MJMeng Jia

Meng Jia

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Meng Jia has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Improved DDPG ‐Based Path Planning for Mobile Robots

    by Xianyong Ruan, Du Jiang, Juntong Yun, Bo Tao, Yuanmin Xie, Baojia Chen, Meng Jia, Li Huang

    With the rapid advancement of robotics technology, path planning has attracted extensive research attention. Reinforcement learning, owing to its ability to acquire optimal policies through continuous interaction with the environment, offers a promising solution for path planning in environments with incomplete or unknown information. However, reinforcement learning‐based path planning methods often suffer from high training complexity and low utilization of effective samples. To address these issues, this paper proposes an improved deep reinforcement learning (DRL) algorithm. The proposed approach builds upon the deep deterministic policy gradient (DDPG) algorithm and incorporates a short‐term goal planning strategy based on local perceptual information, which decomposes the global navigation task into multiple short‐term subgoals, thereby reducing task complexity and enhancing learning efficiency. Furthermore, a reward function integrating the artificial potential field (APF) method is designed to improve obstacle avoidance capability. To tackle the low utilization of effective experiences in DDPG, a dual experience pool strategy is introduced to improve experience utilization efficiency and accelerate model training. The parameters for short‐term goal selection are optimized through multiple comparative experiments, and the proposed method is evaluated against several DRL‐based path planning approaches in a static environment. Experimental results demonstrate that the improved algorithm significantly accelerates convergence. Moreover, dynamic environment simulation experiments verify that the proposed algorithm can effectively avoid moving obstacles and achieve safe navigation to the target position.


  • Lightweight Gesture Recognition Based on Depthwise Separable Convolution and FECAM Attention Mechanism for sEMG

    by Haozhu Wang, Du Jiang, Juntong Yun, Li Huang, Yuanmin Xie, Baojia Chen, Meng Jia, Ying Sun

    Surface electromyography (sEMG) is a promising approach for non-invasive gesture recognition in human – computer interaction and rehabilitation. However, existing high-accuracy models often incur high computational costs, thereby limiting real-time deployment. To address this, we propose FSGR-Net, a lightweight residual network that reconstructs ResNet50 using a small-convolution stacking strategy and a Lite-Fusion Block. The Lite-Fusion Block integrates Depthwise Separable Convolution (DSC), Ghost Convolution, and a channel compression – expansion mechanism to reduce redundancy. In particular, a Frequency-Enhanced Channel Attention Mechanism (FECAM) is introduced after DSC layers to enhance discriminative features while mitigating the Gibbs phenomenon. Furthermore, a joint data augmentation strategy—time-shifting and masking—is applied to improve generalization. Evaluations on NinaPro DB1, DB5, and our SC-Myo Datasets show that FSGR-Net achieves 93.17%, 87.83%, and 93.35% accuracy, respectively, with only 0.85M parameters and 0.22G FLOPs, demonstrating strong potential for deployment in mobile and low-power wearable systems.


  • Open Access

    Fast Noise Level Estimation via the Similarity within and between Patches

    by Jiaxin Wu, Meng Jia, Shiqian Wu, Shoulie Xie

    Patch level-based noise level estimation (NLE) is often inaccurate and inefficient because of the harsh criteria required to select a small number of homogeneous patches. In this paper, a fast image NLE method based on a global search for similar pixels is proposed to solve the above problem. Specifically, the mean square distance (MSD) is first expressed in the form of the standard deviation (std) and mean value of image patches. Afterward, the two values, std and mean, are calculated and stored in advance. Then, a 2D statistical histogram and summed area table are adopted to speed up the search for similar patches. Further, the most similar pixels are selected from similar patches to obtain an initial estimation. Finally, we correct the deviation of the initial estimation by re-injecting noise for secondary estimation. Experimental results show that the proposed method outperforms the state-of-the-art techniques in fast NLE and guided denoising.


  • Computer data processing method of map manifold based on derivative transfer approach

    by Meng Jia, Zheren Sun

    Summary: With the rapid development of Internet technology and the rapid upgrade of hardware devices such as computers and smartphones, the data processed by computers often presents a high dimension. How to reduce data dimension and improve data processing speed is a hot research topic in the computer field. One shortcoming of one‐dimensional manifold calculation methods via sequence points is that the position of new points on the manifold is determined by searching the existing manifold by dichotomy, which is tedious and time‐consuming. This paper improved this method based on derivative transfer. According to the calculation results of one‐dimensional stable and unstable manifolds, they exhibited a special property, that is, for any point on the manifold orbit, its image and preimage must also be on the manifold orbit. Based on this, the derivative transfer method was used to improve the calculation of one‐dimensional manifolds for the map, and a three‐step calculation method of “prediction‐correction‐accuracy verification” was proposed. Then this method was applied to the simulation of high‐dimensional manifolds. The simulation results revealed that the proposed method refined the manifold calculation and greatly enhanced the computation speed.

