Rotation Estimation Research Articles

SRTPN: Scale and Rotation Transform Prediction Net for Multimodal Remote Sensing Image Registration

How to recover geometric transformations is one of the most challenging issues in image registration. To alleviate the effect of large geometric distortion in multimodal remote sensing image registration, a scale and rotate transform prediction net is proposed in this paper. First, to reduce the scale between the reference and sensed images, the image scale regression module is constructed via CNN feature extraction and FFT correlation, and the scale of sensed image can be recovered roughly. Second, the rotation estimate module is developed for predicting the rotation angles between the reference and the scale-recovered images. Finally, to obtain the accurate registration results, LoFTR is employed to match the geometric-recovered images. The proposed registration network was evaluated on GoogleEarth, HRMS, VIS-NIR and UAV datasets with contrast differences and geometric distortions. The experimental results show that the number of correct matches of our model reached 74.6%, and the RMSE of the registration results achieved 1.236, which is superior to the related methods.

CRIN: Rotation-Invariant Point Cloud Analysis and Rotation Estimation via Centrifugal Reference Frame

Various recent methods attempt to implement rotation-invariant 3D deep learning by replacing the input coordinates of points with relative distances and angles. Due to the incompleteness of these low-level features, they have to undertake the expense of losing global information. In this paper, we propose the CRIN, namely Centrifugal Rotation-Invariant Network. CRIN directly takes the coordinates of points as input and transforms local points into rotation-invariant representations via centrifugal reference frames. Aided by centrifugal reference frames, each point corresponds to a discrete rotation so that the information of rotations can be implicitly stored in point features. Unfortunately, discrete points are far from describing the whole rotation space. We further introduce a continuous distribution for 3D rotations based on points. Furthermore, we propose an attention-based down-sampling strategy to sample points invariant to rotations. A relation module is adopted at last for reinforcing the long-range dependencies between sampled points and predicts the anchor point for unsupervised rotation estimation. Extensive experiments show that our method achieves rotation invariance, accurately estimates the object rotation, and obtains state-of-the-art results on rotation-augmented classification and part segmentation. Ablation studies validate the effectiveness of the network design.

Improved Spatial Knowledge Acquisition through Sensory Augmentation.

Sensory augmentation provides novel opportunities to broaden our knowledge of human perception through external sensors that record and transmit information beyond natural perception. To assess whether such augmented senses affect the acquisition of spatial knowledge during navigation, we trained a group of 27 participants for six weeks with an augmented sense for cardinal directions called the feelSpace belt. Then, we recruited a control group that did not receive the augmented sense and the corresponding training. All 53 participants first explored the Westbrook virtual reality environment for two and a half hours spread over five sessions before assessing their spatial knowledge in four immersive virtual reality tasks measuring cardinal, route, and survey knowledge. We found that the belt group acquired significantly more accurate cardinal and survey knowledge, which was measured in pointing accuracy, distance, and rotation estimates. Interestingly, the augmented sense also positively affected route knowledge, although to a lesser degree. Finally, the belt group reported a significant increase in the use of spatial strategies after training, while the groups' ratings were comparable at baseline. The results suggest that six weeks of training with the feelSpace belt led to improved survey and route knowledge acquisition. Moreover, the findings of our study could inform the development of assistive technologies for individuals with visual or navigational impairments, which may lead to enhanced navigation skills and quality of life.

Are we overestimating internal rotation gait by conventional modelling?

