-
Towards Robust and Accurate Myoelectric Controller Design based on Multi-objective Optimization using Evolutionary Computation
Authors:
Ahmed Aqeel Shaikh,
Anand Kumar Mukhopadhyay,
Soumyajit Poddar,
Suman Samui
Abstract:
Myoelectric pattern recognition is one of the important aspects in the design of the control strategy for various applications including upper-limb prostheses and bio-robotic hand movement systems. The current work has proposed an approach to design an energy-efficient EMG-based controller by considering a kernelized SVM classifier for decoding the information of surface electromyography (sEMG) si…
▽ More
Myoelectric pattern recognition is one of the important aspects in the design of the control strategy for various applications including upper-limb prostheses and bio-robotic hand movement systems. The current work has proposed an approach to design an energy-efficient EMG-based controller by considering a kernelized SVM classifier for decoding the information of surface electromyography (sEMG) signals to infer the underlying muscle movements. In order to achieve the optimized performance of the EMG-based controller, our main strategy of classifier design is to reduce the false movements of the overall system (when the EMG-based controller is at the `Rest' position). To this end, we have formulated the training algorithm of the proposed supervised learning system as a general constrained multi-objective optimization problem. An elitist multi-objective evolutionary algorithm $-$ the non-dominated sorting genetic algorithm II (NSGA-II) has been used to tune the hyperparameters of SVM. We have presented the experimental results by performing the experiments on a dataset consisting of the sEMG signals collected from eleven subjects at five different upper limb positions. Furthermore, the performance of the trained models based on the two-objective metrics, namely classification accuracy, and false-negative have been evaluated on two different test sets to examine the generalization capability of the proposed training approach while implementing limb-position invariant EMG classification. It is evident from the presented result that the proposed approach provides much more flexibility to the designer in selecting the parameters of the classifier to optimize the energy efficiency of the EMG-based controller.
△ Less
Submitted 22 May, 2023; v1 submitted 2 April, 2022;
originally announced April 2022.
-
Spelling Error Trends and Patterns in Sindhi
Authors:
Zeeshan Bhatti,
Imdad Ali Ismaili,
Asad Ali Shaikh,
Waseem Javaid
Abstract:
Statistical error Correction technique is the most accurate and widely used approach today, but for a language like Sindhi which is a low resourced language the trained corpora's are not available, so the statistical techniques are not possible at all. Instead a useful alternative would be to exploit various spelling error trends in Sindhi by using a Rule based approach. For designing such techniq…
▽ More
Statistical error Correction technique is the most accurate and widely used approach today, but for a language like Sindhi which is a low resourced language the trained corpora's are not available, so the statistical techniques are not possible at all. Instead a useful alternative would be to exploit various spelling error trends in Sindhi by using a Rule based approach. For designing such technique an essential prerequisite would be to study the various error patterns in a language. This pa per presents various studies of spelling error trends and their types in Sindhi Language. The research shows that the error trends common to all languages are also encountered in Sindhi but their do exist some error patters that are catered specifically to a Sindhi language.
△ Less
Submitted 19 March, 2014;
originally announced March 2014.
-
Transparent caching of virtual stubs for improved performance in ubiquitous environments
Authors:
Lachhman Das Dhomeja,
Yasir Arfat Malkani,
Asad Ali Shaikh,
Ayaz Keerio
Abstract:
Context-awareness is an essential requirement for pervasive computing applications, which enables them to adapt and perform tasks based on context. One of the adaptive features of context-awareness is contextual reconfiguration. Contextual reconfiguration involves discovering remote service(s) based on context and binding them to the application components to realize new behaviors, which may be ne…
▽ More
Context-awareness is an essential requirement for pervasive computing applications, which enables them to adapt and perform tasks based on context. One of the adaptive features of context-awareness is contextual reconfiguration. Contextual reconfiguration involves discovering remote service(s) based on context and binding them to the application components to realize new behaviors, which may be needed to satisfy user needs or to enrich user experience. One of the steps in the reconfiguration process involves a remote lookup to discover the service(s) based on context. This remote lookup process provides the largest contribution to reconfiguration time and this is due to fact that the remote calls are much slower than local calls. Consequently, it affects system performance. In pervasive computing applications, this may turn out to be undesirable in terms of user experience. Moreover, other distributed applications using the network may be affected as every remote method call decreases the amount of bandwidth available on the network. Various systems provide reconfiguration support and offer high-level reconfiguration directives to develop adaptive context-aware applications, but do not address this performance bottleneck. We address this issue and implement seamless caching of virtual stubs within our PCRA1 for improved performance. In this paper we present and describe our transparent caching support and also provide its performance evaluation.
△ Less
Submitted 25 November, 2011;
originally announced January 2012.
