Joint Tracking Research Articles

Multi-Target Tracking with Multiple Unmanned Aerial Vehicles Based on Information Fusion

In high-altitude scenarios, targets tend to occupy a small number of pixels within the UAV’s field of view, resulting in substantial errors when identity recognition is attempted based solely on appearance features during multi-UAV joint tracking. Existing methodologies typically propose projecting multi-view data onto a single plane and leveraging distance information for identity association; however, their accuracy remains low as they are contingent on one-dimensional target information. To address this limitation, this paper introduces the UAVST-HM (UAV Swarm Tracking in High-altitude scenarios for Multiple targets) model, specifically designed to handle the characteristics of targets in the field of view of multiple UAVs at high altitudes. Initially, we develop techniques for extracting targets’ appearance, geometric, and distribution features. Subsequently, adaptive weights, calculated based on the mean of the respective features, are devised to amalgamate these diverse features, thereby constructing a cost matrix for cross-view target identity matching. This matrix is processed through the Hungarian algorithm, and multi-view target identity association is ultimately achieved via threshold filtering. On the MDMT dataset, our method enhances the MDA indicator, which assesses cross-view target identity matching, by 1.78 percentage points compared to the current state of the art. This significant enhancement substantially improves the overall efficacy of multi-UAV joint visual tracking from a high-altitude perspective.

Tensor-Based Joint Symbol Detection, User Localization and Tracking for THz Systems With Dual-Wideband Effects

Tensor-Based Joint Symbol Detection, User Localization and Tracking for THz Systems With Dual-Wideband Effects

An adaptive bionic sensor: enhancing ankle joint tracking with high sensitivity and superior cushioning performance

Flexible sensors are renowned for their rapid responsiveness, high flexibility, and outstanding mechanical properties, making them ideal for applications in wearable devices, sports and health monitoring, and human-machine interfaces. To enhance the sensing performance of these flexible sensors, researchers have drawn inspiration from nature, creating various bionic microstructures. However, these structures often manifest in the macroscopic form of thin films, which do little to improve impact resistance and joint protection. In response, this work achieves a highly sensitive and impact-resistant pressure sensor by integrating bionic principles at both micro and macro levels. Specifically, at the micro level, a muscle-like hydrogel system consisting of MXene nanosheets incorporated into the double network of polyvinyl alcohol and sodium alginate matrix is selected. At the macro level, the bionic prototypes of mimosa leaves and durian spikes with good energy absorption structure, together with octopus’ sucker with good adsorption capacity are selected, and the synergistic construction of the three structures is realized through 3D printing methods. The obtained sensor has an appropriate response range, sensitivity and stability. More importantly, the displacement deformation of the hydrogel can be reduced by 91.22%, while its adhesion strength can be increased by 57.5% compared to the unstructured hydrogel. As a proof-of-concept, the proposed bionic flexible sensor can be used to monitor the motion status of the human ankle joint in real-time, thereby assisting users in making real-time judgments on gait health.

Observation of Discrete Charge States of a Coherent Two-Level System in a Superconducting Qubit.

We report observations of discrete charge states of a coherent two-level system (TLS) that is strongly coupled to an offset-charge-sensitive superconducting transmon qubit. We measure an offset charge of 0.072e associated with the two TLS eigenstates, which have a transition frequency of 2.9GHz and a relaxation time exceeding 3ms. Combining measurements in the strong dispersive and resonant regime, we quantify both transverse and longitudinal coupling of the TLS-qubit interaction. We further perform joint tracking of TLS transitions and quasiparticle tunneling dynamics but find no intrinsic correlations. This Letter demonstrates microwave-frequency TLS as a source of low-frequency charge noise.

High-Precision Multi-Object Tracking in Satellite Videos via Pixel-Wise Adaptive Feature Enhancement.

In this paper, we focus on the multi-target tracking (MOT) task in satellite videos. To achieve efficient and accurate tracking, we propose a transformer-distillation-based end-to-end joint detection and tracking (JDT) method. Specifically, (1) considering that targets in satellite videos usually have small scales and are shot from a bird's-eye view, we propose a pixel-wise transformer-based feature distillation module through which useful object representations are learned via pixel-wise distillation using a strong teacher detection network; (2) targets in satellite videos, such as airplanes, ships, and vehicles, usually have similar appearances, so we propose a temperature-controllable key feature learning objective function, and by highlighting the learning of similar features during distilling, the tracking accuracy for such objects can be further improved; (3) we propose a method that is based on an end-to-end network but simultaneously learns from a highly precise teacher network and tracking head during training so that the tracking accuracy of the end-to-end network can be improved via distillation without compromising efficiency. The experimental results on three recently released publicly available datasets demonstrated the superior performance of the proposed method for satellite videos. The proposed method achieved over 90% overall tracking performance on the AIR-MOT dataset.

