Tracking Precision Research Articles

Toward Real-Time Animal Tracking with Integrated Stimulus Control for Automated Conditioning in Aquatic Eco-Neurotoxicology.

Aquatic eco-neurotoxicology is an emerging field that requires new analytical systems to study the effects of pollutants on animal behaviors. This is especially true if we are to gain insights into one of the least studied aspects: the potential perturbations that neurotoxicants can have on cognitive behaviors. The paucity of experimental data is partly caused by a lack of low-cost technologies for the analysis of higher-level neurological functions (e.g., associative learning) in small aquatic organisms. Here, we present a proof-of-concept prototype that utilizes a new real-time animal tracking software for on-the-fly video analysis and closed-loop, external hardware communications to deliver stimuli based on specific behaviors in aquatic organisms, spanning three animal phyla: chordates (fish, frog), platyhelminthes (flatworm), and arthropods (crustacean). The system's open-source software features an intuitive graphical user interface and advanced adaptive threshold-based image segmentation for precise animal detection. We demonstrate the precision of animal tracking across multiple aquatic species with varying modes of locomotion. The presented technology interfaces easily with low-cost and open-source hardware such as the Arduino microcontroller family for closed-loop stimuli control. The new system has potential future applications in eco-neurotoxicology, where it could enable new opportunities for cognitive research in diverse small aquatic model organisms.

Trajectory optimization and tracking control of free-flying space robots for capturing non-cooperative tumbling objects

This paper investigates the problem of capturing non-cooperative tumbling objects by free-flying space robots. To solve the two challenges of task constraints and base-manipulator coupling, a pseudospectral method based trajectory optimization and a reinforcement learning based tracking control are proposed for the free-flying space robots. Multiple constrains, including dynamics, field of view and obstacle avoidance, are considered in trajectory optimization. The adaptive segmented Radau pseudospectral method is used to discretize the energy-optimal trajectory problem into a nonlinear programming problem. By adaptively dividing the global time interval into multiple subintervals, higher-order interpolation polynomials are avoided. A reinforcement learning based parameter tuning method is proposed for the base controller to suppress the reaction torque caused by the manipulator. Numerical simulations and experiments on air-bearing testbed verify the effectiveness of our methods in terms of planning efficiency and tracking precision.

A target tracking method based on adaptive occlusion judgment and model updating strategy

Target tracking is an important research in the field of computer vision. Despite the rapid development of technology, difficulties still remain in balancing the overall performance for target occlusion, motion blur, etc. To address the above issue, we propose an improved kernel correlation filter tracking algorithm with adaptive occlusion judgement and model updating strategy (called Aojmus) to achieve robust target tracking. Firstly, the algorithm fuses color-naming (CN) and histogram of gradients (HOG) features as a feature extraction scheme and introduces a scale filter to estimate the target scale, which reduces tracking error caused by the variations of target features and scales. Secondly, the Aojmus introduces four evaluation indicators and a double thresholding mechanism to determine whether the target is occluded and the degree of occlusion respectively. The four evaluation results are weighted and fused to a final value. Finally, the updating strategy of the model is adaptively adjusted based on the weighted fusion value and the result of the scale estimation. Experimental evaluations on the OTB-2015 dataset are conducted to compare the performance of the Aojmus algorithm with four other comparable algorithms in terms of tracking precision, success rate, and speed. The experimental results show that the proposed Aojmus algorithm outperforms all the algorithms compared in terms of tracking precision. The Aojmus also exhibits excellent performance on attributes such as target occlusion and motion blur in terms of success rate. In addition, the processing speed reaches 74.85 fps, which also demonstrates good real-time performance.

RETRACTED: Vision-based tracking method of nighttime construction workers by integrating YOLOv5 and Deepsort

Due to poor visibility and fatigue factors, although nighttime construction has been widely used, its safety problems like struck-by accidents have also become increasingly prominent. Most of the current tracking methods of workers are not suitable for direct application in nighttime construction scenarios, so this research proposes a vision-based method, which integrates low-light image enhancement technology, YOLOv5 and Deepsort to track nighttime workers. The proposed method is mainly composed of four modules, including illumination enhancement module, detection module, the Kalman filter and matching module. In the experiment based on nine test videos, the method achieved the average multiple-object tracking accuracy (MOTA) of 89.93% and multiple-object tracking precision (MOTP) of 97.07%. At the same time, the experimental results also show that the method is robust to the common tracking challenges of occlusions, scale variations and posture variations. The proposed method has practical application potential in the monitoring task in nighttime construction, which makes the nighttime construction activities safer and more efficient.

