We propose MFT -- Multi-Flow dense Tracker -- a novel method for dense, pixel-level, long-term tracking. The approach exploits optical flows estimated not only between consecutive frames, but also for pairs of frames at logarithmically spaced intervals. It selects the most reliable sequence of flows on the basis of estimates of its geometric accuracy and the probability of occlusion, both provided by a pre-trained CNN. We show that MFT achieves competitive performance on the TAP-Vid benchmark, outperforming baselines by a significant margin, and tracking densely orders of magnitude faster than the state-of-the-art point-tracking methods. The method is insensitive to medium-length occlusions and it is robustified by estimating flow with respect to the reference frame, which reduces drift.
Synthesizing visual content that meets users' needs often requires flexible and precise controllability of the pose, shape, expression, and layout of the generated objects. Existing approaches gain controllability of generative adversarial networks (GANs) via manually annotated training data or a prior 3D model, which often lack flexibility, precision, and generality. In this work, we study a powerful yet much less explored way of controlling GANs, that is, to "drag" any points of the image to precisely reach target points in a user-interactive manner, as shown in Fig.1. To achieve this, we propose DragGAN, which consists of two main components: 1) a feature-based motion supervision that drives the handle point to move towards the target position, and 2) a new point tracking approach that leverages the discriminative generator features to keep localizing the position of the handle points. Through DragGAN, anyone can deform an image with precise control over where pixels go, thus manipulating the pose, shape, expression, and layout of diverse categories such as animals, cars, humans, landscapes, etc. As these manipulations are performed on the learned generative image manifold of a GAN, they tend to produce realistic outputs even for challenging scenarios such as hallucinating occluded content and deforming shapes that consistently follow the object's rigidity. Both qualitative and quantitative comparisons demonstrate the advantage of DragGAN over prior approaches in the tasks of image manipulation and point tracking. We also showcase the manipulation of real images through GAN inversion.
The quest for optimal operation in environments with unknowns and uncertainties is highly desirable but critically challenging across numerous fields. This paper develops a dual control framework for exploration and exploitation (DCEE) to solve an auto-optimisation problem in such complex settings. In general, there is a fundamental conflict between tracking an unknown optimal operational condition and parameter identification. The DCEE framework stands out by eliminating the need for additional perturbation signals, a common requirement in existing adaptive control methods. Instead, it inherently incorporates an exploration mechanism, actively probing the uncertain environment to diminish belief uncertainty. An ensemble based multi-estimator approach is developed to learn the environmental parameters and in the meanwhile quantify the estimation uncertainty in real time. The control action is devised with dual effects, which not only minimises the tracking error between the current state and the believed unknown optimal operational condition but also reduces belief uncertainty by proactively exploring the environment. Formal properties of the proposed DCEE framework like convergence are established. A numerical example is used to validate the effectiveness of the proposed DCEE. Simulation results for maximum power point tracking are provided to further demonstrate the potential of this new framework in real world applications.
Generic motion understanding from video involves not only tracking objects, but also perceiving how their surfaces deform and move. This information is useful to make inferences about 3D shape, physical properties and object interactions. While the problem of tracking arbitrary physical points on surfaces over longer video clips has received some attention, no dataset or benchmark for evaluation existed, until now. In this paper, we first formalize the problem, naming it tracking any point (TAP). We introduce a companion benchmark, TAP-Vid, which is composed of both real-world videos with accurate human annotations of point tracks, and synthetic videos with perfect ground-truth point tracks. Central to the construction of our benchmark is a novel semi-automatic crowdsourced pipeline which uses optical flow estimates to compensate for easier, short-term motion like camera shake, allowing annotators to focus on harder sections of video. We validate our pipeline on synthetic data and propose a simple end-to-end point tracking model TAP-Net, showing that it outperforms all prior methods on our benchmark when trained on synthetic data.