BackgroundThe determination of the knee joint axis (KJA) via specific calibration movements has become a promising alternative to the conventional approach to determine this axis based on regression equations or directly via marker placement on bony landmarks of the knee. Since the orientation of the KJA may differ between methods, it has direct influence on hip rotation and may therefore influence clinical decision making in context of transverse plane gait deviations.Research question: Does a functional KJA calibration lead to clinically relevant differences in hip rotation estimates during gait compared to the conventional method? Methods95 subjects (age: 19.9 years; BMI: 21.1 kg/m2), including 71 patients with potential rotation malalignment, were prospectively examined and underwent 3D gait analysis. For the conventional approach the KJA was determined by applying a knee alignment device, for the functional approach subjects were asked to perform two different calibration movements. Each procedure was performed twice. Mean hip rotation in stance (mHipRotSt) was determined following the conventional and the functional KJA calibration. ResultsDeming regression analysis for the comparison of conventional and functionally measured hip rotation revealed a systematic and substantial difference between methods (slope: 0.63; intercept: 0.31°). Measurement repetition with the knee alignment device revealed typical errors around 5°, whereas the functional methods lead to profoundly smaller errors around 1-2° with slightly inferior results for SQUAT compared to FLEX. However, when compared to conventional frontal plane video-taping, the conventional method seemed to reflect the more consistent results. SignificanceThe systematic linear discrepancy in mHipRotSt obtained by a functional approach as compared to the conventional approach appears critical since patients with severe internal or external rotation gait may be misjudged in function when receiving corrective procedures such as femoral derotation osteotomy.

Minuscule Corrections to Near-surface Solar Internal Rotation Using Mode Coupling

The observed solar oscillation spectrum is influenced by internal perturbations such as flows and structural asphericities. These features induce splitting of characteristic frequencies and distort the resonant-mode eigenfunctions. Global axisymmertric flow—differential rotation—is a very prominent perturbation. Tightly constrained rotation profiles as a function of latitude and radius are products of established helioseismic pipelines that use observed Dopplergrams to generate frequency-splitting measurements at high precision. However, the inference of rotation using frequency splittings does not consider the effect of mode coupling. This approximation worsens for modes with high angular degrees, as they become increasingly proximal in frequency. Since modes with high angular degrees probe the near-surface layers of the Sun, inversions considering coupled modes could potentially lead to more accurate estimates of rotation very close to the surface. In order to investigate if this is indeed the case, we perform inversions for solar differential rotation, considering coupling of modes for angular degrees 160 ≤ ℓ ≤ 300 in the surface gravity f-branch and first-overtone p modes. In keeping with the character of mode coupling, we carry out a nonlinear inversion using an eigenvalue solver. Differences in inverted profiles for frequency-splitting measurements from MDI and HMI are compared and discussed. We find that the corrections to the near-surface differential rotation profile, when accounting for mode-coupling effects, are smaller than 0.003 nHz and hence are insignificant. These minuscule corrections are found to be correlated with the solar cycle. We also present corrections to even-order splitting coefficients, which could consequently impact inversions for structure and magnetic fields.

Insights into Batch Selection for Event-Camera Motion Estimation.

Event cameras measure scene changes with high temporal resolutions, making them well-suited for visual motion estimation. The activation of pixels results in an asynchronous stream of digital data (events), which rolls continuously over time without the discrete temporal boundaries typical of frame-based cameras (where a data packet or frame is emitted at a fixed temporal rate). As such, it is not trivial to define a priori how to group/accumulate events in a way that is sufficient for computation. The suitable number of events can greatly vary for different environments, motion patterns, and tasks. In this paper, we use neural networks for rotational motion estimation as a scenario to investigate the appropriate selection of event batches to populate input tensors. Our results show that batch selection has a large impact on the results: training should be performed on a wide variety of different batches, regardless of the batch selection method; a simple fixed-time window is a good choice for inference with respect to fixed-count batches, and it also demonstrates comparable performance to more complex methods. Our initial hypothesis that a minimal amount of events is required to estimate motion (as in contrast maximization) is not valid when estimating motion with a neural network.

