PyTorchKR Papers

최신 논문

총 1,050편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2309.06008

A Note on the Extended Kalman Filter on a Manifold

The kinematics of many control systems, especially in the robotics field, naturally live on smooth manifolds. Most classical state-estimation algorithms, including the extended Kalman filter, are posed on Euclidean space. Although any filter algorithm can be adapted to a manifold setting by implementing it in local coordinates and ignoring the geometric structure, it has always been clear that there would be advantages in taking the geometric structure into consideration in developing the algorithm. In this paper, we argue that the minimum geometric structure required to adapt the extended Kalman filter to a manifold is that of an affine connection. With this structure, we show that a naive coordinate implementation of the EKF fails to account for geometry of the manifold in the update step and in the reset step. We provide geometric modifications to the classical EKF based on parallel transport of the measurement covariance (for the update) and a-posteriori state covariance (for the reset) that address these limitations. Preliminary results for attitude estimation with two directional measurements demonstrate that the proposed modifications significantly improve the transient behavior of the filter.

arXiv:2309.05644

Grid-based Hybrid 3DMA GNSS and Terrestrial Positioning

The paper discusses the increasing use of hybridized sensor information for GNSS-based localization and navigation, including the use of 3D map-aided GNSS positioning and terrestrial systems based on different geometric measurement principles. However, both GNSS and terrestrial systems are subject to negative impacts from the propagation environment, which can violate the assumptions of conventionally applied parametric state estimators. Furthermore, dynamic parametric state estimation does not account for multi-modalities within the state space leading to an information loss within the prediction step. In addition, the synchronization of non-deterministic multi-rate measurement systems needs to be accounted. In order to address these challenges, the paper proposes the use of a non-parametric filtering method, specifically a 3DMA multi-epoch Grid Filter, for the tight integration of GNSS and terrestrial signals. Specifically, the fusion of GNSS, Ultra-wide Band (UWB) and vehicle motion data is introduced based on a discrete state representation. Algorithmic challenges, including the use of different measurement models and time synchronization, are addressed. In order to evaluate the proposed method, real-world tests were conducted on an urban automotive testbed in both static and dynamic scenarios. We empirically show that we achieve sub-meter accuracy in the static scenario by averaging a positioning error of $0.64$ m, whereas in the dynamic scenario the average positioning error amounts to $1.62$ m. The paper provides a proof-of-concept of the introduced method and shows the feasibility of the inclusion of terrestrial signals in a 3DMA positioning framework in order to further enhance localization in GNSS-degraded environments.

arXiv:2309.05158

Real-Time Kinematics-Based Sensor-Fault Detection for Autonomous Vehicles Using Single and Double Transport with Adaptive Numerical Differentiation

Sensor-fault detection is crucial for the safe operation of autonomous vehicles. This paper introduces a novel kinematics-based approach for detecting and identifying faulty sensors, which is model-independent, rule-free, and applicable to ground and aerial vehicles. This method, called kinematics-based sensor fault detection (KSFD), relies on kinematic relations, sensor measurements, and real-time single and double numerical differentiation. Using onboard data from radar, rate gyros, magnetometers, and accelerometers, KSFD uniquely identifies a single faulty sensor in real time. To achieve this, adaptive input and state estimation (AISE) is used for real-time single and double numerical differentiation of the sensor data, and the single and double transport theorems are used to evaluate the consistency of data. Unlike model-based and knowledge-based methods, KSFD relies solely on sensor signals, kinematic relations, and AISE for real-time numerical differentiation. For ground vehicles, KSFD requires six kinematics-based error metrics, whereas, for aerial vehicles, nine error metrics are used. Simulated and experimental examples are provided to evaluate the effectiveness of KSFD.

arXiv:2309.05075공식 코드 · Papers with Code

Secure Set-Based State Estimation for Linear Systems under Adversarial Attacks on Sensors