Critical point tracking is a core topic in scientific visualization for understanding the dynamic behavior of time-varying vector field data. The topological notion of robustness has been introduced recently to quantify the structural stability of critical points, that is, the robustness of a critical point is the minimum amount of perturbation to the vector field necessary to cancel it. A theoretical basis has been established previously that relates critical point tracking with the notion of robustness, in particular, critical points could be tracked based on their closeness in stability, measured by robustness, instead of just distance proximities within the domain. However, in practice, the computation of classic robustness may produce artifacts when a critical point is close to the boundary of the domain; thus, we do not have a complete picture of the vector field behavior within its local neighborhood. To alleviate these issues, we introduce a multilevel robustness framework for the study of 2D time-varying vector fields. We compute the robustness of critical points across varying neighborhoods to capture the multiscale nature of the data and to mitigate the boundary effect suffered by the classic robustness computation. We demonstrate via experiments that such a new notion of robustness can be combined seamlessly with existing feature tracking algorithms to improve the visual interpretability of vector fields in terms of feature tracking, selection, and comparison for large-scale scientific simulations. We observe, for the first time, that the minimum multilevel robustness is highly correlated with physical quantities used by domain scientists in studying a real-world tropical cyclone dataset. Such observation helps to increase the physical interpretability of robustness.
This study focuses on the numerical analysis and optimal control of vertical-axis wind turbines (VAWT) using Bayesian reinforcement learning (RL). We specifically address small-scale wind turbines, which are well-suited to local and compact production of electrical energy on a small scale, such as urban and rural infrastructure installations. Existing literature concentrates on large scale wind turbines which run in unobstructed, mostly constant wind profiles. However urban installations generally must cope with rapidly changing wind patterns. To bridge this gap, we formulate and implement an RL strategy using the Markov chain Monte Carlo (MCMC) algorithm to optimize the long-term energy output of a wind turbine. Our MCMC-based RL algorithm is a model-free and gradient-free algorithm, in which the designer does not have to know the precise dynamics of the plant and its uncertainties. Our method addresses the uncertainties by using a multiplicative reward structure, in contrast with additive reward used in conventional RL approaches. We have shown numerically that the method specifically overcomes the shortcomings typically associated with conventional solutions, including, but not limited to, component aging, modeling errors, and inaccuracies in the estimation of wind speed patterns. Our results show that the proposed method is especially successful in capturing power from wind transients; by modulating the generator load and hence the rotor torque load, so that the rotor tip speed quickly reaches the optimum value for the anticipated wind speed. This ratio of rotor tip speed to wind speed is known to be critical in wind power applications. The wind to load energy efficiency of the proposed method was shown to be superior to two other methods; the classical maximum power point tracking method and a generator controlled by deep deterministic policy gradient (DDPG) method.
The current paper presents a Sliding Mode Controller (SMC) for indirect field-oriented Cascaded Doubly Fed Induction Generator (CDFIG) powered through a Matrix Converter (MC). The proposed SMC employs a continuous control strategy to accomplish free chattering fractional-order sliding-mode control and to ensure that the control of the first DFIG stator's reactive and active power is separated. An MC is used to control the current provided to the second stator of the CDFIG as an alternative to standard voltage source inverters. The two MCs are controlled via Space-Vector Pulse-Width Modulation (SVPWM) and Indirect Field Oriented Control (IFOC). The proposed Wind Power Generation System (WPGS) is used with the purpose to ensure Maximum Power Point Tracking (MPPT) sensing under various disturbance variables such as turbulent wind. The simulation results prove the efficiency and robustness of the proposed method.
