YJYi Jiang

Yi Jiang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yi Jiang has a total of 18 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Autonomous Parking via Kinematics-Aware Hybrid A* Path Planning and Learning-Driven Trajectory Control

    by Yanbei Duan, Shamsul Arrieya Ariffin, Yi Jiang

    Advances in autonomous driving have made automatic parking systems a practical means of mitigating urban congestion. This work investigates how an autonomous parking system can consistently reconcile global, kinematically feasible trajectory generation with locally robust, smooth, real-time feedback control under dynamic disturbances. To address this question, we present a closed-loop framework that coordinates Hybrid A* planning, Bézier−B-spline smoothing, reinforcement-learning control (PPO/DDPG), and genetic-algorithm-based global refinement to improve parking accuracy and efficiency. Hybrid A* first produces a kinematically feasible path; PPO subsequently adapts the control policy in dynamic environments, and DDPG further refines steering and throttle commands to achieve smoother motion. The GA periodically revises the global trajectory and the associated control parameters, reducing the risk of convergence to suboptimal solutions. A Bézier−B-spline hybrid smoother is finally applied to eliminate residual curvature discontinuities. Experimental evaluations indicate that the proposed system attains a 98.7% parking success rate, an average curvature variation of 15.3[ ], and a mean completion time of 2.1[ ]s — substantially surpassing state-of-the-art baselines. These findings suggest that explicitly coordinating structured planning with learning-based feedback control is a practical route to reliable, real-time autonomous parking.

  • A Hybrid CNN–LSTM Deep Learning Framework for URLLC-Enabled Intelligent V2X Communication Systems

    by Yi Jiang, Shamsul Arrieya Bin Ariffin

    Safe and efficient vehicular networks in modern Intelligent Transportation Systems (ITS) rely heavily on ultra-reliable and low-latency communication (URLLC) as a fundamental requirement. To address this, researchers propose a hybrid deep learning–based Vehicle-to-Everything (V2X) framework that integrates Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) networks to enhance real-time decision-making. The framework initially collects heterogeneous V2X data from Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I), Vehicle-to-Pedestrian (V2P), and Vehicle-to-Network (V2N) sources, including GPS coordinates, vehicle speed, Received Signal Strength Indicator (RSSI), and channel state information. Data preprocessing employs normalization techniques such as Min–Max scaling, sliding window methods, dimensionality reduction, and time-series transformations to generate model-ready inputs. CNN components extract spatial features such as road layouts, vehicle density, and interference patterns, while LSTM modules capture temporal dynamics including signal fluctuations, handover delays, and congestion evolution. The proposed framework was evaluated using a dataset of 10,000 V2X communication samples collected at a sampling frequency of 10 Hz to capture dynamic vehicular network conditions. By fusing spatial and temporal representations, the framework predicts optimal communication strategies, enabling adaptive routing, dynamic power control, and stable link selection. This hybrid approach improves computational efficiency and predictive accuracy, supporting autonomous driving, enhanced traffic management, real-time safety, and congestion forecasting in next-generation V2X networks.

  • Optimized Deep Learning-Based Intrusion Detection System Using SMOTE and Genetic Algorithms

    by Li Xu, Lin Wang, Yi Jiang

    With the increasing sophistication of network attacks, traditional intrusion detection systems (IDSs) face challenges in detecting new attack types, managing high-dimensional network traffic, and responding to adversarial threats. These systems often lack intelligent, efficient solutions. In this paper, we propose an intelligent intrusion detection system that integrates deep learning with optimization algorithms to address these challenges. Specifically, we introduce the synthetic minority over-sampling technique (SMOTE) to handle data imbalance and use genetic algorithms (GA) to optimize the hyperparameters of a deep neural network (DNN). Our results show that this combined approach significantly improves detection accuracy, enhances the detection capability for minority class samples, and increases computational efficiency. Additionally, the system demonstrates robust performance against adversarial attacks and adapts well to complex network environments. This research not only provides a novel method for network security but also lays a practical foundation for developing real-time, efficient IDS solutions.

  • Open Access

    Enhancement study of electronic component size measurement and defect detection based on machine vision

    by Yi Jiang, Shamsul Arrieya Ariffin, Yanbei Duan

    This paper is based on machine vision technology, combined with Halcon software, based on electronic components’ size measurement and defect detection. First, the application status and research significance of machine vision technology in the field of electronic components are introduced. The camera model MV-SUA1201C is used to take the image of the calibration board and then calibrate the industrial camera through HALCON to obtain the relationship between the world coordinates and the pixel coordinates. Then, the electronic component images were pre-processed, such as grayscale conversion, and the preprocessed images were used for size measurement and defect detection. Size measurement is the automatic measurement of electronic components by adding components to the measurement model. Defect detection is to automatically identify the starting position of the pin, detect each pin one by one, and automatically mark the missing position of the pin. The pin detection result is displayed by the upper machine. Finally, the relative error of the visual measurement results is controlled at about 0.8%, and the visual defect detection can automatically detect the pin defect, saving both manpower and time cost.


  • Open Access

    Toward Ultra-Reliable Low-Latency V2X: A Hybrid Deep Learning Approach for Intelligent Vehicular Networks

    by Yi Jiang, Shamsul Arrieya Bin Ariffin

    Safe and efficient vehicular networks in contemporary intelligent transportation systems necessitate ultra-reliable and low-latency communication (URLLC) requirements acting as the base foundation. Researchers combined Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) networks for creating their Hybrid Deep Learning-Based V2X Framework to improve V2X real-time decision-making abilities. The system's first operation phase acquires diverse Vehicle-to-Everything data from V2V, V2I, V2P and V2N sources which contain GPS locations and vehicle speed readings side by side with Received Signal Strength Indicator (RSSI) measurements along with channel status data. The preprocessing method applies normalization strategies (Min-Max Scaling and Sliding Window Method) together with data reduction methods and time-series transformations to create ready-to-use modelling inputs. Through traffic data sources CNN modules decode road layout features and vehicle distributions next to detecting signal interference sequences but LSTM modules analyze signal variations and handover delay effects and identify congested area evolutions. Processor layers integrate both spatial and temporal elements to produce a unified representation that enables predictions for optimal communication standards. The system maintains dependable communication in dense and mobile environments by enabling adaptive routing and dynamic power control along with stable link selection mechanics. The proposed hybrid framework will benefit the next-generation V2X network by achieving computational efficiency alongside predictive accuracy for autonomous driving and smart traffic management functionalities. The proposed hybrid framework boosts the V2X network by ensuring both computational efficiency and predictive accuracy for autonomous driving, enabling improved traffic management. This integration enhances vehicle coordination, real-time safety, and congestion forecasting for future transportation systems.

  • Iterative RANSAC based adaptive birth intensity estimation in GM-PHD filter for multi-target tracking

    by Jingjing Wu, Ke Li, Qiuju Zhang, Wei An, Yi Jiang, Xueliang Ping, Peng Chen


  • Estimation of monthly-mean global solar radiation using MODIS atmospheric product over China

    by Ji-Long Chen, Bei-Bei Xiao, Chun-Di Chen, Zhao-Fei Wen, Yi Jiang, Ming-Quan Lv, Sheng-Jun Wu, Guo-Sheng Li