Set-based state estimation plays a vital role in the safety verification of dynamical systems, which becomes significantly challenging when the system's sensors are susceptible to cyber-attacks. Existing methods often impose limitations on the attacker's capabilities, restricting the number of attacked sensors to be strictly less than half of the total number of sensors. This paper proposes a Secure Set-Based State Estimation (S3E) algorithm that addresses this limitation. The S3E algorithm guarantees that the true system state is contained within the estimated set, provided the initialization set encompasses the true initial state and the system is redundantly observable from the set of uncompromised sensors. The algorithm gives the estimated set as a collection of constrained zonotopes, which can be employed as robust certificates for verifying whether the system adheres to safety constraints. Furthermore, we demonstrate that the estimated set remains unaffected by attack signals of sufficiently large and also establish sufficient conditions for attack detection, identification, and filtering. This compels the attacker to inject only stealthy signals of small magnitude to evade detection, thus preserving the accuracy of the estimated set. When a few number of sensors (less than half) can be compromised, we prove that the estimated set remains bounded by a contracting set that converges to a ball whose radius is solely determined by the noise magnitude and is independent of the attack signals. To address the computational complexity of the algorithm, we offer several strategies for complexity-performance trade-offs. The efficacy of the proposed algorithm is illustrated through its application to a three-story building model.

arXiv:2309.03112

A Lie-Theoretic Approach to Propagating Uncertainty Jointly in Attitude and Angular Momentum

Dynamic state estimation, as opposed to kinematic state estimation, seeks to estimate not only the orientation of a rigid body but also its angular velocity, through Euler's equations of rotational motion. This paper demonstrates that the dynamic state estimation problem can be reformulated as estimating a probability distribution on a Lie group defined on phase space (the product space of rotation and angular momentum). The propagation equations are derived non-parametrically for the mean and covariance of the distribution. It is also shown that the equations can be approximately solved by ignoring the third and higher moments of the probability distribution. Numerical experiments show that the distribution constructed from the propagated mean and covariance fits the sample data better than an extended Kalman filter.

arXiv:2309.01461

Joint vehicle state and parameters estimation via Twin-in-the-Loop observers

Vehicular control systems are required to be both extremely reliable and robust to different environmental conditions, e.g. load or tire-road friction. In this paper, we extend a new paradigm for state estimation, called Twin-in-the-Loop filtering (TiL-F), to the estimation of the unknown parameters describing the vehicle operating conditions. In such an approach, a digital-twin of the vehicle (usually already available to the car manufacturer) is employed on-board as a plant replica within a closed-loop scheme, and the observer gains are tuned purely from experimental data. The proposed approach is validated against experimental data, showing to significantly outperform the state-of-the-art solutions.

arXiv:2309.01236

BodySLAM++: Fast and Tightly-Coupled Visual-Inertial Camera and Human Motion Tracking

Robust, fast, and accurate human state - 6D pose and posture - estimation remains a challenging problem. For real-world applications, the ability to estimate the human state in real-time is highly desirable. In this paper, we present BodySLAM++, a fast, efficient, and accurate human and camera state estimation framework relying on visual-inertial data. BodySLAM++ extends an existing visual-inertial state estimation framework, OKVIS2, to solve the dual task of estimating camera and human states simultaneously. Our system improves the accuracy of both human and camera state estimation with respect to baseline methods by 26% and 12%, respectively, and achieves real-time performance at 15+ frames per second on an Intel i7-model CPU. Experiments were conducted on a custom dataset containing both ground truth human and camera poses collected with an indoor motion tracking system.

arXiv:2309.04682

DeNoising-MOT: Towards Multiple Object Tracking with Severe Occlusions

Multiple object tracking (MOT) tends to become more challenging when severe occlusions occur. In this paper, we analyze the limitations of traditional Convolutional Neural Network-based methods and Transformer-based methods in handling occlusions and propose DNMOT, an end-to-end trainable DeNoising Transformer for MOT. To address the challenge of occlusions, we explicitly simulate the scenarios when occlusions occur. Specifically, we augment the trajectory with noises during training and make our model learn the denoising process in an encoder-decoder architecture, so that our model can exhibit strong robustness and perform well under crowded scenes. Additionally, we propose a Cascaded Mask strategy to better coordinate the interaction between different types of queries in the decoder to prevent the mutual suppression between neighboring trajectories under crowded scenes. Notably, the proposed method requires no additional modules like matching strategy and motion state estimation in inference. We conduct extensive experiments on the MOT17, MOT20, and DanceTrack datasets, and the experimental results show that our method outperforms previous state-of-the-art methods by a clear margin.

