Special Session 14 会议特别专题 14
Intelligent Target Tracking and Information Fusion
Description:
The evolution of surveillance and situational
awareness systems is entering a transformative
phase, driven by the convergence of artificial
intelligence, advanced sensor fusion, and autonomous
networking. This special session focuses on the core
technologies enabling the next generation of
intelligent perception systems, which must deliver
persistent, accurate, and robust tracking in
increasingly complex, dynamic, and contested
environments. Moving beyond traditional
single-sensor, single-target paradigms, the future
lies in seamlessly integrating heterogeneous data
streams and enabling collaborative reasoning across
distributed platforms.
The design and implementation of such intelligent
tracking and fusion systems present fundamental
research challenges that require a holistic
approach. Below, we highlight the critical
motivations and technological imperatives driving
this field:
(1) The AI Revolution in Target Tracking:
Classical tracking algorithms often struggle with
complex target behaviors, dense clutter, and
occlusions. The integration of machine learning and
deep learning offers a paradigm shift, enabling
predictive tracking, adaptive model learning, and
sophisticated behavior recognition. Research into
AI-driven filters, reinforcement learning for
decision-making, and end-to-end learning frameworks
is crucial to achieving the required levels of
autonomy and robustness.
(2) The Imperative of Multimodal Large Model Fusion:
Modern systems employ a diverse array of sensors
(e.g., radar, EO/IR, LiDAR, SIGINT, acoustic). The
key challenge is no longer just data association but
semantic-level fusion. Leveraging large foundation
models to create a unified, contextual understanding
from multimodal, heterogeneous data streams is a
frontier research area. This enables higher-level
inference, intent prediction, and significantly
enhanced situational awareness beyond simple
kinematic fusion.
(3) The Shift to Distributed Autonomous Cooperative
Surveillance:
Centralized architectures are vulnerable and lack
scalability. The future is in decentralized networks
of autonomous agents (UAVs, ground robots, static
sensors) that collaboratively perceive, track, and
share information. This demands breakthroughs in
distributed consensus algorithms,
communication-efficient data fusion, resource-aware
task allocation, and swarm intelligence to achieve
resilient and scalable collaborative surveillance.
(4) The Emergence of Embodied AI and Multi-Sensor
Fusion:
For autonomous systems like robots and intelligent
vehicles to operate effectively in the physical
world, perception must be tightly coupled with
action. Embodied AI research focuses on how agents
learn through interaction with their environment.
Multi-sensor fusion in this context is critical for
building robust embodied perception, enabling
real-time understanding, navigation, and
manipulation in unstructured settings. This theme
explores the synergy between sensor fusion
algorithms and embodied intelligence for
next-generation autonomous systems.
(5) Addressing the Critical Challenge of
Low-Slow-Small (LSS) Targets:
Drones, birds, and ultralights pose a unique and
persistent threat due to their small radar
cross-section, low altitude, and slow speed.
Detecting, classifying, and continuously tracking
these elusive targets require specialized sensor
modalities, novel signal processing techniques, and
fusion strategies that can separate them from
clutter and environmental noise. This domain tests
the limits of current tracking and fusion
technologies.
(6) The Drive for Real-Time, Trustworthy, and
Explainable Systems:
As systems become more autonomous, the need for
real-time processing, system resilience against
deception, and explainable AI outputs becomes
paramount. Research into efficient algorithms for
edge computing, adversarial robustness in fusion
models, and interpretable decision-making processes
is essential for deploying trustworthy systems in
critical applications.
(7) Global Research Synergy and Application Pull:
This field sits at the intersection of signal
processing, control theory, robotics, computer
vision, and machine learning. There is strong
synergy with global efforts in autonomous systems,
smart cities, perimeter security, and airspace
management. The session aligns with these expansive
research currents, aiming to bridge theoretical
advancements with pressing practical needs.
This special session provides a dedicated forum to
address these interconnected challenges. By bringing
together researchers from academia, industry, and
defense sectors, we aim to foster discussions on
innovative theories, algorithms, and system
implementations. Contributions are expected to push
the boundaries of intelligent tracking and fusion,
paving the way for autonomous systems capable of
operating in the most demanding scenarios. We invite
submissions on topics including but not limited to
AI/ML-enhanced tracking filters, multimodal fusion
architectures, distributed fusion algorithms, sensor
resource management, embodied AI for perception, and
practical system evaluations for LSS and other
challenging targets.
Session organizers
Assoc. Prof. Weihua Wu, Air Force Early Warning
Academy, China
Assoc. Prof. Kai Dong, Naval Aviation University,
China
The topics of interest include, but are not limited to:
▪ AI/ML-driven target
tracking and prediction algorithms
▪ Multimodal (radar, EO/IR, LiDAR, etc.) information
fusion
▪ Large foundation models for situational
understanding and data fusion
▪ Distributed, decentralized, and collaborative
tracking architectures
▪ Embodied AI and Multi-Sensor Fusion for autonomous
systems
▪ Detection, classification, and tracking of
Low-Slow-Small (LSS) targets
▪ Sensor resource management and scheduling for
tracking
▪ Adversarial robustness and secure information
fusion
▪ Track-before-detect and weak signal tracking
techniques
▪ Integration of tracking with automatic target
recognition (ATR)
▪ Real-time embedded systems and hardware
implementations for tracking
▪ Performance evaluation and metrics for intelligent
tracking systems
Submission method
Submit your Full Paper (no less than 5 pages with two colums) or
your paper abstract-without publication (200-400
words) via
Online Submission System, then choose Special
Session 14 (Intelligent Target Tracking and Information Fusion)
Template Download
Introduction of
Session organizers
Assoc. Prof. Weihua Wu,
Air Force Early Warning Academy, China
Weihua Wu received the B.S., M.S., and Ph.D. degrees
from the Air Force Early Warning Academy (AFEW),
Wuhan, China, in 2009, 2011, and 2015, respectively.
He is currently an Associate Professor with AFEW.
His research interests include target tracking and
information fusion, with a focus on AI-driven
methods.
He is a selected young talent in the Youth Support
project of the China Association for Science and
Technology and a recipient of the National Excellent
Doctoral Dissertation Award. He has led research
projects funded by the National Natural Science
Foundation of China (NSFC). He has published over 40
papers in international journals and conferences,
including IEEE Transactions on Signal Processing
(IEEE TSP). He is the author of several influential
books in the field, including Target Tracking With
Random Finite Sets, Moving Sensor Target Tracking
Technology, and Multi-Sensor Data Fusion.
Assoc. Prof. Kai Dong,
Naval Aviation University, China
Associate Prof. Kai Dong received the Ph.D. degree
from the Naval Aviation University(NAU), Yantai,
China, in 2014. He once worked as a postdoctoral
researcher at Beihang University, and a visiting
scholar at Tsinghua University. Currently he is an
Associate Professor with Information Fusion
Institute of NAU. He is a selected young talent in
the “Youth Support” project of the China Association
for Science and Technology and recognized as a
ministerial-level young scientific and technological
talent. He has been funded by the Postdoctoral
Science Foundation of China, and has published over
40 papers in journals and conferences. He is the
co-author of Multi-source Information Fusion and
Application(Third Edition), which is an influential
book in the field. His research interests includes
radar target tracking, multi-source information
fusion, etc.