Target Tracking Research Articles

AI-Driven Adaptive Radar Systems for Real-Time Target Tracking in Urban Environments

Radar systems play a crucial role in target tracking within urban environments, where challenges such as clutter, multipath effects, and electromagnetic interference significantly impact detection accuracy. Traditional radar methods often struggle to adapt to dynamic urban conditions, leading to decreased reliability in real-time target tracking. This study aims to develop and evaluate an AI-driven adaptive radar system that enhances tracking accuracy in urban settings. The research employs a quantitative approach using simulations to model radar signal processing under various environmental conditions. The AI model, based on Convolutional Neural Networks (CNN), is trained to optimize radar performance by filtering out noise and dynamically adjusting detection parameters. The results indicate that the AI-based radar system achieves a tracking accuracy of 95.2%, significantly outperforming traditional radar systems, which only reach 80% accuracy. Additionally, the AI-enhanced radar reduces response time to 120 milliseconds, compared to 250 milliseconds in conventional systems, demonstrating improved real-time processing capabilities. The system also exhibits greater resilience to high-clutter environments, maintaining stable target detection despite signal interference. These findings highlight the potential of AI in enhancing radar functionality for applications such as surveillance, traffic monitoring, and security. Future research should focus on integrating AI-driven radar with real-world radar hardware, exploring multi-sensor fusion, and refining adaptive learning techniques to further optimize tracking performance in complex environments

Abstract 4478: Persistent luminescent nanoparticles for autofluorescence-free tumor imaging: A novel approach to enhanced cancer diagnostics

Abstract Background: Persistent luminescent nanoparticles (PLNs) represent a significant advancement in tumor imaging by providing autofluorescence-free imaging capabilities. Unlike conventional fluorescent probes, PLNs emit light for an extended period after excitation, enabling high-sensitivity imaging without interference from surrounding tissue autofluorescence. Despite their potential, commonly used PLNs suffer from low photoluminescence quantum yield (PLQY), limiting their brightness and imaging efficiency. This challenge necessitates the development of novel materials and synthesis strategies to enhance PLNs' optical properties and stability for bioimaging applications. Method: PLNs were developed Mn2+-doped Zn2GeO4 (ZGOM) nanorods through a modified hydrothermal synthesis method. The approach was designed to optimize crystal structure, doping efficiency, and luminescent properties. The nanorods were characterized using dynamic lightscattering (DLS), transmission electron microscopy (TEM), X-ray diffraction (XRD), andphotoluminescence spectroscopy to evaluate particle size, morphology, crystallinity, and opticalperformance. Their photoluminescence quantum yield (PLQY) was measured, and stability studies were conducted in biologically relevant media. Additionally, in vitro studies were performed to assess biocompatibility and imaging efficacy in cancer cell models. Results: The synthesized ZGOM nanorods exhibited a highly uniform rod-like morphology with an average size below 100 nm. Under UV irradiation, the nanorods displayed intense green luminescence with a quantum yield exceeding 50%, demonstrating significant improvement over conventional PLNs. Stability studies confirmed excellent photostability and minimal degradation in aqueous and biological media. In vitro experiments showed efficient cellular uptake and bright, sustained luminescence, enabling high-resolution imaging of cancer cells. Furthermore, the absence of autofluorescence interference in surrounding tissues highlighted the potential of these nanorods as precise imaging agents for tumor targeting and cell tracking. Conclusion: The Mn2+-doped Zn2GeO4 nanorods synthesized in this study exhibit enhanced optical properties, including high PLQY and photostability, making them a promising candidate for autofluorescence-free tumor imaging. The combination of improved synthesis methods and superior material properties addresses the primary limitations of conventional PLNs, paving the way for their application in advanced bioimaging. These findings establish ZGOM nanorods as aviable platform for cancer diagnosis, with potential for further optimization and translation intoclinical imaging applications. Citation Format: Bhupendra B. Srivastava, Swati Mohan, Murali M. Yallapu, Subhash C. Chauhan. Persistent luminescent nanoparticles for autofluorescence-free tumor imaging: A novel approach to enhanced cancer diagnostics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4478.

