Ninghui Xu

Ninghui Xu

Auto-generated profile, based on publicly available information

Ninghui Xu

Key Laboratory of Micro-Inertial Instrument and Advanced Navigation Technology, Ministry of Education, School of Instrument Science and Engineering, Southeast University, Nanjing, China

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
27 February 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
NXNinghui Xu
Ninghui Xu
Lihui WangLihui Wang
Lihui Wang
JZJiajia Zhao
Jiajia Zhao
ZYZhiting Yao
Zhiting Yao
To effectively denoise the output events of Dynamic Vision Sensor (DVS) in real-time, a denoising method based on augmented spatiotemporal correlation is proposed. In addition, this paper also presents a novel method to evaluate the performance of event denoising. By classifying the events and noise based on their spatiotemporal properties, we use the dynamic response characteristics of DVS under different cases to augment the spatiotemporal correlation, and construct two event filters to process different kinds of events. To benchmark event denoising, we exploit the low-latency of DVS to temporally synchronize raw event stream through motion compensation and obtain the ideal response probability of DVS to quantify the plausibility of events and measure the denoising accuracy. Experiments in various scenes validate that the denoising method is effective at reducing noise with high accuracy and little computational cost, which significantly outperforms the state-of-the-art methods. Moreover, the evaluation method can quantify the denoising accuracy and benchmark different denoising methods objectively and comprehensively.

8 April 2025 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
ZYZhiting Yao
Zhiting Yao
XCXiyuan Chen
Xiyuan Chen
MHMitsuhiro Hayashibe
Mitsuhiro Hayashibe
WZWei Zhu
Wei Zhu
NXNinghui Xu
Ninghui Xu
Collision avoidance is critical for ensuring the safe navigation of unmanned surface vehicles (USVs). This paper presents an end-to-end solution for local path planning of USVs, focusing on enhanced obstacle evasion and smoother navigation. By leveraging deep reinforcement learning (DRL), we enable direct translation of relative distance states into navigational actions, eliminating the need for cumbersome map maintenance and complex feature extraction. A novel observation modality, the “beam map”, is designed to accurately perceive obstacles in all directions, mimicking the functionality of an onboard LiDAR system. To further refine collision avoidance maneuver, a warning zone is introduced, adjusting the agent’s sensitivity to obstacles and allowing ample time and space for decision-making. Additionally, we propose a continuous-time short-distance constraint to calculate the International Regulations for Preventing Collision at Sea (COLREGs) adherence rewards, enabling legal and rational navigation without requiring prior knowledge of the encounter situation. Extensive experimental results, comparing various RL policies and classical methods, demonstrate the planner’s exceptional obstacle avoidance capability and adaptability to changing environments. Using real-world inland ship navigation data, four steering scenarios are designed to further validate the efficacy of the proposed method.

28 September 2023 in IEEE Transactions on Intelligent Vehicles
Verified
Journal Article
Closed access
Lihui WangLihui Wang
Lihui Wang
ZWZhuoxuan Wang
Zhuoxuan Wang
ZYZehua Ying
Zehua Ying
XBXiao Bai
Xiao Bai
NXNinghui Xu
Ninghui Xu
This article investigates a local obstacle avoidance (LOA) problem in terms of the planning efficiency and driving safety of low-speed driving unmanned vehicles. A novel improved weighted heuristic RRT (IWH-RRT) method and an opti-mization strategy are proposed to address the LOA problem with constraints. The sampling space optimization (SPO) and vehicle non-integrity constraint detection (NICD) are first adopted to improve sampling efficiency and guide RRT tree to expand nodes that satisfy driveability towards the target point. Then a cost function combining node attribute and expanding is designed, and the parent node is reselected for the newly expanded ones to reduce the deviation from the reference line and meet the LOA requirements. Finally, the convex space is approximated according to a driving corridor generated based on the initial path, in which the optimized path is obtained by solving an optimal control problem through QP solver. The proposed framework is compared with the baselines and state-of-the-art algorithms in simulation, and real-world experiments are designed afterward. The results verify that our method has good engineering feasibility and real-time capacity in generating smooth, safe and road-shape-liked paths for LOA problems.

26 September 2024 in IEEE Sensors Journal
Verified
Journal Article
Closed access
XTXingbang Tang
Xingbang Tang
XCXianghong Cheng
Xianghong Cheng
NXNinghui Xu
Ninghui Xu
Multi-object tracking (MOT) is an important component of perception system for autonomous vehicles and mobile robots. The 4D millimeter wave (MMW) radar is a novel and promising sensor for MOT for its capability to provide 3D position data and additional Doppler data (3+1D). However, the MOT methods using 4D MMW radar remain scarce, and existing research commonly disregards the potential simplicity of MOT. Moreover, the application of 4D MMW in MOT is constrained by the noise and sparsity of the radar point clouds. To address the above problems, we propose a robust MOT method with a simple modular structure, which integrates 4D MMW radar and monocular vision. The radar data and vision data are first processed in parallel to obtain object detections separately. The detections of the two sensors are then matched and fused using a proposed association method in the camera domain, which can avoid association ambiguities. Finally, a composite structure, mainly consisting of a detection-to-trajectory association module named Fusion-BYTE and a trajectory management module, is designed to exploit the processed sensor detections and yield tracking results. The proposed method is extensively verified on the public View-of-Delft (VoD) dataset and our unmanned aerial vehicle (UAV) dataset. The experimental results demonstrate that our method achieves superior tracking performance compared to the state-of-the-art single-sensor and multi-sensor fusion MOT algorithms, while maintaining high operational efficiency.

31 December 2025 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
ZYZhiting Yao
Zhiting Yao
XCXiyuan Chen
Xiyuan Chen
MHMitsuhiro Hayashibe
Mitsuhiro Hayashibe
NXNinghui Xu
Ninghui Xu
TYTianjian Yuan
Tianjian Yuan
WZWei Zhu
Wei Zhu
The paper proposes a robust collision avoidance framework for USVs navigating high-risk maritime environments without prior assumptions about surrounding obstacles. Current research typically assumes that known static obstacles and dynamic obstacles move in simplified patterns, failing to account for real-world conditions where human-operated vessels exhibit unpredictable and high-dynamic behaviors. To address this, we develop an innovative spatiotemporal feature-encoded navigation policy built on the implicit quantile network. The policy's input is constructed using a sensor-based observation that integrates multiple sensors for environmental data collection, enabling a highly responsive and end-to-end navigation solution. Additionally, by employing rectangle convolutional kernels to extract spatial features and utilizing a continuous-time closed-form network to capture long-term motion dependencies, the proposed method effectively distinguishes between static and dynamic obstacles. The algorithm is compared with one reinforcement learning-based method and three classical methods. The results demonstrate its superior performance in navigating USVs through congested scenarios involving varying ocean flows, aggressive vessels, and static barriers.

25 September 2026
Verified
Book Chapter
Closed access
PDF file
ZGZiren Gong
Ziren Gong
XLXiaohan Li
Xiaohan Li
FTFabio Tosi
Fabio Tosi
NXNinghui Xu
Ninghui Xu
Stefano MattocciaStefano Mattoccia
Stefano Mattoccia
MPMatteo Poggi
Matteo Poggi