With the prevalence of LiDAR sensors in autonomous driving, 3D object tracking has received increasing attention. In a point cloud sequence, 3D object tracking aims to predict the location and orientation of an object in consecutive frames given an object template. Motivated by the success of transformers, we propose Point Tracking TRansformer (PTTR), which efficiently predicts high-quality 3D tracking results in a coarse-to-fine manner with the help of transformer operations. PTTR consists of three novel designs. 1) Instead of random sampling, we design Relation-Aware Sampling to preserve relevant points to the given template during subsampling. 2) We propose a Point Relation Transformer for effective feature aggregation and feature matching between the template and search region. 3) Based on the coarse tracking results, we employ a novel Prediction Refinement Module to obtain the final refined prediction through local feature pooling. In addition, motivated by the favorable properties of the Bird's-Eye View (BEV) of point clouds in capturing object motion, we further design a more advanced framework named PTTR++, which incorporates both the point-wise view and BEV representation to exploit their complementary effect in generating high-quality tracking results. PTTR++ substantially boosts the tracking performance on top of PTTR with low computational overhead. Extensive experiments over multiple datasets show that our proposed approaches achieve superior 3D tracking accuracy and efficiency.
In this work we present a novel dual-port grid-forming control strategy, for permanent magnet synchronous generator wind turbines with back-to-back voltage source converters, that unifies the entire range of functions from maximum power point tracking (MPPT) to providing inertia and fast frequency response without explicit mode switching between grid-following and grid-forming control. The controls impose a well-defined AC voltage at the grid-side converter (GSC) and the machine-side converter (MSC) AC terminals and explicitly stabilize the DC-link capacitor voltage through both GSC and MSC. The wind turbine's kinetic energy storage and curtailment are adjusted through a combination of implicit rotor speed control and pitch angle control and directly determine the operating mode and level of grid support. Moreover, we provide analytical small-signal stability conditions for a simplified system and explicitly characterize the relationship between control gains, curtailment, and the wind turbines steady-state response. Finally, a detailed simulation study is used to validate the results and compared the proposed control with state-of-the-art MPPT control.
Multi-point tracking is a challenging task that involves detecting points in the scene and tracking them across a sequence of frames. Computing detection-based measures like the F-measure on a frame-by-frame basis is not sufficient to assess the overall performance, as it does not interpret performance in the temporal domain. The main evaluation metric available comes from Multi-object tracking (MOT) methods to benchmark performance on datasets such as KITTI with the recently proposed higher order tracking accuracy (HOTA) metric, which is capable of providing a better description of the performance over metrics such as MOTA, DetA, and IDF1. While the HOTA metric takes into account temporal associations, it does not provide a tailored means to analyse the spatial associations of a dataset in a multi-camera setup. Moreover, there are differences in evaluating the detection task for points when compared to objects (point distances vs. bounding box overlap). Therefore in this work, we propose a multi-view higher order tracking metric (mvHOTA) to determine the accuracy of multi-point (multi-instance and multi-class) tracking methods, while taking into account temporal and spatial associations.mvHOTA can be interpreted as the geometric mean of detection, temporal, and spatial associations, thereby providing equal weighting to each of the factors. We demonstrate the use of this metric to evaluate the tracking performance on an endoscopic point detection dataset from a previously organised surgical data science challenge. Furthermore, we compare with other adjusted MOT metrics for this use-case, discuss the properties of mvHOTA, and show how the proposed multi-view Association and the Occlusion index (OI) facilitate analysis of methods with respect to handling of occlusions. The code is available at https://github.com/Cardio-AI/mvhota.
Mathematical models are just models. The desire to describe battery energy storage system (BESS) operation using computationally tractable model formulations has motivated a long-standing discussion in both the scientific and industrial communities. Linear BESS models are the most widely used so far. However, finding suitable linear BESS models has been controversial. This paper focuses on the description of linear BESS models. Four linear BESS formulations are presented, among the most popularly used. A new formulation is also proposed. The 5 BESS models are tested in 100 random BESS and 1.450 random samples of daily profiles of renewable generation. Two classical problems of power systems, namely, the set-point tracking problem and the transmission expansion planning problem, are selected for numerical analysis. Five thousand simulations are used to draw a better interpretation of each linear formulation presented and showcase specific challenges of BESS models. Practical recommendations are provided based on the findings.