Review of UAV based on Reconfigurable Intelligent Surface and ISAC

This paper discusses the application of Reconfigurable Intelligent Surface (RIS) technology combined with Unmanned Aerial Vehicle (UAV) technology, and discusses their innovative potential in 6G communication networks. Especially in the field of Integrated Sensing and Communication (ISAC) application prospects. RIS is a two-dimensional surface based on electromagnetic metamaterial technology. Through dynamic regulation of electromagnetic units, parameters such as phase and amplitude of incident electromagnetic waves can be flexibly controlled, so as to optimize wireless channel and improve communication performance. Compared with traditional communication systems, RIS has the advantages of low cost, low power consumption and easy deployment, which can significantly improve the stability and coverage of UAS communication links. Through in-depth analysis of the working principle, hardware architecture and advantages of RIS combined with UAV, this paper elaborates its application scenarios in UAV network, including disaster relief, urban management, intelligent transportation and battlefield reconnaissance. With RIS’s intelligent beamforming and channel optimization, UAVs can not only provide efficient communication services, but also enable accurate environment awareness and target tracking. In addition, this paper analyzes the key challenges faced by RIS-assisted UAVs in practical engineering applications, such as the complexity of channel estimation, multi-UAV cooperative communication, and adaptability in dynamic environments. By proposing several solutions, including optimizing channel estimation technology and improving beamforming design, this paper provides theoretical basis and technical support for the wide application of RIS and UAV in 6G networks in the future. The research shows that the introduction of RIS technology will greatly improve the communication ability of UAVs in complex environments, and promote the integration of communication and perception in the 6G era.

Multi-UAV collaborative trajectory planning for non-cooperative target tracking in complex environment

Multi-UAV collaborative trajectory planning for non-cooperative target tracking in complex environment

Gaussian Belief Propagation-Based Multiview Multiextended Target Tracking With Occlusion

Gaussian Belief Propagation-Based Multiview Multiextended Target Tracking With Occlusion

A Novel Set-Member Filter for Maneuver Non-Ellipsoid Extended Target Tracking With Unknown but Bounded Noises

A Novel Set-Member Filter for Maneuver Non-Ellipsoid Extended Target Tracking With Unknown but Bounded Noises

Fusion of Kalman filter to improve the KFMS algorithm and its application in parallel robots

An improved Mean Shift target tracking algorithm integrated with the Kalman Filter (KFMS algorithm) is proposed to address the challenge of accurately tracking targets in complex environments using parallel robots. This algorithm combines the local search capability of the Mean Shift algorithm with the state estimation capability of the Kalman Filter, reducing the impact of complex environments on tracking performance. It achieves accurate global tracking of targets by the camera, effectively improving the stability and accuracy of dynamic target tracking. Experimental results with a local camera show that the target tracking accuracy and success rate of the improved KFMS algorithm are increased by 21.3% and 29.6%, respectively, compared to the Mean Shift algorithm, further validating the effectiveness of the enhanced algorithm.

Near-Infrared Hyperspectral Target Tracking Based on Background Information and Spectral Position Prediction

In order to address the problems of in-plane rotation and fast motion during near-infrared (NIR) video target tracking, this study explores the application of capsule networks in NIR video and proposes a capsule network method based on background information and spectral position prediction. First, the history frame background information extraction module is proposed. This module performs spectral matching on the history frame images through the average spectral curve of the groundtruth value of the target and makes a rough distinction between the target and the background. On this basis, the background information of history frames is stored as a background pool for subsequent operations. The proposed background target routing module combines the traditional capsule network algorithm with spectral information. Specifically, the similarity between the target capsule and the background capsule in the spectral feature space is calculated, and the capsule weight allocation mechanism is dynamically adjusted. Thus, the discriminative ability of the target and background is strengthened. Finally, the spectral information position prediction module locates the center of the search region in the next frame by fusing the position information and spectral features of adjacent frames with the current frame. This module effectively reduces the computational complexity of feature extraction by capsule networks and improves tracking stability. Experimental evaluations demonstrate that the novel framework achieves superior performance compared to current methods, attaining a 70.3% success rate and 88.4% accuracy on near-infrared (NIR) data. Meanwhile, for visible spectrum (VIS) data analysis, the architecture maintains competitive effectiveness with a 59.6% success rate and 78.8% precision.

