In this paper, the vibration suppression and trajectory tracking control of a flexible joint manipulator (FJM) based on particle swarm optimization (PSO) and fixed-time nonsingular terminal sliding mode control (NTSMC) are studied. Firstly, in order to suppress the residual vibration of the FJM, an optimal trajectory planning method based on higher-order trajectory planning (HOTP) and the PSO algorithm is proposed. Then, to ensure that the FJM can track the optimized trajectory without being affected by the initial value of the trajectory, a novel fixed-time NTSMC scheme is proposed. Compared with the cubic spline trajectory, the proposed HOTP is smoother and can more accurately suppress the residual vibration of the FJM. By combining the HOTP with the PSO algorithm, the vibration amplitude of FJM can be suppressed to around 0.002 mm. Unlike finite-time NTSMC, the rate of convergence of the proposed fixed-time NTSMC does not depend on the initial value of FJM’s joint trajectory. Especially when the initial value of the trajectory is large, the FJM can still quickly track the optimal trajectory within 0 to 1 s. Finally, the effectiveness of this method is verified through simulation and comparison.

This paper addresses the control problem of the flexible joint manipulator (FJM) with unmeasurable system states and mismatched uncertainties. First, the control system is transformed as a matched uncertain system with unmeasurable states based on differentiation method. Then, the uncertainties and unmeasurable states are estimated by designing a fixed‐time observer (FTO). Based on the integral sliding‐mode control (SMC), the bi‐limit homogeneity technique and the estimation of FTO, a finite‐time SMC is proposed for FJM. Compared with the existing finite‐time SMC method for FJM, the most attractive feature of the proposed method is that not only the finite‐time convergence is guaranteed but also the two angular velocity sensors for rotation angles of link and motor are simplified. Moreover, the proposed SMC can suppress the mismatched and matched uncertainties by using chattering‐free control input. The fixed‐time stability of FTO and finite‐time stability of proposed controller are proved. Finally, the efficiency of proposed scheme is shown by the numerical simulation.

Hashing methods have proven to be useful for a variety of tasks and have attracted extensive attention in recent years. Various hashing approaches have been proposed to capture similarities between textual, visual, and cross-media information. However, most of the existing works use a bag-of-words methods to represent textual information. Since words with different forms may have similar meaning, semantic level text similarities can not be well processed in these methods. To address these challenges, in this paper, we propose a novel method called semantic cross-media hashing (SCMH), which uses continuous word representations to capture the textual similarity at the semantic level and use a deep belief network (DBN) to construct the correlation between different modalities. To demonstrate the effectiveness of the proposed method, we evaluate the proposed method on three commonly used cross-media data sets are used in this work. Experimental results show that the proposed method achieves significantly better performance than state-of-the-art approaches. Moreover, the efficiency of the proposed method is comparable to or better than that of some other hashing methods.

Keyphrases can provide highly condensed and valuable information that allows users to quickly acquire the main ideas.The task of automatically extracting them have received considerable attention in recent decades.Different from previous studies, which are usually focused on automatically extracting keyphrases from documents or articles, in this study, we considered the problem of automatically extracting keyphrases from tweets.Because of the length limitations of Twitter-like sites, the performances of existing methods usually drop sharply.We proposed a novel deep recurrent neural network (RNN) model to combine keywords and context information to perform this problem.To evaluate the proposed method, we also constructed a large-scale dataset collected from Twitter.The experimental results showed that the proposed method performs significantly better than previous methods.

Hashing methods have proven to be useful for a variety of tasks and have attracted extensive attention in recent years. Various hashing approaches have been proposed to capture similarities between textual, visual, and cross-media information. However, most of the existing works use a bag-of-words methods to represent textual information. Since words with different forms may have similar meaning, semantic level text similarities can not be well processed in these methods. To address these challenges, in this paper, we propose a novel method called semantic cross-media hashing (SCMH), which uses continuous word representations to capture the textual similarity at the semantic level and use a deep belief network (DBN) to construct the correlation between different modalities. To demonstrate the effectiveness of the proposed method, we evaluate the proposed method on three commonly used cross-media data sets are used in this work. Experimental results show that the proposed method achieves significantly better performance than state-of-the-art approaches. Moreover, the efficiency of the proposed method is comparable to or better than that of some other hashing methods.