Given its importance, fault diagnosis has attracted considerable attention in the literature, and several machine learning methods have been proposed to discover the characteristics of different aspects in fault diagnosis. In this paper, we propose a Hybrid Deep Belief Network (HDBN) learning model that integrates data in different ways for intelligent fault diagnosis in motor drive systems, such as a vehicle drive system. In particular, we propose three data fusion methods: data union, data join, and data hybrid, based on detailed data fusion research. Additionally, the significance of the fusion is explained from the energy perspective of the signal. In particular, the appropriate fusion methods and data structures suitable for model training requirements can help improve the accuracy of fault diagnosis. Moreover, mixed-precision training is used as a special fusion method to further improve the performance of the model. Experiments with the datasets obtained from the simulation platform demonstrate the superiority of our proposed model over the state-of-the-art methods.
In view of the future lack of human resources due to the aging of the population, the automatic, Intelligent Mechatronic Systems (IMSs) and Intelligent Transportation Systems (ITSs) have broad application prospects. However, complex application scenarios and limited open design resources make designing highly efficient ITS systems still a challenging task. In this paper, the optimal load factor solving solution is established. By converting the three user requirements including working distance, time and load into load-related factors, the optimal result can be obtained among system complexity, efficiency and system energy consumption. A specialized visual navigation and motion control system has been proposed to simplify the path planning, navigation and motion control processes and to be accurately calculated in advance, thereby further improving the efficiency of the ITS system. The validity of the efficiency calculation formula and navigation control method proposed in this paper is verified. Under optimal conditions, the actual working mileage is expected to be 99.7%, and the energy consumption is 83.5% of the expected value, which provides sufficient redundancy for the system. In addition, the individual ITS reaches the rated operating efficiency of 95.86%; in other words, one ITS has twice the ability of a single worker. This proves the accuracy and efficiency of the designed ITS system.
25 March 2022 in ACM Transactions on Intelligent Systems and Technology
Crime poses a major threat to human life and property, which has been recognized as one of the most crucial problems in our society. Predicting the number of crime incidents in each region of a city before they happen is of great importance to fight against crime. There has been a great deal of research focused on crime prediction, ranging from introducing diversified data sources to exploring various prediction models. However, most of the existing approaches fail to offer fine-scale prediction results and take little notice of the intricate spatial-temporal-categorical correlations contained in crime incidents. In this article, we propose a tailor-made framework called CrimeTensor to predict the number of crime incidents belonging to different categories within each target region via tensor learning with spatiotemporal consistency. In particular, we model the crime data as a tensor and present an objective function which tries to take full advantage of the spatial, temporal, and categorical correlations contained in crime incidents. Moreover, a well-designed optimization algorithm which transforms the objective into a compact form and then applies CP decomposition to find the optimal solution is elaborated to solve the objective function. Furthermore, we develop an enhanced framework which takes a set of pre-selected regions to conduct prediction so as to further improve the computational efficiency of the optimization algorithm. Finally, extensive experiments are performed on both proprietary and public datasets and our framework significantly outperforms all the baselines in terms of each evaluation metric.
In High Efficiency Video Coding, the Residual Quad-Tree (RQT) coding is used to encode the prediction residual for both intra and inter Coding Units (CU) and provides improved coding gains. However, this results in much higher computational complexities. To address this problem, we develop two fast RQT algorithms for intra- and inter-prediction residual coding respectively. For intra coding, the proposed algorithm selects the best prediction mode in the rate distortion mode decision process using a Prediction Unit (PU) size-dependent fast RQT depth decision on a reduced prediction mode candidates set from the rough mode decision process. For inter coding, in addition to CU size-dependent fast RQT depth decisions, we propose a discriminant analysis-based fast depth decision algorithm to determine the best transform unit size. Experimental results show that on average, we can realise a 21.29% encoding time saving and 0.03% bit-rate reduction for intra coding, while 15% of the encoding time can be saved with a negligible coding performance loss for inter coding.
Articulated robot is now widely employed in manufacture, such as, welding, painting, and assembly, with high precision and endurance. It plays an important role in scientific and technological innovation. Trajectory planning of articulated robot is one of the key researches in industrial robot. The commonly used trajectory planning algorithm of articulated robot, such as, polynomial interpolation algorithm in joint space and linear interpolation in Cartesian space are introduced. Researches on articulated robot trajectory planning are surveyed. Meanwhile these articulated robot trajectory planning algorithms are analysed. Some further researches and developing trend of articulated robot trajectory planning are indicated.