Chunxu Li

Chunxu Li

Auto-generated profile, based on publicly available information

Chunxu Li

Mechanical and Electrical Engineering Hohai University, Changzhou, Jiangsu, China

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2016 in Neural Regeneration Research
Verified
Journal Article
Open access
PDF file
THTing-Ting Hou
Ting-Ting Hou
XYXiaoyu Yang
Xiaoyu Yang
JLJian-An Li
Jian-An Li
CZChun-Fang Zan
Chun-Fang Zan
PXPeng Xia
Peng Xia
ZLZhi-Gang Liu
Zhi-Gang Liu
The temporal expression of microRNA after spinal cord ischemia/reperfusion injury is not yet fully understood. In the present study, we established a model of spinal cord ischemia in Sprague-Dawley rats by clamping the abdominal aorta for 90 minutes, before allowing reperfusion for 24 or 48 hours. A sham-operated group underwent surgery but the aorta was not clamped. The damaged spinal cord was removed for hematoxylin-eosin staining and RNA extraction. Neuronal degeneration and tissue edema were the most severe in the 24-hour reperfusion group, and milder in the 48-hour reperfusion group. RNA amplification, labeling, and hybridization were used to obtain the microRNA expression profiles of each group. Bioinformatics analysis confirmed four differentially expressed microRNAs (miR-22-3p, miR-743b-3p, miR-201-5p and miR-144-5p) and their common target genes (Tmem69 and Cxcl10). Compared with the sham group, miR-22-3p was continuously upregulated in all three ischemia groups but was highest in the group with no reperfusion, whereas miR-743b-3p, miR-201-5p and miR-144-5p were downregulated in the three ischemia groups. We have successfully identified the key genes expressed at different stages of spinal cord ischemia/reperfusion injury, which provide a reference for future investigations into the mechanism of spinal cord injury.


13 July 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
ZZZhiyuan Zeng
Zhiyuan Zeng
XGXiongfei Geng
Xiongfei Geng
JWJie Wen
Jie Wen
JLJianan Luo
Jianan Luo
CLChunxu Li
Chunxu Li
YZYue Zheng
Yue Zheng
Low-visibility conditions such as smoke, haze, and dense fog severely degrade the perception and localization capabilities of autonomous systems, yet most existing multi-sensor SLAM methods rely on fixed-weight or hard-switching fusion strategies that cannot adapt to dynamically varying visibility. This manuscript studies a visibility-adaptive multi-sensor fusion SLAM framework that couples a 4D millimeter-wave radar with a LiDAR and a monocular camera. A 4D radar front-end odometry is developed that jointly exploits Doppler velocity, spatial geometry, and Radar Cross Section (RCS) through an improved RCS-weighted correlative scan matching scheme. A lightweight Multi-Scale Spatial-Channel Attention Network (MS-SCANet) is constructed to estimate a continuous environmental visibility score from camera images in real time, using depthwise separable convolutions for multi-scale feature extraction and a spatial-channel dual-attention module for robust degradation representation. The estimated visibility then drives a Cross-Modal Attention Fusion (CMAF) module, which generates sensor-specific weights and embeds them into an Extended Kalman Filter to fuse LiDAR and radar observations for final state estimation. The method is evaluated on a self-collected wheeled-robot dataset covering a dim subterranean tunnel, an outdoor road with patchy smoke, and a smoke-filled indoor corridor, and is further validated on public 4D-radar datasets. Quantitative comparisons against single-sensor and conventional fusion baselines, together with ablation studies, visibility-estimation accuracy analysis, weight-factor analysis, and per-module runtime profiling, show that the proposed method keeps the absolute trajectory error below 0.2m in smoke-filled scenes while running at about 80ms per frame.

13 September 2024
Verified
Conference Paper
Open access
XZXuanming Zhang
Xuanming Zhang
CLChunxu Li
Chunxu Li
BZBing Zhao
Bing Zhao
XLXinyu Li
Xinyu Li
This paper proposes a reinforcement learning based motion control system for ROV underwater robots. In the past, the control system of ROV underwater robots often relied on pre-set path planning and control algorithms, which limited the performance and application scope of underwater robots, making it difficult for ROVs to cope with complex and changing underwater environments. To address this issue, this paper introduces reinforcement learning algorithms to achieve autonomous learning and adaptive control through the interaction learning between robots and their environment. Specifically, this article uses deep reinforcement learning algorithms to train underwater robots, enabling them to autonomously adjust their motion control strategies based on changes in the environment and task requirements. The experimental results show that the reinforcement learning based underwater robot motion control system exhibits better adaptability and robustness in complex underwater environments, with high control accuracy and efficiency.