Parallelized Monte Carlo Optimization with Efficient Pipelining on an FPGA for Onboard Ballistic Target Tracking

Parallelized Monte Carlo Optimization with Efficient Pipelining on an FPGA for Onboard Ballistic Target Tracking

Video Tracking of Targets for Vibration Measurement of Large‐Scale Structures Under Seismic Excitation

ABSTRACTDevelopment of video‐based vibration techniques and reliability of displacement measurements have important implications in several areas, including structural testing and monitoring of large structures. Through the use of numerous targets placed on the structure, 3‐dimensional (3D) analysis, and modal identification can be performed with minimal installation costs, no mass increase, and contactlessly. This study applies and validates video tracking of targets for vibration analysis of a large‐scale structure, namely, a reinforced concrete U‐shaped wall, tested experimentally under seismic excitation on a shake table. 2D grids of targets are placed on two orthogonal surfaces and filmed by several cameras. Values obtained from a motion capture system, digital image correlation, and traditional displacement sensors are compared. Results confirm that the 3D motion of the U‐shaped wall surfaces and foundation can be accurately extracted from video analysis.

Robust UAV Target Tracking Algorithm Based on Saliency Detection

Due to their high efficiency and real-time performance, discriminant correlation filtering (DCF) trackers have been widely applied in unmanned aerial vehicle (UAV) tracking. However, the robustness of existing trackers is still poor when facing complex scenes, such as background clutter, occlusion, camera motion, and scale variations. In response to this problem, this paper proposes a robust UAV target tracking algorithm based on saliency detection (SDBCF). Using saliency detection methods, the DCF tracker is optimized in three aspects to enhance the robustness of the tracker in complex scenes: feature fusion, filter-model construct, and scale-estimation methods improve. Firstly, this article analyzes the features from both spatial and temporal dimensions, evaluates the representational and discriminative abilities of different features, and achieves adaptive feature fusion. Secondly, this paper constructs a dynamic spatial regularization term using a mask that fits the target, and integrates it with a second-order differential regularization term into the DCF framework to construct a novel filter model, which is solved using the ADMM method. Next, this article uses saliency detection to supervise the aspect ratio of the target, and trains a scale filter in the continuous domain to improve the tracker’s adaptability to scale variations. Finally, comparative experiments were conducted with various DCF trackers on three UAV datasets: UAV123, UAV20L, and DTB70. The DP and AUC scores of SDBCF on the three datasets were (71.5%, 58.9%), (63.0%, 57.8%), and (72.1%, 48.4%), respectively. The experimental results indicate that SDBCF achieves a superior performance.

Performance Analysis of a Small-Scale Dynamic Platform for Target Tracking Using Enlarged Template Matching

Abstract Vision-based navigation is essential for mobile robotics in dynamic environments. This study develops a four-wheeled robotic platform with an ESP32 camera, Arduino Uno, and mecanum wheels for omnidirectional movement, enabling vision-guided target tracking. The system employs an enlarged template matching method, where image processing is performed on a user computer using a Python-based program. Tracking performance was evaluated using a checkerboard-patterned accuracy board under four template scale values (M = 1.5, 2, 2.5, and 3). Results show that template scale significantly affects tracking accuracy and processing time. The optimal scale (M = 2) achieved 100% accuracy with an average processing time of 24.95 seconds, while other scales exhibited reduced accuracy due to spatial information loss or noise. This study highlights the effectiveness of enlarged template matching for precise navigation and suggests adaptive template scaling for improved performance. The findings contribute to advancing low-cost, vision-based navigation for mobile robotics.