arXiv:2309.03979공식 코드 · Papers with Code

Separable Self and Mixed Attention Transformers for Efficient Object Tracking

The deployment of transformers for visual object tracking has shown state-of-the-art results on several benchmarks. However, the transformer-based models are under-utilized for Siamese lightweight tracking due to the computational complexity of their attention blocks. This paper proposes an efficient self and mixed attention transformer-based architecture for lightweight tracking. The proposed backbone utilizes the separable mixed attention transformers to fuse the template and search regions during feature extraction to generate superior feature encoding. Our prediction head performs global contextual modeling of the encoded features by leveraging efficient self-attention blocks for robust target state estimation. With these contributions, the proposed lightweight tracker deploys a transformer-based backbone and head module concurrently for the first time. Our ablation study testifies to the effectiveness of the proposed combination of backbone and head modules. Simulations show that our Separable Self and Mixed Attention-based Tracker, SMAT, surpasses the performance of related lightweight trackers on GOT10k, TrackingNet, LaSOT, NfS30, UAV123, and AVisT datasets, while running at 37 fps on CPU, 158 fps on GPU, and having 3.8M parameters. For example, it significantly surpasses the closely related trackers E.T.Track and MixFormerV2-S on GOT10k-test by a margin of 7.9% and 5.8%, respectively, in the AO metric. The tracker code and model is available at https://github.com/goutamyg/SMAT

arXiv:2309.03058공식 코드 · Papers with Code

Bayesian KalmanNet: Quantifying Uncertainty in Deep Learning Augmented Kalman Filter

Recent years have witnessed a growing interest in tracking algorithms that augment Kalman Filters (KFs) with Deep Neural Networks (DNNs). By transforming KFs into trainable deep learning models, one can learn from data to reliably track a latent state in complex and partially known dynamics. However, unlike classic KFs, conventional DNN-based systems do not naturally provide an uncertainty measure, such as error covariance, alongside their estimates, which is crucial in various applications that rely on KF-type tracking. This work bridges this gap by studying error covariance extraction in DNN-aided KFs. We begin by characterizing how uncertainty can be extracted from existing DNN-aided algorithms and distinguishing between approaches by their ability to associate internal features with meaningful KF quantities, such as the Kalman Gain (KG) and prior covariance. We then identify that uncertainty extraction from existing architectures necessitates additional domain knowledge not required for state estimation. Based on this insight, we propose Bayesian KalmanNet, a novel DNN-aided KF that integrates Bayesian deep learning techniques with the recently proposed KalmanNet and transforms the KF into a stochastic machine learning architecture. This architecture employs sampling techniques to predict error covariance reliably without requiring additional domain knowledge, while retaining KalmanNet's ability to accurately track in partially known dynamics. Our numerical study demonstrates that Bayesian KalmanNet provides accurate and reliable tracking in various scenarios representing partially known dynamic systems.

arXiv:2309.00498

Application of Deep Learning Methods in Monitoring and Optimization of Electric Power Systems

This PhD thesis thoroughly examines the utilization of deep learning techniques as a means to advance the algorithms employed in the monitoring and optimization of electric power systems. The first major contribution of this thesis involves the application of graph neural networks to enhance power system state estimation. The second key aspect of this thesis focuses on utilizing reinforcement learning for dynamic distribution network reconfiguration. The effectiveness of the proposed methods is affirmed through extensive experimentation and simulations.

arXiv:2309.00189

Data-Driven Safety Filter: An Input-Output Perspective

Implementation of learning-based control remains challenging due to the absence of safety guarantees. Safe control methods have turned to model-based safety filters to address these challenges, but this is paradoxical when the ultimate goal is a model-free, data-driven control solution. Addressing the core question of "Can we ensure the safety of any learning-based algorithm without explicit prediction models and state estimation?" this paper proposes a Data-Driven Safety Filter (DDSF) grounded in Behavioral System Theory (BST). The proposed method needs only a single system trajectory available in an offline dataset to modify unsafe learning inputs to safe inputs. This contribution addresses safe control in the input-output framework and therefore does not require full state measurements or explicit state estimation. Since no explicit model is required, the proposed safe control solution is not affected by unmodeled dynamics and unstructured uncertainty and can provide a safe solution for systems with unknown time delays. The effectiveness of the proposed DDSF is illustrated in simulation for a high-order six-degree-of-freedom aerial robot and a time-delay adaptive cruise control system.

arXiv:2309.00055

Precision Enhancement of Distribution System State Estimation via Tri-Objective Micro Phasor Measurement Unit Deployment