Mode Mixing and Rotational Splittings. II. Reconciling Different Approaches to Mode Coupling

In the mixed-mode asteroseismology of subgiants and red giants, the coupling between the p- and g-mode cavities must be understood well in order to derive localized estimates of interior rotation from measurements of mode multiplet rotational splittings. There exist now two different descriptions of this coupling: one based on an asymptotic quantization condition, and the other arising from the coupling matrices associated with “acoustic molecular orbitals.” We examine the analytic properties of both, and derive closed-form expressions for various quantities—such as the period-stretching function τ—which previously had to be solved for numerically. Using these, we reconcile both formulations for the first time, deriving relations by which quantities in each formulation may be translated to and interpreted within the other. This yields an information criterion for whether a given configuration of mixed modes may meaningfully constrain the parameters of the asymptotic construction, which is likely not satisfied by the majority of first-ascent red giant stars in our observational sample. Since this construction has been already used to make rotational measurements of such red giants, we examine—through a hare-and-hounds exercise—whether, and how, such limitations affect these existing measurements. While averaged estimates of core rotation seem fairly robust, template-matching using the asymptotic construction has difficulty reliably assigning rotational splittings to individual multiplets, or estimating the mixing fractions ζ of the most p-dominated mixed modes, where such estimates are most needed. We finally discuss implications for extending the two-zone model of radial differential rotation, e.g., via rotational inversions, with these methods.

An Improved RGB-D SLAM Method Based on Manhattan World and Point-Line-Plane Features

At present, most feature-based SLAM methods are based on feature points. But in indoor environments where texture is scarce and light varies greatly, these methods will reduce the accuracy of pose estimation or even lead to pose tracking failure. To tackle the above problem, we propose an improved RGB-D SLAM method which employs point-line-plane feature extraction and matching as well as Manhattan Frames-based rotation estimation. Finally, the proposed method is proven to have good reliability and robustness in open datasets and real indoor scenes.

IRAv3: Hierarchical Incremental Rotation Averaging on the Fly

We present IRAv3, which is built upon the state-of-the-art rotation averaging method, IRA++, to push this fundamental task in 3D computer vision one step further. The key observation of this letter lies in that during IRA++, the community detection-based Epipolar-geometry Graph (EG) clustering is preemptive and permanent, which is not relevant to the follow-up rotation averaging task and limits the upper bound of absolute rotation estimation accuracy. In this letter, however, the EG clustering is performed along with the cluster-wise absolute rotation estimation, i.e. instead of pre-determination, the affiliation of each vertex to which EG cluster is determined “on the fly”, and the EG clustering finishes until all the vertices find the clusters they belong to, together with their absolute rotations estimated (in the local coordinate systems of the clusters they attached). By this way, a rotation averaging-targeted and -friendly EG clustering is obtained, which facilitates the rotation averaging task in turn. Experiments on both 1DSfM and KITTI odometry datasets demonstrate the effectiveness of our proposed IRAv3 on large-scale rotation averaging problems and its advantages over its previous works (IRA and IRA++) and other state of the arts.

Attitude control via a feedback integrator based observer

A unified observer based attitude control is designed in this article. The novel feature of the proposed design is that the observer is amenable to Euclidean numerical integration schemes such as Euler integration. A feedback integrator term is introduced to ensure that the obtained rotation estimate converges to the manifold of rotations. This observer based control design affords significantly simpler numerical integration as opposed to discrete mechanics based integration. A formal proof of this procedure is constructed via novel application of height function based invariance methods. Simulation results indicate robustness of the proposed scheme to noise and asymmetry in inertia modeling.

Crack Monitoring in Rotating Shaft Using Rotational Speed Sensor-Based Torsional Stiffness Estimation with Adaptive Extended Kalman Filters

In this study, we present an alternative solution for detecting crack damages in rotating shafts under torque fluctuation by directly estimating the reduction in torsional shaft stiffness using the adaptive extended Kalman filter (AEKF) algorithm. A dynamic system model of a rotating shaft for designing AEKF was derived and implemented. An AEKF with a forgetting factor (λ) update was then designed to effectively estimate the time-varying parameter (torsional shaft stiffness) owing to cracks. Both simulation and experimental results demonstrated that the proposed estimation method could not only estimate the decrease in stiffness caused by a crack, but also quantitatively evaluate the fatigue crack growth by directly estimating the shaft torsional stiffness. Another advantage of the proposed approach is that it uses only two cost-effective rotational speed sensors and can be readily implemented in structural health monitoring systems of rotating machinery.

