Cognitive Computation and Systems最新文献

In microservice systems, availability can be ensured through a variety of measures, such as fault tolerance and flow limiting, which are collectively called the flow control. In the current mainstream system design, the flow control rules are usually fixed and set manually, which cannot be dynamically adjusted according to the flow shape. The performance of the system is thus not fully explored. To mitigate this problem, an adaptive dynamic flow control algorithm is proposed. Based on the system's monitoring data and current flow, the algorithm calculates the flow-limiting threshold in real time, and then it implements fine-grained service adaptive flow control to improve the resource utilization. Experimental results show that the performance of the adaptive automatic flow control is better than that of the traditional static method on resource utilization.

Machine-based language translation has been certainly picking up. Still, machines lag behind the cognitive powers of human beings. Neural Machine Translation (NMT) methods require huge datasets and computational power for high-quality translation. A novel Sequential Adaptive Memory (SAM) cognitive model-based machine translation system for English to Hindi translation, was proposed. This model is an augmented version of the Cortical Learning Algorithm (CLA). The SAM is based on the architecture of the neocortex region of the brain, where speech and language comprehension and production take place. The proposed model is capable of learning with smaller datasets. This model employs the sequence to sequence learning approach, which provides better quality translation. It enables the creation of word pairs, dictionaries, and rules for translation. The results of the proposed approach are compared with the traditional phrase-based SMT approach as well as with the state-of-the-art NMT approach. The results are comparable with the results of the conventional approaches. We illustrate that the limitations of the approaches are won over by the proposed SAM approach. It is observed that SAM is capable of exhibiting satisfactory quality translation for low resource languages as well.

Soft grippers are compliant and self-adaptive, and can be highly compatible with the surrounding environment in grasping tasks. Currently, most soft pneumatic grippers are developed with a single grasping configuration, which leads to poor universality for different objects. Additionally, the oscillation caused by actuator's elastic bodies will result in poor stability during grasping and transportation, which can be improved by stiffness enhancement. A four-fingered soft pneumatic gripper is proposed by integrating multi-configuration and variable-stiffness functionality. The multi-configuration was realised by using the motion characteristics of a tangent mechanism. Meanwhile, a damping method based on electrorheological fluids was applied on a pneumatic actuator to improve the grasping stability. Besides, a machine vision technique was adopted to automatically adjust the grasping posture during manipulation. As a result, the proposed multi-configuration gripper can self-adaptively grasp different shapes of objects, especially two classical types, a pen canister as the flat cylinder and a cuboid box as the long cylinder. In addition, the electrorheological variable-stiffness method was manifested to be applicable for reducing pneumatic finger vibration. This research is expected to improve the versatility and grasping stability of soft pneumatic grippers.

Cancer is a group of related diseases with high mortality rate characterized by abnormal cell growth which attacks the body tissues. Microarray cancer data is a prominent research topic across many disciplines focused to address problems related to the higher curse of dimensionality, a small number of samples, noisy data and imbalance class. A random forest (RF) tree-based feature selection and ensemble learning based on hard voting and soft voting is proposed to classify microarray cancer data using six different base classifiers. The selected features due to RF tree are submitted to the base classifiers as the training set. Then, an ensemble learning method is applied to the base classifiers in which case each base classifier predicts class label individually. The final prediction is carried out hard and soft voting techniques that use majority voting and weighted probability on the test set. The proposed ensemble learning method is validated on eight different standard microarray cancer datasets, of which three of the datasets are binary class and the remaining five datasets are multi-class datasets. Experimental results of the proposed method show 1.00 classification accuracy on six of the datasets and 0.96 on two of the datasets.