Based on improved joint detection and tracking of UAV for multi-target detection of livestock

In agriculture, specifically livestock monitoring, drones' ability to track multiple targets is essential for advancing the field. However, limited computing resources and unpredictable drone movements often cause issues like ambiguous video frames, object obstructions, and size deviations. These inconsistencies reduce tracking accuracy, making traditional algorithms inadequate for handling drone footage. This study introduces an enhanced deep learning-based multi-target drone tracker framework that enables real-time processing. The proposed method combines object detection and tracking by leveraging consecutive frame pairs to extract and share features, enhancing computational efficiency. It employs diverse loss functions to address class and sample distribution imbalances and includes a composite deblurring module to enhance detection accuracy. Object association utilizes a dual regress bounding box technique, aiding in object identification verification and predictive motion. Live tracking is achieved by predicting object locations in subsequent frames, enabling real-time tracking. Evaluation against leading benchmarks shows that the system improves precision and speed, achieving a 4.3 % increase in Multi-Object Tracking Accuracy (MOTA) and a 7.7 % boost in F1 score.

A novel combined control of flexible single-link manipulator based on singularly perturbed theory

PurposeThis study aims to improve the control accuracy and antidisturbance performance of the manipulator with the flexible link, a combined controller, which combines the novel backstepping sliding mode controller based on the extended state observer (ESO) and super-twisting sliding mode controller.Design/methodology/approachFirst, the dynamic of the system is constructed by Lagrange method and assumed mode method, and then the dynamic is decoupled by the singular perturbation theory to obtain the slow-varying subsystem and fast-varying subsystem. For the slow-varying subsystem, the novel backstepping sliding mode controller based on ESO is used to achieve joint tracking. For the fast-varying subsystem, the super-twisting sliding mode controller is used for vibration suppression. At the same time, to suppress chattering, the tanh function is used to replace the sign function in the reaching law.FindingsThe simulation results show that the combined control has better trajectory tracking performance, antiinterference performance and vibration suppression performance than traditional sliding mode control (SMC).Originality/valueA novel backstepping sliding mode controller based on ESO is designed to guarantee the performance of the tracking trajectory. The new controller improves the converge rate. A super-twisting sliding mode controller, which can stabilize the fast-varying subsystem, is used to suppress the vibration of flexible link.

Double-Level Torsional Osteotomy in Severe Torsional Malalignment: A Surgical Technique

Background: Torsional malalignment of the femur and/or tibia is associated with patellar maltracking, and torsional osteotomies have shown to improve clinical symptoms. In patients with severe torsional malalignment, a double-level torsional osteotomy may be necessary. Indications: Symptomatic torsional malalignment leading to patellofemoral maltracking associated with anterior knee pain and/or patellofemoral instability. Technique Description: A double-level torsional osteotomy correcting both increased internal femoral torsion and increased external tibial torsion, is described. Meticulous preoperative deformity analysis and planning of the osteotomy is mandatory. An arthroscopy of the knee is performed first to evaluate the patellofemoral joint and patellar tracking. Tibial torsional correction is performed by a lateral approach. A biplanar osteotomy is performed, and the amount of torsional correction is controlled by 2 Schanz screws. Osteosynthesis can be done via a bended 5-hole DC-Plate or an angle-stable plate. The femoral osteotomy is performed by a medial approach. A uniplanar osteotomy is performed perpendicular to the mechanical axis of the femur. The amount of torsional correction is controlled by 2 Schanz screws. Osteosynthesis is achieved by an angle-stable plate. Results: Double-level torsional osteotomy has been shown to be an effective treatment for patients with patellar dislocation or subluxation associated with severe torsional malalignment. In a series of 18 patients, double-level osteotomy led to improved patellofemoral stability, decreased pain, and increases subjective outcome scores. Discussion/Conclusion: In patients with patellofemoral problems caused by a combined increased tibial external torsion and increased femoral internal torsion, a double-level torsional osteotomy is able to correct torsional angles to normal. Available clinical data on this procedure are promising. Patient Consent Disclosure Statement: The author(s) attests that consent has been obtained from any patient(s) appearing in this publication. If the individual may be identifiable, the author(s) has included a statement of release or other written form of approval from the patient(s) with this submission for publication.