RETRACTED: Vision-based tracking method of nighttime construction workers by integrating YOLOv5 and Deepsort (REDIRECT)

Due to poor visibility and fatigue factors, although nighttime construction has been widely used, its safety problems like struck-by accidents have also become increasingly prominent. Most of the current tracking methods of workers are not suitable for direct application in nighttime construction scenarios, so this research proposes a vision-based method, which integrates low-light image enhancement technology, YOLOv5 and Deepsort to track nighttime workers. The proposed method is mainly composed of four modules, including illumination enhancement module, detection module, the Kalman filter and matching module. In the experiment based on nine test videos, the method achieved the average multiple-object tracking accuracy (MOTA) of 89.93% and multiple-object tracking precision (MOTP) of 97.07%. At the same time, the experimental results also show that the method is robust to the common tracking challenges of occlusions, scale variations and posture variations. The proposed method has practical application potential in the monitoring task in nighttime construction, which makes the nighttime construction activities safer and more efficient.

A Novel Type of Pseudo-Decoupling Method for Two Degree-of-Freedom Piezoelectrically Driven Compliant Mechanisms Based on Elliptical Parameter Compensation.

At present, a large number of two-degree-of-freedom piezoelectrically driven compliant mechanisms (2-DOF PDCMs) have been widely adopted to construct various elliptical vibration machining (EVM) devices employed in precisely fabricating functional micro-structured surfaces on difficult-to-cut materials, which have broad applications in many significant fields like optical engineering and precision manufacturing. For a higher precision of conventional 2-DOF PDCMs on tracking elliptical trajectories, a novel type of pseudo-decoupling method is proposed based on phase difference compensation (PDC). With finite element analysis (FEA), the dependences of elliptical trajectory tracking precision on PDC angles will then be investigated for optimizing PDC angles under different elliptical parameters. As the modification of the PDC-based method, another type of pseudo-decoupling method will be improved based on elliptical parameter compensation (EPC) for much higher tracking precision, an amplification coefficient and a coupling coefficient will be introduced to mathematically construct the EPC-based model. A series of FEA simulations will also be conducted on a conventional 2-DOF PDCM to calculate the amplification and coupling coefficients as well as optimize the EPC parameters under four series of elliptical parameters. The tracking precision and operational feasibility of these two new pseudo-decoupling methods on four series of elliptical trajectories will be further analyzed and discussed in detail. Meanwhile, a conventional 2-DOF PDCM will be practically adopted to build an experimental system for investigating the pseudo-decoupling performances of an EPC-based method, the input and output displacements will be measured and collected to actually calculate the amplification coefficients and coupling coefficients, further inversely solving the actual input elliptical parameters with EPC. The error distances between the expected and experimental elliptical trajectories will also be calculated and discussed. Finally, several critical conclusions on this study will be briefly summarized.

Three-dimensional trajectory tracking guidance against near-space maneuvering targets with multiple constraints under wind field

This paper presents a three-dimensional disturbance observer-based adaptive tracking guidance scheme for the trajectory tracking problem of a missile intercepting a hypersonic vehicle with the wind field, autopilot lag dynamics, and input saturation. Unlike most of the existing missile trajectory tracking guidance systems without velocity controllability and external wind field model, the wind field model is added to the dynamic model, and the velocity control channel is introduced into the three-dimensional trajectory tracking guidance problem to prevent impractical scenarios and increase tracking precision. The proposed tracking guidance methodology incorporates the sliding mode and adaptive control techniques, in which the sliding mode motion is finite time stable independently of the tracking guidance system's initial conditions. Also, the super twisting switch control law is added to cope with the chattering problem. In this context, external uncertainty disturbances and unmeasured states are lumped into lump disturbance and estimated using the backpropagation neural network extended state observer. To this end, numerical simulations demonstrate the tracking to reach zero between 3 s and 5 s against various uncertainties in engineering practical, and position tracking errors are within 0.15 m, superior to the existing tracking guidance methods.

