制新方法。该方法针对开关磁阻电机的非线性，利用具有自学习和自适应能力的单神经元来构成开关磁阻电机的单神经元自适应控制器，不但结构简单，而且能适应环境变化，具有较强的鲁棒性。并构造了一个RBF网络对系统进行在线辨识。建立其在线参考模型．由单神经元控制器完成控制器参数的自学习，从而实现控制器参数的在线调整，能取得更好的控制效果。样机的实验结果表明，文中所提出的基于RBF神经网络辨识的开关磁阻电机单神经元自适应PID控制方法，通过在线辨识建立了过程模型并为神经元控制器提供了梯度信息，达到了在线辨识在线控制的目的，控制精度高，动态特性好。