A Deconvolutive Neural Network for Speech Classification With Applications to Home Service Robot

Wang, Donglin; Leung, Henry; Kurian, Ajeesh P.; Kim, Hye-Jin; Yoon, Ho-Sub

doi:10.1109/tim.2010.2047551

Cited by 25 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use original data points and the data transformed by wavelet packet in this paper. The angle of elbow, knee, backbone, and distance of each elbow were used for motion analysis and coaching [ 33 , 34 ].…”

Section: Machine Learning Algorithmsmentioning

confidence: 99%

Classification of Horse Gaits Using FCM-Based Neuro-Fuzzy Classifier from the Transformed Data Information of Inertial Sensor

Lee

Byeon

et al. 2016

Sensors

View full text Add to dashboard Cite

In this study, we classify four horse gaits (walk, sitting trot, rising trot, canter) of three breeds of horse (Jeju, Warmblood, and Thoroughbred) using a neuro-fuzzy classifier (NFC) of the Takagi-Sugeno-Kang (TSK) type from data information transformed by a wavelet packet (WP). The design of the NFC is accomplished by using a fuzzy c-means (FCM) clustering algorithm that can solve the problem of dimensionality increase due to the flexible scatter partitioning. For this purpose, we use the rider’s hip motion from the sensor information collected by inertial sensors as feature data for the classification of a horse’s gaits. Furthermore, we develop a coaching system under both real horse riding and simulator environments and propose a method for analyzing the rider’s motion. Using the results of the analysis, the rider can be coached in the correct motion corresponding to the classified gait. To construct a motion database, the data collected from 16 inertial sensors attached to a motion capture suit worn by one of the country’s top-level horse riding experts were used. Experiments using the original motion data and the transformed motion data were conducted to evaluate the classification performance using various classifiers. The experimental results revealed that the presented FCM-NFC showed a better accuracy performance (97.5%) than a neural network classifier (NNC), naive Bayesian classifier (NBC), and radial basis function network classifier (RBFNC) for the transformed motion data.

show abstract

Section: Machine Learning Algorithmsmentioning

confidence: 99%

Classification of Horse Gaits Using FCM-Based Neuro-Fuzzy Classifier from the Transformed Data Information of Inertial Sensor

Lee

Byeon

et al. 2016

Sensors

View full text Add to dashboard Cite

show abstract

“…With the development of robotics, many intelligent walking robots are widely applied in the life of human [1], outdoor exploration [2,3], and so on. Intelligence [4], high trafficability [5], low energy dissipation [6,7], and high load rate [8] are generally viewed as parts of the important development directions for ground walking roots.…”

Section: Introductionmentioning

confidence: 99%

Analysis Method of Articulated Torque of Heavy-Duty Six-Legged Robot under Its Quadrangular Gait

2016

View full text Add to dashboard Cite

Abstract:The articulated torque is an important base for developing heavy-duty multi-legged robots.To quickly obtain accurate articulated torques, a method is presented for an electrically driven heavy-duty six-legged robot under its quadrangular gait. First, the typical walking ways and the working conditions of support phase are analyzed. The static articulated torques are solved though Matrix Laboratory (MATLAB) software. The variable tendency charts of torques are acquired with the changes in the joint angles. The static simulation of robot is carried out by Automatic Dynamic Analysis of Mechanical Systems (ADAMS) software. A comparative analysis of the maximum static articulated torques shows that the theoretical calculation values are smaller than the static simulation values. The maximum error value is about 5%. Finally, a prototype of the electrically driven heavy-duty six-legged robot is developed. Its climbing experiments prove that the proposed method can be applied to quickly determine accurate articulated torques.

show abstract

“…Data classification by using ANN applied to many engineering problems (speech classification, power system, image processing, human-robot interface and machine vision system) [10]- [14]. Just the same as the other applied examples of ANN, one of the important issues for using ANN is the learning of ANN.…”

Section: Introductionmentioning

confidence: 99%

Genetic based feed-forward neural network training for chaff cluster detection

Lee

Jeong

et al. 2012

2012 International Conference on Fuzzy Theory and Its Applications (iFUZZY2012)

View full text Add to dashboard Cite

Data classification is one of the most important and fundamental problems in many decision making tasks. As traditional methods for data classification, discriminant analysis, k-nearest neighbor (k-NN) and support vector machine (SVM) is widely used. Also, artificial neural net-works (ANN) have emerged as an important tool for data classification. In this paper, we propose a learning method of ANN for data classification by combining genetic algorithm (GA) and performance criterion (PC). We can prevent ANN from trapping in local minimum by using GA and also avoid over-fitting problems of training data by using PC. The data used in the simulations has four attribute and can be classified by two classes. We compare the classification performance of BP learning, SVM and proposed learning method by using the k-fold cross validation technique. Among the methods used in the simulations, we can demonstrate that our proposed method shows the best performance.

show abstract

A Deconvolutive Neural Network for Speech Classification With Applications to Home Service Robot

Cited by 25 publications

References 22 publications

Classification of Horse Gaits Using FCM-Based Neuro-Fuzzy Classifier from the Transformed Data Information of Inertial Sensor

Classification of Horse Gaits Using FCM-Based Neuro-Fuzzy Classifier from the Transformed Data Information of Inertial Sensor

Analysis Method of Articulated Torque of Heavy-Duty Six-Legged Robot under Its Quadrangular Gait

Genetic based feed-forward neural network training for chaff cluster detection

Contact Info

Product

Resources

About