Collision avoidance control for redundant manipulators in construction operations: a null-space-based task-priority adjustment approach

PurposeThis paper aims to use redundant manipulators to solve the challenge of collision avoidance in construction operations such as welding and painting.Design/methodology/approachIn this paper, a null-space-based task-priority adjustment approach is developed to avoid collisions. The method establishes the relative position of the obstacle and the robot arm by defining the “link space,” and then the priority of the collision avoidance task and the end-effector task is adjusted according to the relative position by introducing the null space task conversion factors.FindingsNumerical simulations demonstrate that the proposed method can realize collision-free maneuvers for redundant manipulators and guarantee the tracking precision of the end-effector task. The experimental results show that the method can avoid dynamic obstacles in redundant manipulator welding tasks.Originality/valueA new formula for task priority adjustment for collision avoidance of redundant manipulators is proposed, and the original task tracking accuracy is guaranteed under the premise of safety.

Fixed-time control for free-floating space manipulators with prescribed constraints and input saturation

Abstract This paper investigates the issue of tracking control for a free-floating space manipulator with prescribed performance constraints, considering the inertia uncertainties, internal disturbances and input saturation. An inherently continuous adaptive controller is proposed by incorporating non-singular fixed-time sliding mode control, prescribed performance control (PPC), and auxiliary compensation. First, a modified non-singular fast fixed-time terminal sliding surface is constructed, which has a shorten convergence time than the conventional fixed-time sliding surface. Unlike the existing complicated PPCs, a simple structure controller is developed to satisfy prescribed performance constraints through a unique tangent-type PPC technique. The input saturation is then compensated adaptively by an auxiliary mechanism. The Lyapunov theory thoroughly validates the stability and fixed-time convergence of the closed-loop tracking system. With the suggested control scheme, the system states can converge quickly to a small neighbourhood around the origin within a preassigned time, while the position tracking error can be confined within a prescribed performance bounds even in the presence of input saturation. Compared to the existing tracking methods, the suggested control approach has the advantages of faster transient convergence, higher steady-state tracking precision, and stronger robustness. Simulation comparisons demonstrate the effectiveness and superiority of the proposed controller.

Adaptive neural approximated inverse control for photovoltaic power generation servo systems with all states constrained

For PV power generation servo systems with motor hysteresis, an all-state constrained decentralized adaptive approximation inversion control strategy is suggested in order to further increase the tracking precision of photovoltaic (PV) panels to the sun and the efficiency of solar energy utilization. Firstly, the hysteresis phenomenon in the servo motors of photovoltaic panels is considered, and the hysteresis-approximate inverse compensator is first studied in the servo motors of photovoltaic panels. Secondly, to deal with the constrained states such as rotor angle, angular velocity, and stator current in practical servo motors, and to ensure all states are strictly confined within each set of constraints with pre-set tracking performance, the asymmetric barrier Lyapunov functions and the error transformed functions are designed. Finally, to test the effectiveness of the proposed control strategy, a hardware-in-the-loop simulation platform and a photovoltaic servo system for PV power generation are built.

Hybrid Energy Regulated Constant Gain Kalman-Filter for Optimized Target Detection and Tracking in Wireless Sensor Networks

Wireless Sensor Networks (WSNs) are one of the most researched areas worldwide as the wide-scale networks possess low cost, are small in size, consume low power, and can be deployed in various environments. Among various applications of WSNs, target tracking is a highly demanding and broadly investigated application of wireless sensor networks. The parameter of accurate tracking is restricted because of the limited resources present in the wireless sensor networks, noise of the network, environmental factors, and faulty sensor nodes. Our work aims to enhance the accuracy of the tracking process as well as energy utilization by combing the mechanism of clustering with the prediction. Here, we present a hybrid energy-regulated constant gain Kalman filter-based target detection and tracking method, which is an algorithm to make the best use of energy and enhance precision in tracking. Our proposed algorithm is compared with the existing approaches where it is observed that the proposed technique possesses efficient energy utilization by decreasing the transference of unimportant data within the sensor network, achieving accurate results.

Development and evaluation of ultrasound image tracking technology based on Mask R-CNN applied to respiratory motion compensation system.