Platform Design for Optical Screening and Conditioning for Injury Resilience.

Musculoskeletal injuries (MSKIs) among active duty soldiers result in more than 10 million limited duty days each year and account for more than 70% of the medically nondeployable population. Overuse injuries in lower limbs from running, foot marching long distances with heavy loads, and lifting heavy objects are the most common types of injuries in the military. Physical training and rehabilitation exercises for greater resiliency through aerobic, muscle strength, endurance, and agility conditioning programs can prevent or reduce the effects of MSKIs if Soldiers adhere to proper biomechanics and training techniques. We are introducing a three-dimensional (3D) camera-based platform for Optical Screening and Conditioning for Injury Resilience (OSCIR) that is designed to identify and correct high-risk movement patterns based on quantifiable biomechanical measurements in clinical or field settings. Our goal is to improve resilience to MSKI by offering greater access to quality of movement skills in warfighters through an autonomous device that can be used in Sports Medicine and Reconditioning Team (SMART) clinics and High-Intensity Tactical Training (HITT) sites. OSCIR fuses four pairs of Kinect Azure cameras into a concise footprint to achieve suitable sampling rates and an unobstructed field of view for accurate dynamic movement tracking using a custom point cloud solution. We designed a unique multistage 3D joint tracking algorithm architecture to methodically isolate the human body point cloud from the background, identify individual limb segments, and perform iterative joint optimization at the global and local joint levels. We evaluated the feasibility of our prototype system among N = 12 control participants (6 M/6 F; 21-37 years) in compliance with the Western Institutional Review Board (Tracking #20225920, approved on November 4, 2022). Five task-specific MSKI outcome metrics identified by end-user physical therapists and athletic trainers as indicators for movement quality were assessed across 7 lower-extremity exercises derived from standardized MSK assessment/conditioning batteries used in the military. Data were recorded concurrently by OSCIR and a reference standard Vicon motion capture system for validating system accuracy. Task-specific MSKI indicators for knee flexion and hip flexion range of motion achieved an average error of 4.05 ± 2.34°, while 3D position-based postural outcomes of left-right foot distance, left-right hand distance, and step length obtained mean absolute errors of 2.58 ± 2.30 cm. Results support the feasibility of our system in achieving outcomes that are comparable to currently accepted laboratory standards. Our study describes the integration process for a 3D camera-based clinical system for MSKI conditioning and rehabilitation. The impact of our system will enable key stakeholders in the military to manage MSKIs in warfighters by automating key assessment and rehabilitation test batteries; making tests more readily accessible, and interpretations more accurate by providing objective biomechanical measures. OSCIR is undergoing turn-key design features to serve as a screening tool for warfighters to readily assess susceptibility to MSKI or as a training platform to help guide exercise techniques to achieve resiliency against future injuries.

A cooperative timestamp-free clock synchronization scheme based on fast unscented Kalman filtering for time-sensitive networking

Clock synchronization, built on the classical two-way message exchange scheme, is the key prerequisite for the normal operation of time-sensitive networking (TSN). In practical TSN, the imperfect oscillator caused by environmental changes leads to clock parameters drift. Moreover, synchronization errors accumulate in multi-hop networks, making it difficult for nodes at the edge of the network to achieve precise synchronization performance. Additionally, in some industrial and vehicular scenarios, the energy consumption and complexity of clock synchronization are important factors that need to be considered. To address these problems, this paper proposes a cooperative synchronization clock offset and clock skew joint tracking algorithm based on fast Unscented Kalman filter (FUKF). To further reduce the computation and energy consumption caused by clock synchronization, we introduce randomized singular value decomposition and timestamp-free exchange. The former uses small sub-matrices approximations to replace extremely high-dimensional matrices, reducing computational time in the update stage of the UKF. The latter reduces energy consumption by setting response intervals at the receiving end, eliminating the need for timestamp exchange during the synchronization process. Therefore, this algorithm can achieve long-term synchronization without requiring excessive computational and communication overhead. The results show that the proposed method, while maintaining accuracy unchanged, reduced the running time by 20% to 90% as the number of observations increased, thus verifying the effectiveness of the algorithm.

