Inverse kinematics solution of demolition manipulator based on global mapping

Article Computer Science, Hardware & Architecture

Intelligent Diagnosis Using Continuous Wavelet Transform and Gauss Convolutional Deep Belief Network

Huimin Zhao et al.

Summary: This paper proposes a novel method for extracting vibration amplitude spectrum imaging features using continuous wavelet transform and image conversion, as well as a new CDBN for bearing fault classification. The proposed method outperforms traditional methods in performance, as shown in experiments on motor bearing datasets.

IEEE TRANSACTIONS ON RELIABILITY (2023)

Add to Collection

Article Computer Science, Artificial Intelligence

An adaptive differential evolution algorithm based on belief space and generalized opposition-based learning for resource allocation

Wu Deng et al.

Summary: This paper introduces an improved adaptive DE algorithm ACDE/F, which addresses the premature convergence and local optimization issues commonly seen in DE algorithms by incorporating belief space strategy, generalized opposition-based learning strategy, and parameter adaptive strategy. Experimental results demonstrate that ACDE/F outperforms other algorithms in optimization performance and exhibits good performance in practical applications.

APPLIED SOFT COMPUTING (2022)

Add to Collection

Article Automation & Control Systems

Parameter adaptation-based ant colony optimization with dynamic hybrid mechanism

Xiangbing Zhou et al.

Summary: This paper proposes a parameter adaptation-based ant colony optimization (ACO) algorithm called PF3SACO, which combines particle swarm optimization (PSO), fuzzy system, and 3-Opt algorithm to improve the optimization ability and convergence, and avoid falling into local optima. The PF3SACO utilizes dynamic parameter adjustment and adaptive search to achieve better optimization performance, and applies 3-Opt algorithm to optimize the generated path.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2022)

Add to Collection

Proceedings Paper Computer Science, Interdisciplinary Applications

On the Use of Low-discrepancy Sequences in the Training of Neural Networks

E. Atanassov et al.

Summary: Quasi-Monte Carlo methods utilize specially designed deterministic sequences to achieve higher convergence rates compared to random numbers. The usefulness of low-discrepancy sequences, such as Sobol and Halton sequences, has been established in various fields including Mathematical Finance, optimization, and machine learning. This study explores different approaches to efficiently employ low-discrepancy sequences in the training process of neural networks, highlighting their advantage in benchmark problems and discussing practical issues in real-life applications.

LARGE-SCALE SCIENTIFIC COMPUTING (LSSC 2021) (2022)

Add to Collection

Article Automation & Control Systems

An analytical inverse kinematics solution with joint limits avoidance of 7-DOF anthropomorphic manipulators without offset

Xinyang Tian et al.

Summary: The paper proposes an analytical inverse kinematics solution for 7-DOF anthropomorphic manipulators, allowing combined control of global arm configuration and local self-motion. It addresses singularity and joint limits issues, achieving fast calculation speed, high precision, and stability according to simulation results.

JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS (2021)

Add to Collection

Article Automation & Control Systems

An intelligent multi-robot path planning in a dynamic environment using improved gravitational search algorithm

P. K. Das et al.

Summary: This study introduces a new methodology to optimize the trajectory of multi-robots in clutter environment using an improved gravitational search algorithm (IGSA), which enables autonomous decision-making, collaboration, and adaptation to harsh environments.

INTERNATIONAL JOURNAL OF AUTOMATION AND COMPUTING (2021)

Add to Collection

Article Engineering, Electrical & Electronic

A Hyperspectral Image Classification Method Using Multifeature Vectors and Optimized KELM

Huayue Chen et al.

Summary: A feature extraction method combining PCA and LBP is developed to optimize the KELM model for hyperspectral image classification in this article. Experimental results demonstrate that the proposed method achieves higher classification accuracy and better generalization ability for small samples, providing a new idea for hyperspectral image processing.

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING (2021)

Add to Collection

Article Automation & Control Systems