For respiration induced tumor displacement during a radiation therapy, a common method to prevent the extra radiation is image-guided radiation therapy. Moreover, mask region-based convolutional neural networks (Mask R-CNN) is one of the state-of-the-art (SOTA) object detection frameworks capable of conducting object classification, localization, and pixel-level instance segmentation. We developed a novel ultrasound image tracking technology based on Mask R-CNN for stable tracking of the detected diaphragm motion and applied to the respiratory motion compensation system (RMCS). For training Mask R-CNN, 1800 ultrasonic images of the human diaphragm are collected. Subsequently, an ultrasonic image tracking algorithm was developed to compute the mean pixel coordinates of the diaphragm detected by Mask R-CNN. These calculated coordinates are then utilized by the RMCS for compensation purposes. The tracking similarity verification experiment of mask ultrasonic imaging tracking algorithm (M-UITA) is performed. The correlation between the input signal and the signal tracked by M-UITA was evaluated during the experiment. The average discrete Fréchet distance was less than 4 mm. Subsequently, a respiratory displacement compensation experiment was conducted. The proposed method was compared to UITA, and the compensation rates of three different respiratory signals were calculated and compared. The experimental results showed that the proposed method achieved a 6.22% improvement in compensation rate compared to UITA. This study introduces a novel method called M-UITA, which offers high tracking precision and excellent stability for monitoring diaphragm movement. Additionally, it eliminates the need for manual parameter adjustments during operation, which is an added advantage.

An agent-based approach for energy-efficient sensor networks in logistics

As part of the fourth industrial revolution, logistics processes are augmented with connected information systems to improve their reliability and sustainability. Above all, customers can analyse process data obtained from the networked logistics operations to reduce costs and increase margins. The logistics of managing liquid goods is particularly challenging due to the strict transport temperature requirements involving monitoring via sensors attached to containers. However, these sensors transmit much redundant information that, at times, does not provide additional value to the customer, while consuming the limited energy stored in the sensor batteries. This paper aims to explore and study alternative approaches for location tracking and state monitoring in the context of liquid goods logistics. This problem is addressed by using a combination of data-driven sensing and agent-based modelling techniques. The simulation results show that the longest life span of batteries is achieved when most sensors are put into sleep mode yielding an increase of ×21.7 and ×3.7 for two typical routing scenarios. However, to allow for situations in which high quality sensor data is required to make decisions, agents need to be made aware of the life cycle phase of individual containers. Key contributions include (1) an agent-based approach for modelling the dynamics of liquid goods logistics to enable monitoring and detect inefficiencies (2) the development and analysis of three sensor usage strategies for reducing the energy consumption, and (3) an evaluation of the trade-offs between energy consumption and location tracking precision for timely decision making in resource constrained monitoring systems.

The use of web-scraped data to analyse the dynamics of clothing and footwear prices

Web scraping is a technique that makes it possible to obtain information from websites automatically. As online shopping grows in popularity, it became an abundant source of information on the prices of goods sold by retailers. The use of scraped data usually allows, in addition to a significant reduction of costs of price research, the improvement of the precision of inflation estimates and real-time tracking. For this reason, web scraping is a popular research tool both for statistical centers (Eurostat, British Office of National Statistics, Belgian Statbel) and universities (e.g. the Billion Prices Project conducted at Massachusetts Institute of Technology). However, the use of scraped data to calculate inflation brings about many challenges at the stage of their collection, processing, and aggregation. The aim of the study is to compare various methods of calculating price indices of clothing and footwear on the basis of scraped data. Using data from one of the largest online stores selling clothing and footwear for the period of February 2018–November 2019, the author compared the results of the Jevons chain index, the GEKS-J index and the GEKS-J expanding and updating window methods. As a result of the calculations, a high chain index drift was confirmed, and very similar results were found using the extension methods and the updated calculation window (excluding the FBEW method).

MBT3D: Deep learning based multi-object tracker for bumblebee 3D flight path estimation.