A Terminal Residual Vibration Suppression Method of a Robot Based on Joint Trajectory Optimization

Vibration problems have become one of the most important factors affecting robot performance. To this end, a terminal residual vibration suppression method based on joint trajectory optimization is proposed to improve the accuracy and stability of robot motion. Firstly, based on the characteristics of the friction nonlinearity due to joint coupling and physical feasibility of dynamic parameters, a semidefinite programming method is used to identify dynamic parameters with actual physical meaning, thereby obtaining an accurate dynamic model. Then, based on the result of the residual vibration time domain analysis, a joint trajectory optimization model with the goal of minimizing joint tracking error is established. The Chebyshev collocation method is used to discretize the optimization model. The dynamic model is used as the optimization constraint, and barycentric interpolation is used to obtain the optimized joint motion trajectory. Finally, industrial robot experiments prove that the vibration suppression method proposed in this article can reduce the maximum acceleration amplitude of residual vibration by 62% and the vibration duration by 71%. Compared with the input shaping method, the method proposed in this paper can reduce the terminal residual vibration more effectively and ensure the consistency of running time and trajectory.

Finite-time composite learning control for flexible-link manipulator based on disturbance observer

With the development of robotic technology and theory, there is an expectation for robots to operate in complex environments. This paper addresses the joint tracking control problem of flexible-link manipulator system under unknown external disturbances, modelling uncertainties, and actuator saturation constraints. A research is conducted, proposing a composite learning control scheme based on adaptive finite-time disturbance observer (AFTDO). The designed AFTDO can estimate the upper bound of unknown disturbances by adaptive laws, and a neural network controller is developed based on the AFTDO algorithm to compensate and suppress disturbances from multiple sources, ensuring the stability of all variables within a finite time. Simultaneously, a barrier Lyapunov function (BLF) is chosen to guarantee that the system satisfies constraints under flexible vibrations. Finally, through simulation comparisons with other control strategies, the feasibility and superiority of the proposed method are convincingly demonstrated.

Multiple Object Tracking in Drone Aerial Videos by a Holistic Transformer and Multiple Feature Trajectory Matching Pattern

Drone aerial videos have immense potential in surveillance, rescue, agriculture, and urban planning. However, accurately tracking multiple objects in drone aerial videos faces challenges like occlusion, scale variations, and rapid motion. Current joint detection and tracking methods often compromise accuracy. We propose a drone multiple object tracking algorithm based on a holistic transformer and multiple feature trajectory matching pattern to overcome these challenges. The holistic transformer captures local and global interaction information, providing precise detection and appearance features for tracking. The tracker includes three components: preprocessing, trajectory prediction, and matching. Preprocessing categorizes detection boxes based on scores, with each category adopting specific matching rules. Trajectory prediction employs the visual Gaussian mixture probability hypothesis density method to integrate visual detection results to forecast object motion accurately. The multiple feature pattern introduces Gaussian, Appearance, and Optimal subpattern assignment distances for different detection box types (GAO trajectory matching pattern) in the data association process, enhancing tracking robustness. We perform comparative validations on the vision-meets-drone (VisDrone) and the unmanned aerial vehicle benchmarks; the object detection and tracking (UAVDT) datasets affirm the algorithm’s effectiveness: it obtained 38.8% and 61.7% MOTA, respectively. Its potential for seamless integration into practical engineering applications offers enhanced situational awareness and operational efficiency in drone-based missions.

Power Allocation Scheme for Multi-Static Radar to Stably Track Self-Defense Jammers

Due to suppression jamming by jammers, the signal-to-interference-plus-noise ratio (SINR) during tracking tasks is significantly reduced, thereby decreasing the target detection probability of radar systems. This may result in the interruption of the target track. To address this issue, we propose a multi-static radar power allocation algorithm that enhances the detection and tracking performance of multiple radars in relation to their targets by optimizing power resource allocation. Initially, the echo signal model and measurement model of multi-static radar are formulated, followed by the derivation of the Bayesian Cramér–Rao lower bound (BCRLB). The multi-objective optimization method is utilized to establish the objective function for joint tracking and detection, with dynamic adjustment of the weight coefficient to balance the tracking and detection performance of multiple radars. This ensures the reliability and anti-jamming capability of the multi-static radar system. Simulation results indicate that the proposed algorithm can prevent the interruption of jammer tracking and maintain robust tracking performance.

