Improvement of Echo State Network Generalization by Selective Ensemble Learning Based on BPSO
Xiaodong Zhang,
Xuefeng Yan
Issue:
Volume 4, Issue 6, December 2016
Pages:
84-88
Received:
29 November 2016
Published:
1 December 2016
Abstract: The Echo State Network (ESN) is a novel and special type of recurrent neural network that has become increasingly popular in machine learning domains such as time series forecasting, data clustering, and nonlinear system identification. This network is characterized by large randomly constructed recurrent neural networks (RNN) called “reservoir”, in which the neurons are sparsely connected and the weights remain unchanged during training, leaving the simple training of the output layer. However, the reservoir is criticized for its randomness and instability because of the random initialization of the connectivity and weights. In this article, we introduced the selective ensemble learning based on BPSO to improve the generalization performance of ESN. Two widely studied tasks are used to prove the feasibility and priority of the selective ESN ensemble based on BPSO(SESNE-BPSO) model. And the results indicate that the SESNE-BPSO model performs better than the general ESN ensemble, the single standard ESN and several other improved ESN models.
Abstract: The Echo State Network (ESN) is a novel and special type of recurrent neural network that has become increasingly popular in machine learning domains such as time series forecasting, data clustering, and nonlinear system identification. This network is characterized by large randomly constructed recurrent neural networks (RNN) called “reservoir”, i...
Show More
Barcode Recognizable System Implementing Based on AM5728
Xicai Li,
Junsheng Shi,
Xiaoqiao Huang,
Yonghang Tai,
Chongde Zi,
Huan Yang,
Xingyu Yang,
Zhiwei Deng,
Feiyan Li
Issue:
Volume 4, Issue 6, December 2016
Pages:
89-94
Received:
29 November 2016
Published:
1 December 2016
Abstract: To refine the implementation of industrial camera requirements in terms of barcode identification, speeding the barcode image acquisition and processing challenges, as well as the defect of low accuracy. We proposed a barcode recognition framework based on AM5728 embedded system, which employed industrial CCD to scan the barcode image, moreover, integrated with AM5728 visual development platform to manipulate the collected images. After that, decoding information is yielded from series of algorithms refer to convolution filtering, barcode positioning as well as recognition facilitated by AM5728 visual development platform. Experimental outcomes validated that the accuracy of our system recognition rate can reach up to satisfied 100% in the threshold condition, with 20 frames per second barcode images recognition rate.
Abstract: To refine the implementation of industrial camera requirements in terms of barcode identification, speeding the barcode image acquisition and processing challenges, as well as the defect of low accuracy. We proposed a barcode recognition framework based on AM5728 embedded system, which employed industrial CCD to scan the barcode image, moreover, in...
Show More
Control of Flexible Joint Robot Using Integral Sliding Mode and Backstepping
Sungha Kwon,
Abner Asignacion,
Seungkyu Park
Issue:
Volume 4, Issue 6, December 2016
Pages:
95-100
Received:
6 December 2016
Accepted:
20 December 2016
Published:
16 January 2017
Abstract: The control of flexible joint robot is getting more attentions because its applications are more frequently used for robot systems in these days. This paper proposes a robust impedance controller for flexible joint robots by using integral sliding mode control and backstepping control. The sliding mode control decouple disturbances completely but requires matching condition of disturbances. The dynamic model of FJR is divided into motor side and link side and the disturbance of the link side does not satisfy matching condition and cannot be decoupled directly by the actual input in the motor side. To overcome this difficulty, backstepping control technique is used with sliding mode control. The mismatched disturbance in the link side is changed into matched one in the respect to virtual control input which is the state controlled by actual input in the motor side. Integral sliding mode control is used to preserve the impedance control performance and the improved robustness at the same time.
Abstract: The control of flexible joint robot is getting more attentions because its applications are more frequently used for robot systems in these days. This paper proposes a robust impedance controller for flexible joint robots by using integral sliding mode control and backstepping control. The sliding mode control decouple disturbances completely but r...
Show More