The increasing electricity demand and the need for clean and renewable energy resources to satisfy this demand in a cost-effective manner, imposes new challenges on researchers and developers to maximize the output of these renewable resources at all times. However, the increasing penetration of renewable energy into the grid imposes new challenges on the grid operators. All of these challenges and issues gave rise to the need of Maximum Power Point Tracker (MPPT) and Flexible Power Point Trackers (FPPT) in order to maximize the power extracted from Photovoltaic (PV) systems and meet the grid operation constraints. Existing solutions for these algorithms do not take into consideration the very high dynamical nature of weather conditions that affects the output power that can be extracted from the PV modules, whereas in practice, the weather changes dynamically faster than what the algorithms time needed to converge. The work in this document is an attempt to address this shortcoming address shortcoming by utilizing online optimization algorithms for this purpose. Numerical analysis and verification are presented in the document. Code for the algorithms can be found at this link
Using Photovoltaic systems is gradually expanded by increasing energy demand. Abundance and availability of this energy, has turned to one of the most important sources of renewable energy. Unfortunately, photovoltaic systems have two big problems: first, those have very low energy conversion efficiency (in act between 12 and 42 percent under certain circumstances). Second, the power produced by the solar cell depends on nonlinear conditions such as solar radiation, temperature and charge feature. According to this, received power maximum of photovoltaic cells depends on different non-linear variables, it is necessary to be continuously traced, as maximum received power of the cell (by controller). In this research, the increasing efficiency of photovoltaic systems has been investigated by using Maximum Power Point Tracking (MPPT) in two different modes contained connected to the Grid and disconnected from the grid with simulation by MATLAB software. The obtained results showed that the proposed technique is able to improve the current, voltage and power output of photovoltaic cells.
Facial feature tracking is a key component of imaging ballistocardiography (BCG) where accurate quantification of the displacement of facial keypoints is needed for good heart rate estimation. Skin feature tracking enables video-based quantification of motor degradation in Parkinson's disease. Traditional computer vision algorithms include Scale Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF), and Lucas-Kanade method (LK). These have long represented the state-of-the-art in efficiency and accuracy but fail when common deformations, like affine local transformations or illumination changes, are present. Over the past five years, deep convolutional neural networks have outperformed traditional methods for most computer vision tasks. We propose a pipeline for feature tracking, that applies a convolutional stacked autoencoder to identify the most similar crop in an image to a reference crop containing the feature of interest. The autoencoder learns to represent image crops into deep feature encodings specific to the object category it is trained on. We train the autoencoder on facial images and validate its ability to track skin features in general using manually labeled face and hand videos. The tracking errors of distinctive skin features (moles) are so small that we cannot exclude that they stem from the manual labelling based on a $\chi^2$-test. With a mean error of 0.6-4.2 pixels, our method outperformed the other methods in all but one scenario. More importantly, our method was the only one to not diverge. We conclude that our method creates better feature descriptors for feature tracking, feature matching, and image registration than the traditional algorithms.
In a point cloud sequence, 3D object tracking aims to predict the location and orientation of an object in the current search point cloud given a template point cloud. Motivated by the success of transformers, we propose Point Tracking TRansformer (PTTR), which efficiently predicts high-quality 3D tracking results in a coarse-to-fine manner with the help of transformer operations. PTTR consists of three novel designs. 1) Instead of random sampling, we design Relation-Aware Sampling to preserve relevant points to given templates during subsampling. 2) Furthermore, we propose a Point Relation Transformer (PRT) consisting of a self-attention and a cross-attention module. The global self-attention operation captures long-range dependencies to enhance encoded point features for the search area and the template, respectively. Subsequently, we generate the coarse tracking results by matching the two sets of point features via cross-attention. 3) Based on the coarse tracking results, we employ a novel Prediction Refinement Module to obtain the final refined prediction. In addition, we create a large-scale point cloud single object tracking benchmark based on the Waymo Open Dataset. Extensive experiments show that PTTR achieves superior point cloud tracking in both accuracy and efficiency.