Multi-target joint tracking and classification based on joint decision and estimation

Multi-target joint tracking and classification based on joint decision and estimation

Self-supervised multi-object tracking based on metric learning

The current paradigm of joint detection and tracking still requires a large amount of instance-level trajectory annotation, which incurs high annotation costs. Moreover, treating embedding training as a classification problem would lead to difficulties in model fitting. In this paper, we propose a new self-supervised multi-object tracking based on the real-time joint detection and embedding (JDE) framework, which we termed as self-supervised multi-object tracking (SS-MOT). In SS-MOT, the short-term temporal correlations between objects within and across adjacent video frames are both considered as self-supervised constraints, where the distances between different objects are enlarged while the distances between same object of adjacent frames are brought closer. In addition, short trajectories are formed by matching pairs of adjacent frames using a matching algorithm, and these matched pairs are treated as positive samples. The distances between positive samples are then minimized for futher the feature representation of the same object. Therefore, our method can be trained on videos without instance-level annotations. We apply our approach to state-of-the-art JDE models, such as FairMOT, Cstrack, and SiamMOT, and achieve comparable results to these supevised methods on the widely used MOT17 and MOT20 challenges.

Trajectory tracking control of wearable upper limb rehabilitation robot based on Laguerre model predictive control

Wearable rehabilitation robots have become an important auxiliary tool in rehabilitation therapy, providing effective rehabilitation training and helping to recover damaged muscles and joints. In response to the difficulty of traditional control methods in solving various constraints in the trajectory tracking process of the Upper Limb Rehabilitation Robot (ULRR), this study uses model predictive control to study the trajectory tracking problem of the upper limb rehabilitation robot. Firstly, based on the Lagrangian dynamic model of wearable rehabilitation robots, an extended state space model with pseudo linearization of the system was established. Given the performance indicators and various constraints of the system, a corresponding model predictive controller is designed based on the Laguerre model to ensure system performance while greatly reducing the computational complexity of predictive control. Secondly, the stability of the model predictive controller is demonstrated, and a disturbance observer is introduced into the controller to achieve compensation for slow-varying perturbations; a joint space sliding mode variable is also introduced to achieve simultaneous tracking of the joint’s desired position and desired velocity. Finally, taking a planar two bar robot as an example, comparative simulation verification was conducted on unconstrained joint trajectory tracking and constrained joint trajectory tracking. The simulation results show that the model predictive controller can achieve simultaneous tracking of joint expected trajectory and expected speed while meeting various constraints. It has good effects in improving patient motion control ability and reducing patient fatigue, providing new research ideas and methods for the field of rehabilitation therapy.

Rules-reduced fuzzy neural network-based learning control for multiple constraints robots using online identification and compensation methods

This paper presents a novel fuzzy-learning adaptive control approach for uncertain robotic systems with input dead zone and output saturation constraints. To address the input dead zone and output saturation and improve the control stability of the robot, an internal model compensation method was proposed, which utilizes online identification of an arctanh function to approximate the output saturation of actuators. The paper also introduces an adaptive learning control system based on a rules-reduced fuzzy neural network (RRFNN) algorithm, which considers the dead zone and output saturation function to enhance control performance, and an adaptive law driven by approximation mistakes is employed to handle multiple constraints. Furthermore, the controller utilizes the integral Lyapunov stability theorem and RRFNN to design the fuzzy-learning adaptive control law, ensuring global convergence and stability of the control system. Extensive simulations and experiments on a robotic manipulator are conducted to verify the feasibility of the proposed control method. Without the proposed method, the robot’s joint tracking error exceeds 0.5 rad, while with the proposed controller, it is less than 0.05 rad and does not violate output saturation constraints. Thus, the proposed method can realize the desired performance and overcome multiple constraints.

Long-term object tracking based on joint tracking and detection strategy with Siamese network

Long-term object tracking based on joint tracking and detection strategy with Siamese network

Joint track initiation of resolvable group target

This paper studies the problem of resolvable group target track initiation (RGTI) and proposes a novel method to obtain the discrete association vectors and continuous trajectory parameters of individual targets within the group, which benefits from the fitting representations of target tracks and the utilization of group target motion constraint. Specifically, every target track within the group is modeled as a weighted sum of a same set of time basis functions with different weights, and the aim is to solve multiple pairs of association vectors and trajectory parameters, i.e., unknown weights, according to the measurement equation and group target motion constraint. Next, a joint RGTI framework, which includes track extraction, track purification, trajectory parameter and association vector correction, is proposed to transform the original complex constraint set solving problem into multiple solvable sub-problems. And the convergence of RGTI is ensured through iterative optimization of track purification, trajectory parameter correction and association vector correction. Finally, simulation example demonstrates the superiority of the proposed method compared with existing multi-target track initiation methods.