Fully Distributed Fuzzy Adaptive Output Time‐Varying Formation Tracking for Heterogeneous Nonlinear Multiagent Systems

ABSTRACTThis article investigates the fully distributed output time‐varying formation tracking issue for heterogeneous nonlinear multiagent systems on directed graphs, in which the tracking target is a general linear dynamic leader while the followers are composed of first‐order and second‐order dynamic individuals with unknown nonlinearities and velocity measurement. To solve the problem of computational complexity explosion and different order of heterogeneous systems, a fully distributed dynamic compensator is presented. Then, an adaptive observer is designed using fuzzy logic systems to estimate the unknown velocity information. With reference to these observers, a fuzzy adaptive output formation tracking control protocol is designed. Lyapunov stability theory is used to show that the output formation tracking error is uniformly ultimately bounded. Finally, numerical simulation is given to verify the feasibility of the theory.

From Camera Image to Active Target Tracking: Modelling, Encoding and Metrical Analysis for Unmanned Underwater Vehicles

From Camera Image to Active Target Tracking: Modelling, Encoding and Metrical Analysis for Unmanned Underwater Vehicles

Fixed-time adaptive cooperative control architecture for air-ground target tracking using UAV and UGV with dynamic constraints

Tracking targets using localisation methods often involves coordinated efforts between different types of vehicles. In robotics, airground visual cooperation has become increasingly prominent. However, localisation accuracy remains a challenge due to the lack of precise cooperative mechanisms for identifying targets. To address this, we propose a cooperative tracking framework using a UAV and a UGV, operating under line-of-sight and relative distance constraints. Our approach employs symmetric and asymmetric barrier functions within a backstepping design to manage multiple performance and safety constraints. A radial basis function (RBF) neural network is used to estimate the target's velocity, with adaptive laws learning the ideal weight matrix. Additionally, a terminal sliding surface method is integrated with the neural network to handle attitude uncertainties and external disturbances. The proposed control strategy guarantees that tracking errors converge to a small neighbourhood around zero within a fixed time. Numerical simulations demonstrate the effectiveness of the approach.

Multi-Scale Localization Grouping Weighted Weakly Supervised Video Instance Segmentation and Air Cruiser Application

Implementing video instance segmentation (VIS) to detect, segment, and track targets based on vision system is important research for air cruiser. Large data with high sampling difficulty result in inefficient network training and limit the air cruisers in adapting to natural scenes during mission. A multi-scale localization grouping weighted weakly supervised VIS (MLGW-VIS) is proposed. Firstly, a spatial information refinement module is designed to supplement the multi-scale spatial location information of the high-level features of the feature pyramid. Secondly, feature interaction among the channels in each sub-space of mask features is strengthened by grouping weighting module. Thirdly, projection and color similarity loss are introduced to achieve weak supervised learning. The experimental results on the data from YouTube-VIS 2019 show that MLGW-VIS has increased the average segmentation accuracy by 5.7% and reached 37.9%, and has achieved positive effects on the perception and location accuracy of objects on the air cruiser platform.

Improved DIMP Tracking Method for Terminal Guidance Applications

We propose an enhanced algorithm for the DIMP target tracking system, addressing its limitations in feature capability and its vulnerability to interference from similar targets in terminal guidance scenarios. The spatial and color feature weights of the target are enhanced by introducing CBAM attention mechanism in the feature extraction stage. The fusion module processes the extracted multi-stage feature map to acquire enhanced positional and semantic information. After the mutual correlation generates the target confidence, adaptive Kalman filtering is used for trajectory prediction and classification, and accurate target tracking is finally realized. EXPERIMENTAL DEMONSTRATION: The experimental findings indicate that the algorithm enhances both the success rate and accuracy by approximately 3%-4% on the Guidance-UAV and UAV123 datasets and improves the EAO metric of VOT2018 by 1.81%. It demonstrates good, generalized tracking capability and advantages of terminal guidance application, and can meet the relevant requirements.

Enhancing Resolvable Group Target Tracking: Integration of Labeled MultiBernoulli Filter With Deep Learning Approaches

Enhancing Resolvable Group Target Tracking: Integration of Labeled MultiBernoulli Filter With Deep Learning Approaches

S4-KD: A single step spiking SiamFC+ + for object tracking with knowledge distillation.

S4-KD: A single step spiking SiamFC+ + for object tracking with knowledge distillation.

Radar target tracking based on motion characteristic and distribution pattern matching

Radar target tracking based on motion characteristic and distribution pattern matching