As the traditional single biological signal or physical signal is not good at predicting the angle value of the knee joint, the innovative fusion of biological signals and physical signals is used to analyze the movement posture of the lower limbs. In order to solve the problem of human movement intention recognition, a wearable is designed. The signal-acquisition experiment platform uses muscle electrical signals and joint angle signals as motion data. After the signals are processed, the KNN algorithm is used to identify the four gait motion modes of the human body to walk naturally, climb stairs, descend stairs, and cross obstacles. The test results show that it is feasible to use the KNN algorithm to analyze the strength of the active and passive muscles of the knee joint movement according to different thigh lift heights, and to predict the knee joint angle value when the human body goes up and down the stairs. The comprehensive prediction accuracy rate reaches 91.45%.

Manipulation skill learning and generalisation have gained increasing attention due to the wide applications of robot manipulators and the spurt of robot learning techniques. Especially, the learning from demonstration method has been exploited widely and successfully in the robotic community, and it is regarded as a promising direction to realise the manipulation skill learning and generalisation. In addition to the learning techniques, the immersive teleoperation enables the human to operate a remote robot with an intuitive interface and achieve the telepresence. Thus, it is a promising way to transfer manipulation skills from humans to robots by combining the learning methods and teleoperation, and adapting the learned skills to different tasks in new situations. This review, therefore, aims to provide an overview of immersive teleoperation for skill learning and generalisation to deal with complex manipulation tasks. To this end, the key technologies, for example, manipulation skill learning, multimodal interfacing for teleoperation and telerobotic control, are introduced. Then, an overview is given in terms of the most important applications of immersive teleoperation platform for robot skill learning. Finally, this survey discusses the remaining open challenges and promising research topics.

Exoskeleton has been applied in the field of medical rehabilitation and assistance. However, there are still some problems in the interaction between human and exoskeleton, such as time delay, the existence of certain constraints on the human body, and the movement in time is hard to follow. A human motion pattern recognition model based on the long short-term memory (LSTM) is proposed, which can recognise the state of the human body. Meanwhile, the orthogonalisation method is integrated to make personal-specific disentangling, and it can effectively improve the generalisation ability of different groups of people, so as to improve the effective follower ability of the exoskeleton. Compared with some other traditional methods, this model has better performance and stronger generalisation ability, which has certain significance in the field of exoskeleton algorithm.

Multi-tag anti-collision is an important problem in radio frequency identification (RFID) application. Solving the problem is of great significance to the RFID technology application and the future internet of things; therefore, an adaptive frame slotted ALOHA anti-collision algorithm based on tag grouping (IGA) is proposed. First, a novel method for estimating the number of tags accurately is proposed. Through theoretical research and the experimental verification, a relationship is obtained between the ratio of the collision time slot in the frame and the average number of tags in each collision slot, which helps us to calculate the number of tags. Second, the method of estimating the number of tags is applied to the IGA algorithm. The reader randomly groups the tags after the number of tags are estimated, and recognises the tags by grouping. In the identification process, the idle time slot is skipped automatically, and the collided tags can be identified with an additional frame until all tags are identified. The simulation results show that the total time slot of the IGA algorithm is relatively small, and the identification efficiency is about 71%, which is 30% better than the the improved RFID anti-collision algorithm and 90% higher than the traditional ALOHA algorithm.

It is of significant practical importance to ensure the operational safety of the smart grid, which requires real-time fault diagnosis and identifying what causes it based on an enormous amount of data. This article further studies the intelligent fault-identification method based on the combination of multi-machine learning methods on the bases of researching on Fault Diagnosis of Smart Grid Data Acquisition System. Firstly, we should apply statistical analysis and feature extraction for fault data. Then, we can use fuzzy neural network (FNN) to calculate the probability of fault prediction of power distribution stations, manufacturers and operation businesses, and use the membership function to calculate the corresponding fault membership and uncertainty. Secondly, it makes use of Dempster/Shafer (DS) evidence sequential fusion method to realize fault membership fusion, and gives the corresponding decision criteria for failure causes. Thirdly, a fault-identification method of smart grid data-acquisition system is established based on FNN and DS Evidence Fusion. Finally, the experimental results based on the actual operation data of smart grid show that the new method has a very good application effect at fault cause identification.