  • Open Access

    Micro-Shaping of Pure Aluminum in Long-Duration Wire Electrochemical Micromachining Using Bipolar Nanosecond Pulses

    by Xiaolei Bi, Meng Jia, Lingchao Meng

    With the increasing application of three-dimensional pure aluminum microstructures in micro-electromechanical systems (MEMS) and for fabricating terahertz components, high-quality micro-shaping of pure aluminum has gradually attracted attention. Recently, high-quality three-dimensional microstructures of pure aluminum with a short machining path have been obtained through wire electrochemical micromachining (WECMM), owing to its sub-micrometer-scale machining precision. However, machining accuracy and stability decrease owing to the adhesion of insoluble products on the surface of the wire electrode in long-duration WECMM, which limits the application of pure aluminum microstructures with a long machining path. In this study, the bipolar nanosecond pulses are used to improve the machining accuracy and stability in long-duration WECMM of pure aluminum. A negative voltage of −0.5 V was considered appropriate based on experimental results. Compared with the traditional WECMM using unipolar pulses, the machining accuracy of the machined micro-slit and the duration of stable machining were significantly improved in long-duration WECMM using bipolar nanosecond pulses.


  • Computational investigation of MnIr and FeIr electrocatalyst for nitrogen reduction reaction

    by Pengfei Ma, Chensi Li, Wei Zhang, Meng Jia, Wei Song

    NH3 is not only an important component of agricultural and industrial production, but also an extremely promising energy carrier and storage intermediate. Currently, the Haber‐Bosch process used in industry for NH3 production has shortcomings such as high energy consumption and low output. The electrocatalytic nitrogen reduction reaction (NRR) can improve the route and conditions of NH3 synthesis through high‐efficient electrocatalyst, and realize the production mode of high efficiency and low energy consumption. Therefore, the design and synthesis of the NRR electrocatalysts with high catalytic performance are very important. Here, the first principles calculation based on density functional theory was used to form alloy catalysts by using Mn and Fe atoms instead of nine Ir atoms on the surface of Ir(100), and the electrocatalytic performance of the NRR was systematically studied. The results showed that N2 could be stably adsorbed on Mn9@Ir(100) and Fe9@Ir(100) in the side‐on configuration. The possible reaction pathways were analyzed and discussed, and the enzymatic pathway was determined to be the best. Through the simulation of the entire NRR process, it was found that the limit potential was only −0.659 and −0.647 V for Mn9@Ir(100) and Fe9@Ir(100). In addition, the electronic properties of Mn9@Ir(100) and Fe9@Ir(100) were analyzed utilizing charge density difference and density of states, and the reasons for their high activity were obtained. We hope this work can not only reduce the number of noble metals and develop highly active catalysts, but also provide theoretical support and guidance for the catalytic mechanism of alloy electrocatalysts.


  • Open Access

    Low-level wind shear of wind field modeling and simulation

    by Ye Zhang, Meng Jia

    Three wind shear wind field modeling methods has been introduced in this paper. By use of computer numerical simulation, the model can not only simulate the real wind shear of wind field, buy can also provide other useful physical quantities. In the study of low-level wind shear detection system process, the wind field model that can truly reflect the physical characteristics change of wind field is establised. The experiments results show that the most typical shear wind field is caused by the down flow.


  • Open Access

    Image Encryption Based on High-Dimensional Manifold Computing and Block Dividing Algorithm

    by Meng Jia

    The image encryption schemes developing more sensitive and more chaotic maps are used as a key sequence generator, such as cascade chaotic maps and high-dimension maps. High-dimension chaotic maps can generate sequences with little correlation after hundred times of iteration. In fact, the sequence is just a flow of the manifold of the chaotic system. A fast way to select sensitive flows of the dynamic systems is introduced in the paper. Combining with the divided blocks diffusion algorithm, the novel scheme yields the cipher image more randomly. Experimental results show that the sensitive flows of the high-dimension system can generate a series with better ergodicity and less correlation compared with the common flows of the same system. The encryption efficiency is enhanced by choosing the sensitive flows of the high-dimension system. The analysis proved that the novel image encryption scheme can resist all common kinds of attacks.