ContextAVO: Local context guided and refining poses for deep visual odometry

Learning-based monocular visual odometry (VO) has lately drawn significant attention for its robustness to camera parameters and environmental variations. The correlation of ego-motion in the local time dimension, denoted as the local context, is crucial for alleviating accumulated errors of VO problems. Unlike most current learning-based methods, our approach, called ContextAVO, focuses on the effectiveness of local contexts to improve the estimation recovered from consecutive multiple optical flow snippets. To retain the pose consistency in the temporal domain, we design the Context-Attention Refining component to adaptively ameliorate current inference by exploiting the continuity of camera motions and aligning corresponding observations with local contexts. Besides, we employ the multi-length window to make ContextAVO more suitable for general scenarios and less dependent on the fixed length of the input snippet. Extensive experiments on outdoor KITTI, Malaga, ApolloScape, and indoor TUM RGB-D datasets have demonstrated that our approach efficiently produces competitive results against classic algorithms. It outperforms state-of-the-art methods by large margins, improving up to 7.40% and 48.56% for translational and rotational estimation, respectively.

Robust Reduced Order Thau Observer With the Adaptive Fault Estimator for the Unmanned Air Vehicles

Developing fault detection and diagnoses algorithms for the unmanned air vehicles such as the quadrotors is challenging since they are intrinsically non-linear, time-varying, unstable, and uncertain. This paper develops a reduced order Thau observer by only considering the uncertain rotational dynamics, which are re-constructed as the dominant linear and non-linear for the design purpose. Therefore, the proposed Thau observer is just third order and can reveal a rotational state estimation error in the presence of the quadrotor faults. This paper also equips the proposed Thau observer with a simple online adaptive fault estimation law, which is able to recognize up to two faulty actuators instantly using the estimated rotational state error. Lyapunov analysis confirms the error convergence in both the Thau observer states and the adaptive fault estimates. In addition, this paper constructs a batch type least-squares projection approach to quantify the magnitude percentages of the actuator failures. Moreover, to show the feasibility of the proposed algorithm, this paper extensively analyses the fault detection and diagnosis results performed in the simulation and real-time environments. Finally, to demonstrate the superiority of the proposed algorithm, it is compared with a recent Kalman filter based quadrotor fault estimation research under the equal conditions.

A multi-scale approach for identification of potential pesticide use sites impacting vernal pool critical habitat in California.

Spatially explicit ecological risk assessment (ERA) requires estimating the overlap between chemical and receptor distribution to evaluate the potential impacts of exposure on nontarget organisms. Pesticide use estimation at field level is prone to error due to inconsistencies between ground-reporting and geospatial data coverage; attempts to rectify these inconsistencies have been limited in approach and rarely scaled to multiple crop types. We built upon a previously developed Bayesian approach to combine multiple crop types for a probabilistic determination of field-crop assignments and to examine co-occurrence of critical vernal pool habitats and bifenthrin application within a 5-county area in California (Madera, Merced, Sacramento, San Joaquin, and Stanislaus counties). We incorporated a multi-scale repeated sampling approach with an area constraint to improve the delineation of field boundaries and better capture variability in crop assignments and rotation schemes. After comparing the accuracy of the spatial probabilistic approach to USDA Census of Agriculture crop acreage data, we found our approach allows more specificity in the combination of crop types represented by the potential application area and improves acreage estimates when compared to traditional deterministic approaches. In addition, our multi-scale sampling scheme improved estimates of bifenthrin acreage variability for co-occurrence analysis and allowed for estimates of crop rotations that were previously uncaptured. Our approach could be leveraged for more realistic, spatially resolved exposure and effects models both in and outside of California.