This work presents the Multi-Bees-Tracker (MBT3D) algorithm, a Python framework implementing a deep association tracker for Tracking-By-Detection, to address the challenging task of tracking flight paths of bumblebees in a social group. While tracking algorithms for bumblebees exist, they often come with intensive restrictions, such as the need for sufficient lighting, high contrast between the animal and background, absence of occlusion, significant user input, etc. Tracking flight paths of bumblebees in a social group is challenging. They suddenly adjust movements and change their appearance during different wing beat states while exhibiting significant similarities in their individual appearance. The MBT3D tracker, developed in this research, is an adaptation of an existing ant tracking algorithm for bumblebee tracking. It incorporates an offline trained appearance descriptor along with a Kalman Filter for appearance and motion matching. Different detector architectures for upstream detections (You Only Look Once (YOLOv5), Faster Region Proposal Convolutional Neural Network (Faster R-CNN), and RetinaNet) are investigated in a comparative study to optimize performance. The detection models were trained on a dataset containing 11359 labeled bumblebee images. YOLOv5 reaches an Average Precision of AP = 53, 8%, Faster R-CNN achieves AP = 45, 3% and RetinaNet AP = 38, 4% on the bumblebee validation dataset, which consists of 1323 labeled bumblebee images. The tracker's appearance model is trained on 144 samples. The tracker (with Faster R-CNN detections) reaches a Multiple Object Tracking Accuracy MOTA = 93, 5% and a Multiple Object Tracking Precision MOTP = 75, 6% on a validation dataset containing 2000 images, competing with state-of-the-art computer vision methods. The framework allows reliable tracking of different bumblebees in the same video stream with rarely occurring identity switches (IDS). MBT3D has much lower IDS than other commonly used algorithms, with one of the lowest false positive rates, competing with state-of-the-art animal tracking algorithms. The developed framework reconstructs the 3-dimensional (3D) flight paths of the bumblebees by triangulation. It also handles and compares two alternative stereo camera pairs if desired.

Design and Characterization of a single-axis solar tracker system for Small Scale Parabolic trough collectors

ABSTRACT This paper introduces a detailed design and development of a solar tracker (ST) prototype for small-sized parabolic trough collectors (PTCs) with one degree of freedom. The PTCs are intended to produce industrial heat at a moderate temperature. The contribution of this work lies in its introduction of a new and optimized solar tracker, which prioritizes both accuracy and cost-efficiency in its design. This aspect was emphasized throughout the experimental investigation, along with a comparison of findings from existing literature. Therefore, the open-loop strategy based on the solar position algorithm (SPA) is adopted here to determine the tracking-elevation angle. A software algorithm is proposed and implemented in a control unit managed by low-cost board. Then, the tracking angle is updated by controlling the electro-mechanical system using an on-off hysteresis approach. An experimental characterization of the tracker is carried out regarding the tracking precision by directly measuring the errors in positioning angle and their impact on the optical losses. The obtained results show good performance of the tracker with a standard deviation of tracking errors of 0.55°, which is adequate for such collectors with tolerance angles of more than 2°. This makes the 96th percentile of the optical loss efficiency almost zero for our PTC.

Updatable Siamese tracker with two-stage one-shot learning

Offline-trained Siamese networks have realized very promising tracking precision and efficiency. However, the performance is still limited by the drawbacks in online update. Traditional strategies cannot tackle the irregular variations of object and the sampling noise, so it is quite risky to adopt them to update Siamese trackers. In this paper, we present a two-stage one-shot learner by exploring the learning scheme of Siamese network, which reveals there are two key issues during online update, i.e., feature fusion and feature comparison. Based on this finding, we propose an updatable Siamese tracker by introducing two independent transformers (SiamTOL). Concretely, a Cross-aware transformer is designed to combine the features of the initial and the dynamic templates, while a Decoder-favored transformer is exploited to compare the fusing template and the search region. By combining these transformers, our tracker is able to adequately model the feature dependencies between multi-frame object samples. Extensive experimental results on several popular benchmarks well manifest that the proposed approach achieves the leading performance, and outperforms other state-of-the-art trackers.

