Recursive least-squares temporal difference algorithm (RLS-TD) is deduced, which can use data more efficiently with fast convergence and less computational burden. Reinforcement learning based on recursive least-squar...
详细信息
This paper presents the hardware implementation of a neural network controller for a nonlinear system. As a learning algorithm for a neural network, the reference compensation technique has been implemented on a low c...
详细信息
This paper presents the hardware implementation of a neural network controller for a nonlinear system. As a learning algorithm for a neural network, the reference compensation technique has been implemented on a low cost micro-controller unit (MCU), while PID controllers with counters and PWM generators are implemented on an FPGA chip. Interface between an MCU and a field programmable gate array (FPGA) chip has been developed to complete hardware implementation of a neural controller. The neural controller has been tested for controlling the inverted pendulum as a nonlinear system. Reference compensation technique
in particle navigation problem strategy development is crucial. The difficulties encountered by the particles during their navigation tasks require different approaches in problem solving. One way to overcome the diff...
详细信息
This paper presents experimental studies of impedance force tracking control algorithm for a crack sealing robot. The robot is built to find and seal cracks on the pavement. Regulating contact force improves the perfo...
详细信息
This paper presents experimental studies of impedance force tracking control algorithm for a crack sealing robot. The robot is built to find and seal cracks on the pavement. Regulating contact force improves the performance of cleaning process before sealing. The proposed impedance force control method is robust to perform tasks under unknown surface condition such as stiffness and position of the environment. Experimental studies show that the robot regulates a desired force quite well on the curved unknown environment.
We study the non-linear behavior of the KIII model for natural image classification. The KIII model is designed to be a dynamic computational model that simulates the sensory cortex. The KIII model has been explored f...
详细信息
We study the non-linear behavior of the KIII model for natural image classification. The KIII model is designed to be a dynamic computational model that simulates the sensory cortex. The KIII model has been explored for rudimentary pattern recognition and classification in noisy environment [1-3]. We extend the study of KIII models in understanding whether self-organized neural populations can be exploited into perceptual and memory producing systems such as in natural image classification. Our goal is to obtain a quantitative index on how well the KIII model behaves when it is assigned the task to identify and distinguish one class of natural image from the other based on color and texture features. For twenty training data, twenty validation data and eighty test data set for four image classes, we obtain 80% correct classification using the KIII. We compare a standard non linear neural network tools such as back propagation for the classification of the same set of natural images and obtain 65% correct classification. We conclude that dynamic neural computational models such as KIII may be suitable candidates for natural image classification.
Combined with a single circuit UHV power transmission project in one million voltage from Jincheng to Jingzhou via Nanyang which is reasoned by State Power Grid Corp. of China, its switching overvoltages, including th...
详细信息
Combined with a single circuit UHV power transmission project in one million voltage from Jincheng to Jingzhou via Nanyang which is reasoned by State Power Grid Corp. of China, its switching overvoltages, including three-phases closing with no load and single-phase auto-reclosing, are considered. The switching overvoltages are calculated with Electromagnetic Transient Program (EMTDC). The calculation shows that the switching overvoltage is too high up to 1.80 pu if only with parallel resistor in circuit breaker or high performance ZnO arrester. Combining arrester and parallel resistor is a good measure to limit switching overvoltage without such high requirement on arrester's or resister's thermal capacity.
In the standard support vector machines for classification, the use of training sets with uneven class sizes results in classification biases towards the class with the large training size. The main causes lie in that...
详细信息
As Tenney and Sandell pointed out, the optimal local compression/decision rule has the form of a likelihood ratio when the local observations are not correlated [3]. However, this does not hold in general for correlat...
详细信息
Area-covering operation is a special kind of path planning, which requires the robot path to cover every part of the workspace. In this paper, a neural dynamics based algorithm is proposed for real-time map building a...
详细信息
In this paper, a design methodology for enhancing the stability of humanoid robots is presented. Fuzzy Q-Learning (FQL) is applied to improve the Zero Moment Point (ZMP) performance by intelligentcontrol of the trunk...
详细信息
暂无评论