A tri-objective optimal Micro Phasor Measurement Units ({\mu}-PMUs) Placement method is presented, with a focus on minimizing the following three parameters: i) the total number of {\mu}-PMU channels, (ii) the maximum state estimation uncertainty, and (iii) the sensitivity of state estimation to line parameter tolerances. The suggested formulation takes single-line and {\mu}-PMU failures into consideration while guaranteeing the complete observability of the system in the presence and absence of contingencies. It also takes into account the impact of zero injection nodes and the quantity of {\mu}-PMU channels carried out at every node. The suggested placement issue is addressed using a customized version of the nondominated sorting genetic algorithm II (NSGA-II). According to the results achieved utilizing three test systems of varying sizes, {\mu}-PMU channels beyond predetermined thresholds only result in higher costs and negligible further decreases in state estimation uncertainty and sensitivity to line parameter tolerances. Additionally, we may omit to instrument between 30 and 40% of buses if {\mu}-PMUs with only two three-phase channels are utilized, with only a modest negative effect on state estimate performance even in the event of contingencies.

arXiv:2309.00480공식 코드 · Papers with Code

Learning-based NLOS Detection and Uncertainty Prediction of GNSS Observations with Transformer-Enhanced LSTM Network

The global navigation satellite systems (GNSS) play a vital role in transport systems for accurate and consistent vehicle localization. However, GNSS observations can be distorted due to multipath effects and non-line-of-sight (NLOS) receptions in challenging environments such as urban canyons. In such cases, traditional methods to classify and exclude faulty GNSS observations may fail, leading to unreliable state estimation and unsafe system operations. This work proposes a deep-learning-based method to detect NLOS receptions and predict GNSS pseudorange errors by analyzing GNSS observations as a spatio-temporal modeling problem. Compared to previous works, we construct a transformer-like attention mechanism to enhance the long short-term memory (LSTM) networks, improving model performance and generalization. For the training and evaluation of the proposed network, we used labeled datasets from the cities of Hong Kong and Aachen. We also introduce a dataset generation process to label the GNSS observations using lidar maps. In experimental studies, we compare the proposed network with a deep-learning-based model and classical machine-learning models. Furthermore, we conduct ablation studies of our network components and integrate the NLOS detection with data out-of-distribution in a state estimator. As a result, our network presents improved precision and recall ratios compared to other models. Additionally, we show that the proposed method avoids trajectory divergence in real-world vehicle localization by classifying and excluding NLOS observations.

arXiv:2308.13943

Robust Control Barrier Functions for Safe Control Under Uncertainty Using Extended State Observer and Output Measurement

Control barrier functions-based quadratic programming (CBF-QP) is gaining popularity as an effective controller synthesis tool for safe control. However, the provable safety is established on an accurate dynamic model and access to all states. To address such a limitation, this paper proposes a novel design combining an extended state observer (ESO) with a CBF for safe control of a system with model uncertainty and external disturbances only using output measurement. Our approach provides a less conservative estimation error bound than other disturbance observer-based CBFs. Moreover, only output measurements are needed to estimate the disturbances instead of access to the full state. The bounds of state estimation error and disturbance estimation error are obtained in a unified manner and then used for robust safe control under uncertainty. We validate our approach's efficacy in simulations of an adaptive cruise control system and a Segway self-balancing scooter.

arXiv:2308.11629

Autonomous Vehicles as a Sensor: Simulating Data Collection Process

Urban traffic state estimation is pivotal in furnishing precise and reliable insights into traffic flow characteristics, thereby enabling efficient traffic management. Traditional traffic estimation methodologies have predominantly hinged on labor-intensive and costly techniques such as loop detectors and floating car data. Nevertheless, the relentless progression in autonomous driving technology has catalyzed an increasing interest in capitalizing on the extensive potential of on-board sensor data, giving rise to a novel concept known as "Autonomous Vehicles as a Sensor" (AVaaS). This paper innovatively refines the AVaaS concept by simulating the data collection process. We take real-world sensor attributes into account and employ more accurate estimation techniques based on the on-board sensor data. Such data can facilitate the estimation of high-resolution, link-level traffic states and, more extensively, online cluster- and network-level traffic states. We substantiate the viability of the AVaaS concept through a case study conducted using a real-world traffic simulation in Ingolstadt, Germany. The results attest to the ability of AVaaS in estimating both microscopic (link-level) and macroscopic (cluster- and network-level) traffic states, thereby highlighting the immense potential of the AVaaS concept in effecting precise and reliable traffic state estimation and also further applications.