Ionospheric measurement using a linearly polarized tri-band small satellite beacon

Radio waves propagating the ionosphere are subject to a number of unpredictable and corrupting effects, the most significant of which are phase path reduction, Faraday rotation, and scintillation. A method of measuring these effects is necessary in order to study and understand the solar-terrestrial connections, as well as to anticipate their impact on radio systems. However, limitations in current measurement solutions, most notably the lack of direct ionospheric Faraday rotation measurement capability, restricts the ability to characterize the ionosphere to a sufficient degree for applications such as the detection of coronal mass ejections. The rise in popularity of small satellites presents a new cost-effective opportunity to study and observe the ionosphere in detail, however the challenge of realizing a beacon within the constraints of a small satellite system must be overcome. In this paper, we introduce a measurement technique that builds on past multi-frequency beacons, incorporating the well established differential phase technique, as well as introducing a new differential polarization technique. We achieve this by employing linearly polarized beacon signals, which enables us to separate phase path reduction and Faraday rotation into a phase and polarization error respectively. The introduction of linear polarization overcomes a key limitations in current ionospheric measurement techniques: eliminating errors associated with indirect Faraday rotation measurement. The additional polarization data also enables absolute unambiguous measurement of both parameters with fewer and lower frequencies than conventional solutions, as well as without the need for precise satellite clock, phase biases and positioning data. These characteristics correspond with a reduction in the size, power and complexity of the satellite payload, and as a result, our technique better suited to small satellite applications. To validate the capability of measuring the ionosphere within the constraints of a small satellite package, we establish the influence of system noise, weak scattering, clock errors and antenna characteristics on the measurement technique. We then use this to inform a starting point for a CubeSat beacon and receiver design solution which achieves unambiguous and high-resolution measurement of ionospheric parameters with practicable system requirements and constraints. This starting point demonstrates that a resolution of 12mTECU and 0.03rad/m2 for the electron density and Faraday rotation estimates respectively, as well as useful scintillation measurements dominated by scintillation noise may be achieved utilizing a tri-band 3U CubeSat beacon with frequencies 144.2MHz, 432.6MHz, and 1297.8MHz, a transmitter power of no more than 4watts, and antenna requirements consistent with a dipole type antenna.

Motion Estimation of Non-Cooperative Space Objects Based on Monocular Sequence Images

The motion estimation of non-cooperative space objects is an important prerequisite for successfully completing on-orbit servicing, debris removing and asteroid exploration missions. The motion estimation of non-cooperative space objects is a challenging task because of the lack of prior information about the object in unknown scenarios. In this paper, a motion estimation method of non-cooperative space objects for autonomous navigation based on a monocular sequence of images is proposed. The rotational attitude estimation of the non-cooperative space object is realized by matching the feature points of adjacent frame images, and the conditions required for translational estimation are discussed. This method has small calculation costs and runs stably, which meets the requirements of real-time computation on-board. As a result, this method implements the estimation of the complete unknown information of the object, including the attitude quaternion and angular velocity. Numerical simulations and results verify the effectiveness of the proposed method.

Propeller's Rotational Speed Measurement by WLAN Sensing

This paper focuses on the propeller's rotational speed measurement by WLAN sensing using the channel state information (CSI) that indicates the multipath propagation characteristic between transmitting and receiving antennas. To measure the rotation speed by the measurement device via light and reflecting label with contactless, the distance from the target object has to be short, and such devices are not ideal for long-distance measurements from the target object. We propose a rotational speed estimation algorithm using CSI comprising time, frequency, and space domain information in MIMO-OFDM transmission to overcome this problem. The proposed algorithm transforms the time series data for each subcarrier into the normalized amplitude spectrum. After that, the rotational speed is estimated using the sum of the normalized amplitude spectrum in all subcarriers. The experimental results reveal that the proposed algorithm achieves high accuracy at a distance of 5 m from the propeller fabricated with an acrylic material and a blade length of 15cm. Moreover, if the propeller is fabricated with an aluminum, the proposed algorithm achieves high accuracy at a distance of 5 m even if the blade length is short, 5cm.