A Lightweight Model for Real-Time Monitoring of Ships

Real-time monitoring of ships is crucial for inland navigation management. Under complex conditions, it is difficult to balance accuracy, real-time performance, and practicality in ship detection and tracking. We propose a lightweight model, YOLOv8-FAS, to address this issue for real-time ship detection and tracking. First, FasterNet and the attention mechanism are integrated and introduced to achieve feature extraction simply and efficiently. Second, the lightweight GSConv convolution method and a one-shot aggregation module are introduced to construct an efficient network neck to enhance feature extraction and fusion. Furthermore, the loss function is improved based on ship characteristics to make the model more suitable for ship datasets. Finally, the advanced Bytetrack tracke is added to achieve the real-time detection and tracking of ship targets. Compared to the YOLOv8 model, YOLOv8-FAS reduces computational complexity by 0.8×109 terms of FLOPs and reduces model parameters by 20%, resulting in only 2.4×106 parameters. The mAP-0.5 is improved by 0.9%, reaching 98.50%, and the real-time object tracking precision of the model surpasses 88%. The YOLOv8-FAS model combines light weight with high precision, and can accurately perform ship detection and tracking tasks in real time. Moreover, it is suitable for deployment on hardware resource-limited devices such as unmanned surface ships.

Robotic calvarial bone sampling

AbstractThe aim of this study was to assess the feasibility of complex unicortical calvarial harvesting by using the Cold Ablation Robot-Guided Laser Osteotome (CARLO® primo+). A cadaveric study was performed with a progressive complexity of the bone harvesting. This preliminary study on the cadaveric cranial vault area examined the tracking precision, the strategies, settings and durations of harvesting, the accuracy of the unicortical bone cutting, and the risk of dura exposition.All sampling was realised with no more difficulty than that experienced during the standard procedure. No bicortical cutting occurred during CARLO® primo + robot-guided laser cutting. During the second sampling, dura was partially exposed due to improper angulation of the curved osteotome during harvesting.Complex unicortical calvarial harvesting using robot-guided laser appears to be feasible and safe. In the future, robotic approaches will probably replace current surgical techniques using cutting guides and help reduce intraoperative inaccuracies due to the human factor.

Si-MSPDNet: A multiscale Siamese network with parallel partial decoders for the 3-D measurement of spines in 3D ultrasonic images.

Early screening and frequent monitoring effectively decrease the risk of severe scoliosis, but radiation exposure is a consequence of traditional radiograph examinations. Additionally, traditional X-ray images on the coronal or sagittal plane have difficulty providing three-dimensional (3-D) information on spinal deformities. The Scolioscan system provides an innovative 3-D spine imaging approach via ultrasonic scanning, and its feasibility has been demonstrated in numerous studies. In this paper, to further examine the potential of spinal ultrasonic data for describing 3-D spinal deformities, we propose a novel deep-learning tracker named Si-MSPDNet for extracting widely employed landmarks (spinous process (SP)) in ultrasonic images of spines and establish a 3-D spinal profile to measure 3-D spinal deformities. Si-MSPDNet has a Siamese architecture. First, we employ two efficient two-stage encoders to extract features from the uncropped ultrasonic image and the patch centered on the SP cut from the image. Then, a fusion block is designed to strengthen the communication between encoded features and further refine them from channel and spatial perspectives. The SP is a very small target in ultrasonic images, so its representation is weak in the highest-level feature maps. To overcome this, we ignore the highest-level feature maps and introduce parallel partial decoders to localize the SP. The correlation evaluation in the traditional Siamese network is also expanded to multiple scales to enhance cooperation. Furthermore, we propose a binary guided mask based on vertebral anatomical prior knowledge, which can further improve the performance of our tracker by highlighting the potential region with SP. The binary-guided mask is also utilized for fully automatic initialization in tracking. We collected spinal ultrasonic data and corresponding radiographs on the coronal and sagittal planes from 150 patients to evaluate the tracking precision of Si-MSPDNet and the performance of the generated 3-D spinal profile. Experimental results revealed that our tracker achieved a tracking success rate of 100% and a mean IoU of 0.882, outperforming some commonly used tracking and real-time detection models. Furthermore, a high correlation existed on both the coronal and sagittal planes between our projected spinal curve and that extracted from the spinal annotation in X-ray images. The correlation between the tracking results of the SP and their ground truths on other projected planes was also satisfactory. More importantly, the difference in mean curvatures was slight on all projected planes between tracking results and ground truths. Thus, this study effectively demonstrates the promising potential of our 3-D spinal profile extraction method for the 3-D measurement of spinal deformities using 3-D ultrasound data.