Jia Song - Intelligence theory and applications - Best Researcher Award

School of Astronautics, Beihang University - China

AUTHOR PROFILE

SCOPUS

🎓 ACADEMIC BACKGROUND AND EXPERTISE

Jia Song earned her B.S. degree in Electrical Engineering and Automation, and Intelligent Systems from Harbin Engineering University in 2005, followed by a Ph.D. in Control Theory and Control Engineering from the same institution in 2009. Her profound expertise in these fields laid the foundation for her future endeavors in advanced control systems and intelligent decision-making.

🏛️ PROFESSOR AND DOCTORAL SUPERVISOR

Currently a Professor and doctoral supervisor at the School of Astronautics, Beihang University, Jia Song plays a crucial role in shaping the next generation of engineers and researchers. Her work focuses on advanced flight control systems and collaborative decision-making, where she mentors doctoral students, guiding them through complex research challenges.

🚀 RESEARCH IN ADVANCED FLIGHT CONTROL

Jia Song's research is centered on the development of advanced flight control systems, particularly in high-stakes scenarios involving high dynamic vehicles. Her studies address critical issues such as control precision and margin adequacy in environments with multiple constraints, such as no-fly zones and high-velocity interceptors, making significant strides in the field of aerospace engineering.

🎯 INNOVATIVE PENETRATION GAME STRATEGY

One of her notable contributions is the investigation of penetration game strategies for high dynamic vehicles. Her innovative approach involves an enhanced artificial potential field method that improves lateral penetration guidance strategies, effectively balancing obstacle avoidance and target reachability. This research not only advances theoretical understanding but also offers practical solutions for real-world applications.

🔧 APPLICATION OF ARTIFICIAL INTELLIGENCE AND PREDICTIVE MODELS

Jia Song integrates advanced artificial intelligence techniques, such as the Kalman filter and Transformer network, into her research. These tools are used to denoise detection information and predict multi-step state outcomes, significantly increasing the success rate of high dynamic vehicles in confronting high-velocity interceptors. Her work exemplifies the fusion of AI with traditional engineering to solve complex problems.

📊 NUMERICAL SIMULATIONS AND VALIDATION

Her research is rigorously validated through numerical simulations, which demonstrate the effectiveness and performance of the proposed penetration game guidance strategies. These simulations confirm the practicality and reliability of her methods, ensuring that they can be applied to real-world scenarios with confidence.

🏆 CONTRIBUTIONS TO AEROSPACE ENGINEERING

Jia Song’s contributions to aerospace engineering, particularly in the areas of flight control and collaborative decision-making, are highly regarded in the academic and professional communities. Her innovative research and commitment to excellence continue to push the boundaries of what is possible in the field of astronautics, making her a leading figure in her domain.

NOTABLE PUBLICATION

ADRC-Based Compound Control Strategy for Spacecraft Multi-Body Separation
Authors: Hu, Y., Wu, M., Zhao, K., Song, J., He, B.
Year: 2023
Journal: Aerospace Science and Technology

Survey on Mission Planning of Multiple Unmanned Aerial Vehicles
Authors: Song, J., Zhao, K., Liu, Y.
Year: 2023
Journal: Aerospace

Time-Cooperative Trajectory Optimization Method for Hypersonic Vehicle Based on Improved Grey Wolf Artificial Potential Field Method
Authors: Teng, B., Xu, X., Song, J.
Year: 2023
Conference: 2023 China Automation Congress (CAC 2023)

Fault Location and Separation Method of Distributed Inertial Measurement Units Based on IAC
Authors: Song, J., Shang, W., Wu, B., Ai, S.
Year: 2023
Conference: 2023 10th International Conference on Dependable Systems and Their Applications (DSA 2023)

Real-Time Trajectory Planning for Hypersonic Vehicle with Dynamic No-Fly Zone Constraints
Authors: Xiaowei, X., Jia, S., Kai, Z., Xindi, T., Yanxue, Z.
Year: 2023
Book Series: Lecture Notes in Electrical Engineering (LNEE)

📚 SCIENTIFIC RESEARCH ACHIEVEMENTS

Jiaming Zhong has made significant contributions to the fields of video classification and tactile sensing. His groundbreaking papers include "Exploring Cross-video Matching for Few-shot Video Classification via Dual-Hierarchy Graph Neural Network Learning," published in Image and Vision Computing, and "Text-guided Graph Temporal Modeling for Few-Shot Video Classification," featured in Engineering Applications of Artificial Intelligence. These studies, published in top-tier journals, highlight Zhong's innovative approaches in utilizing graph neural networks and multimodal models for advanced video analysis and classification.

🛠️ PATENTS AND TECHNOLOGICAL INNOVATIONS

Zhong holds several patents that showcase his expertise in developing practical solutions for various technological challenges. His patents include methods for video anomaly classification, chip defect detection, and mobile robot obstacle avoidance. These patents reflect his commitment to translating theoretical research into tangible technological advancements that address real-world problems.

🔬 PROJECT EXPERIENCE: PEEL RECOGNITION

In a project focused on the precise identification of Chenpi years using a multimodal model, Zhong's work involved designing lightweight modules and fine-tuning models to achieve high recognition accuracy. His use of the CLIP multimodal model for feature extraction led to a remarkable 99% accuracy in recognizing Chenpi years with limited sample data. This project, detailed on GitHub, demonstrates his proficiency in applying advanced machine learning techniques to practical problems.

🎥 PROJECT EXPERIENCE: FEW-SHOT VIDEO CLASSIFICATION

Zhong's research in video behavior classification involved addressing challenges related to data scarcity and model capabilities. Collaborating with Macau University of Science and Technology and Wuyi University, he developed a dual-hierarchy graph neural network that significantly improved classification performance through cross-video frame matching. This innovative approach was published in Image and Vision Computing and showcased Zhong's ability to enhance model performance through sophisticated temporal modeling.

🔍 PROJECT EXPERIENCE: MULTIMODAL REPRESENTATION LEARNING

In a project focused on multimodal video behavior analysis, Zhong led efforts to develop a novel framework for self-supervised learning using multimodal data. This project, supported by a 500,000 RMB research grant, involved developing a text-guided feature optimization module and a query text token learning mechanism. His research aimed to leverage multimodal knowledge to improve the classification performance of few-shot video behaviors, with results published in top journals.

📈 IMPACTFUL RESEARCH AND PUBLICATIONS

Zhong's work has significantly impacted the fields of video classification and sensor technology. His papers in renowned journals and his patents contribute to advancing the understanding and application of these technologies. His research not only addresses current challenges but also paves the way for future innovations in these areas.

NOTABLE PUBLICATION

Ultra-sensitive and stable All-Fiber iontronic tactile sensors under high pressure for human movement monitoring and rehabilitation assessment
Authors: K. Ma, D. Su, B. Qin, Y. Xin, X. He
Year: 2024
Journal: Chemical Engineering Journal

Real-time citrus variety detection in orchards based on complex scenarios of improved YOLOv7
Authors: F. Deng, J. Chen, L. Fu, J. Li, N. Li
Year: 2024
Journal: Frontiers in Plant Science

Exploring cross-video matching for few-shot video classification via dual-hierarchy graph neural network learning
Authors: F. Deng, J. Zhong, N. Li, D. Wang, T.L. Lam
Year: 2023
Journal: Image and Vision Computing