arXiv:2308.11818

Incorporating Nonlocal Traffic Flow Model in Physics-informed Neural Networks

This research contributes to the advancement of traffic state estimation methods by leveraging the benefits of the nonlocal LWR model within a physics-informed deep learning framework. The classical LWR model, while useful, falls short of accurately representing real-world traffic flows. The nonlocal LWR model addresses this limitation by considering the speed as a weighted mean of the downstream traffic density. In this paper, we propose a novel PIDL framework that incorporates the nonlocal LWR model. We introduce both fixed-length and variable-length kernels and develop the required mathematics. The proposed PIDL framework undergoes a comprehensive evaluation, including various convolutional kernels and look-ahead windows, using data from the NGSIM and CitySim datasets. The results demonstrate improvements over the baseline PIDL approach using the local LWR model. The findings highlight the potential of the proposed approach to enhance the accuracy and reliability of traffic state estimation, enabling more effective traffic management strategies.

arXiv:2308.08790

Remote State Estimation with Privacy Against Active Eavesdroppers

This paper considers a cyber-physical system under an active eavesdropping attack. A remote legitimate user estimates the state of a linear plant from the state information received from a sensor. Transmissions from the sensor occur via an insecure and unreliable network. An active eavesdropper may perform an attack during system operation. The eavesdropper intercepts transmissions from the sensor, whilst simultaneously sabotaging the data transfer from the sensor to the remote legitimate user to harm its estimation performance. To maintain state confidentiality, we propose an encoding scheme that is activated on the detection of an eavesdropper. Our scheme transmits noise based on a pseudo-random indicator, pre-arranged at the legitimate user and sensor. The transmission of noise harms the eavesdropper's performance, more than that of the legitimate user. Using the proposed encoding scheme, we impair the eavesdropper's expected estimation performance, whilst minimising expected performance degradation at the legitimate user. We explore the trade-off between state confidentiality and legitimate user performance degradation through selecting the probability that the sensor transmits noise. Under certain design choices, the trace of the expected estimation error covariance of the eavesdropper is greater than that of the legitimate user. Numerical examples are provided to illustrate the proposed encoding scheme.

arXiv:2308.08074

Real-Time Numerical Differentiation of Sampled Data Using Adaptive Input and State Estimation

Real-time numerical differentiation plays a crucial role in many digital control algorithms, such as PID control, which requires numerical differentiation to implement derivative action. This paper addresses the problem of numerical differentiation for real-time implementation with minimal prior information about the signal and noise using adaptive input and state estimation. Adaptive input estimation with adaptive state estimation (AIE/ASE) is based on retrospective cost input estimation, while adaptive state estimation is based on an adaptive Kalman filter in which the input-estimation error covariance and the measurement-noise covariance are updated online. The accuracy of AIE/ASE is compared numerically to several conventional numerical differentiation methods. Finally, AIE/ASE is applied to simulated vehicle position data generated from CarSim.

arXiv:2308.08229

Composite Disturbance Filtering: A Novel State Estimation Scheme for Systems With Multi-Source, Heterogeneous, and Isomeric Disturbances

State estimation has long been a fundamental problem in signal processing and control areas. The main challenge is to design filters with ability to reject or attenuate various disturbances. With the arrival of big data era, the disturbances of complicated systems are physically multi-source, mathematically heterogenous, affecting the system dynamics via isomeric (additive, multiplicative and recessive) channels, and deeply coupled with each other. In traditional filtering schemes, the multi-source heterogenous disturbances are usually simplified as a lumped one so that the "single" disturbance can be either rejected or attenuated. Since the pioneering work in 2012, a novel state estimation methodology called {\it composite disturbance filtering} (CDF) has been proposed, which deals with the multi-source, heterogenous, and isomeric disturbances based on their specific characteristics. With the CDF, enhanced anti-disturbance capability can be achieved via refined quantification, effective separation, and simultaneous rejection and attenuation of the disturbances. In this paper, an overview of the CDF scheme is provided, which includes the basic principle, general design procedure, application scenarios (e.g. alignment, localization and navigation), and future research directions. In summary, it is expected that the CDF offers an effective tool for state estimation, especially in the presence of multi-source heterogeneous disturbances.