Quaternion-Based Iterative Extended Kalman Filter for Sensor Fusion of Vision Sensor and IMU in 6-DOF Displacement Monitoring

As civil structures are exposed to various external loads, their periodic evaluation is paramount to ensure their safety. By estimating the 6-degree-of-freedom (DOF) displacement of structures, structural behavior can be monitored directly. Therefore, this study aims to develop a translational and rotational displacement estimation method by fusing a vision sensor and inertial measurement unit (IMU) using a quaternion-based iterative extended Kalman filter (QIEKF). The QIEKF algorithm was applied to reduce the nonlinear influence on the measurement model. The 6-DOF displacement is predicted using the integral of the gyroscope output and updated via a combination of an accelerometer and a magnetometer through a vector matching process in the Kalman filter framework. Subsequently, the 6-DOF displacement estimation result is updated through a vision sensor using a 2-D planar marker and homography transformation in the Kalman filter framework. The performance of the proposed sensor fusion method was verified with experiments using a motorized motion stage, and the results show that the displacements can be estimated with high accuracy regardless of measurement noise and slowly varying signal drift.

Estimation of Rotational Speeds Based on Gearbox Vibrations via Artificial Neural Networks

This work investigates the estimation of rotational speeds based on neural networks and high sampled gearbox vibration data. A total of 18 different network architectures consisting of fully connected network layers, long short-term memory cells, and different activation functions are compared. In addition, the impact of the sampling rate of the vibration data, the signal length, and the input data type is investigated. Vibration data in the time domain, frequency domain, and a combination of both are examined as input data. Finally, the number of neurons is varied to obtain the best possible rotational speed estimation. In order to generate an appropriate dataset for the training of the neural networks, an experimental setup is presented, which is used to record the vibration data of the gearboxes. It becomes apparent that especially vibration data in the frequency domain are suitable for the estimation of the rotational speed. Furthermore, it could be shown that the low sampling rates result in an inaccurate speed estimation. Likewise, signal lengths that are too short result in inaccurate estimates. Overall, the speed can be estimated with an accuracy of 7.88 rpm at an average speed of the test dataset of 202 rpm.

Effects of Dynamic IMU-to-Segment Misalignment Error on 3-DOF Knee Angle Estimation in Walking and Running.

The inertial measurement unit (IMU)-to-segment (I2S) alignment is an important part of IMU-based joint angle estimation, and the accurate estimation of the three degree of freedom (3-DOF) knee angle can provide practical support for the evaluation of motions. In this paper, we introduce a dynamic weight particle swarm optimization (DPSO) algorithm with crossover factor based on the joint constraint to obtain the dynamic alignment vectors of I2S, and use them to perform the quaternion-based 3-DOF knee angle estimation algorithm. The optimization algorithm and the joint angle estimation algorithm were evaluated by comparing with the optical motion capture system. The range of 3-DOF knee angle root mean square errors (RMSEs) is 1.6°-5.9° during different motions. Furthermore, we also set up experiments of human walking (3 km/h), jogging (6 km/h) and ordinary running (9 km/h) to investigate the effects of dynamic I2S misalignment errors on 3-DOF knee angle estimation during different motions by artificially adding errors to I2S alignment parameters. The results showed differences in the effects of I2S misalignment errors on the estimation of knee abduction, internal rotation and flexion, which indicate the differences in knee joint kinematics among different motions. The IMU to thigh misalignment error has the greatest effect on the estimation of knee internal rotation. The effect of IMU to thigh misalignment error on the estimation of knee abduction angle becomes smaller and then larger during the two processes of switching from walking to jogging and then speeding up to ordinary running. The effect of IMU to shank misalignment error on the estimation of knee flexion angle is numerically the largest, while the standard deviation (SD) is the smallest. This study can provide support for future research on the accuracy of 3-DOF knee angle estimation during different motions.