The PID controller is an elegant and versatile controller for set point tracking in double integrator systems of which mechanical systems evolving on Euclidean space constitute a large class. But since mechanical systems are typically constrained interconnections of rigid bodies whose configuration space is $SE(3)$, which is not even topologically Euclidean, a geometric PID controller has been developed for mechanical systems evolving on Lie groups. In this work, we extend the framework to such systems which have nonholonomic constraints. It encompasses many practically applicable mechanical systems encountered in robotics as robots are constrained interconnections of rigid bodies where the constraints could either be holonomic or nonholonomic.
Introduction. This article deals with the optimization of the energy conversion of a grid-connected photovoltaic system. The novelty is to develop an intelligent maximum power point tracking technique using artificial neural network algorithms. Purpose. Intelligent maximum power point tracking technique is developed in order to improve the photovoltaic system performances under the variations of the temperature and irradiation. Methods. This work is to calculate and follow the maximum power point for a photovoltaic system operating according to the artificial intelligence mechanism is and the latter is used an adaptive modified perturbation and observation maximum power point tracking algorithm based on function sign to generate an specify duty cycle applied to DC-DC converter, where we use the feed forward artificial neural network type trained by Levenberg-Marquardt backpropagation. Results. The photovoltaic system that we chose to simulate and apply this intelligent technique on it is a stand-alone photovoltaic system. According to the results obtained from simulation of the photovoltaic system using adaptive modified perturbation and observation artificial neural network the efficiency and the quality of the production of energy from photovoltaic is increased. Practical value. The proposed algorithm is validated by a dSPACE DS1104 for different operating conditions. All practice results confirm the effectiveness of our proposed algorithm.
A class of data-driven control methods has recently emerged based on Willems' fundamental lemma. Such methods can ease the modelling burden in control design but can be sensitive to disturbances acting on the system under control. In this paper, we propose a restructuring of the problem to incorporate segmented prediction trajectories. The proposed segmentation leads to reduced tracking error for longer prediction horizons in the presence of unmeasured disturbance and noise when compared to an unsegmented formulation. The performance characteristics are illustrated in a set-point tracking case study in which the segmented formulation enables more consistent performance over a wide range of prediction horizons. The method is then applied to a building energy management problem using a detailed simulation environment. The case studies show that good tracking performance is achieved for a range of horizon choices, whereas performance degrades with longer horizons without segmentation.
In this paper, two maximum power point tracking (MPPT) methods for Fuel Cell (FC) systems based on Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Imperialist Competitive Algorithm trained Neural Network (ICANN) are presented. The first operation voltage of the fuel cell corresponding to maximum power point is determined based on the data, and then the duty cycle of a DC/DC converter is adjusted using fuzzy logic controller to force the system that operates in conditions which match up with its maximum power point, in order to minimize the fuel consumption. The proposed systems and conventional fuzzy controller system are simulated in the MATLAB environment and results show acceptable operation under fast variation of conditions as well as normal conditions in minimum time.
This paper presents a fast power-setpoint tracking algorithm to enable utility-scale photovoltaic (PV) systems to provide high quality grid services such as power reserves and fast frequency response. The algorithm unites maximum power-point estimation (MPPE) with flexible power-point tracking (FPPT) control to improve the performance of both algorithms, achieving fast and accurate PV power-setpoint tracking even under rapid solar irradiance changes. The MPPE is developed using a real-time, nonlinear curve-fitting approach based on the Levenberg-Marquardt algorithm. A modified adaptive FPPT based on the Perturb and Observe technique is developed for the power-setpoint tracking. By using MPPE to decouple the impact of irradiance changes on the measured PV output power, we develop a fast convergence technique for tracking power-reference changes within three FPPT iterations. Furthermore, to limit the maximum output power ripple, a new design is introduced for the steady-state voltage step size of the adaptive FPPT. The proposed algorithm is implemented on a testbed consisting of a 500 kVA three-phase, single-stage, utility-scale PV system on the OPAL-RT eMEGASIM platform. Results show that the proposed method outperforms the state-of-the-art.