Search

The problem of state estimation for non‐linear systems with unknown inputs is discussed. The objective is to construct a non‐linear filter where the unknown input affects both the state equation and measurement equation. An auto‐calibration Kalman filter is proposed with three stages, where the first and second stages are the extension of the two‐stage Kalman filter in non‐linear system cases, and the third stage is the extended Kalman filter with correlated noise. Compared with the two‐stage extended Kalman filter, the presented algorithm conducts a more reasonable linearization of the measurement equation in the third stage. As a result, auto‐calibration is realized and more accurate and more stable state estimation can be obtained. Simulative and practical examples both demonstrate the validity and superiority of the presented auto‐calibration Kalman filter. Furthermore, positioning experiments illustrate that the auto‐calibration Kalman filter can obtain an accurate long‐term heading estimation in pedestrian navigation and is easy to apply in engineering applications.

Surface functional electrical stimulation (FES) is a well-recognized technique for motor function rehabilitation of some paralyzed patients. Finding the proper positions of the electrodes on the forearm for selective control of finger movements is a challenge for FES applications. This work developed a FES system towards hand rehabilitation based on customised high-density electrodes in order to achieve optimal selective stimulation. The proposed system is composed of a customised multi-channel electronic stimulator, an electrode array of 32 electrode pads with each electrode pad size of 1.2 cm ×1.5 cm, and motion feedback sensors. An auto-calibration algorithm was developed to determine the optimal stimulation parameters and electrode pads automatically. The algorithm was based on the prior knowledge gained through manual stimulation. The FES system was tested on eight healthy subjects. The results indicated that the auto-calibration algorithm could obtain the optimal electrode pads for the activation of at least three fingers for six subjects in an isolated manner, and the time required for optimal electrode pads and stimulation parameters detection was significantly reduced, demonstrating the potential for finger movement control in hand rehabilitation.

The so-called problem of identifiability and uniqueness of both the parameter set and the simulated hydrograph (as the output of the rainfall-runoff model), while applying an automatic calibration ...

In this demo, we present a real-time social distancing system with auto-calibration using human pose information. Our system first calculates geometric parameters of a camera in 3D space, i.e. position and rotation, then, measures absolute distances between pedestrians. Detection results are visualized as 3D circles in input images and a bird’s eye view. All processing steps are completely automatic, therefore, our system can be applied for uncalibrated surveillance cameras, which are already installed in town. A demo video is available in the supplemental material.

The rapid localization of ground faults in compensated overhead line networks at the medium-voltage level presents challenges for power supply companies. Existing fault detection methods cannot be used in overhead line indicators because the high-precision voltage measurement required for this, is not yet available with state-of-the-art technology. Motivated by this problem, a sensor concept for voltage measurement with overhead line indicators is presented. The innovative calibration procedure compensates for the disturbing influence of the overhead line arrangement with knowledge of the boundary conditions. It is analysed and successfully validated in realistic laboratory setups.

Two search methods are currently available for calibrating the HEC-HMS program: the Univariate Gradient (UG) and the Nelder Mead (NM) methods; both having poor performance in continuous mod...

Industrial robots play important roles in manufacturing automation for smart manufacturing. Some high-precision applications, for example, robot drilling, robot machining, robot high-precision assembly, and robot inspection, require higher robot accuracy compared with traditional part handling operations. The monitoring and assessment of robot accuracy degradation become critical for these applications. A novel vision-based sensing system for 6-D measurement (six-dimensional x, y, z, yaw, pitch, and roll) is developed at the National Institute of Standards and Technology (NIST) to measure the dynamic high accuracy movement of a robot arm. The measured 6-D information is used for robot accuracy degradation assessment and improvement. This paper presents an automatic calibration method for a vision-based 6-D sensing system. The stereo calibration is separated from the distortion calibration to speed up the on-site adjustment. Optimization algorithms are developed to achieve high calibration accuracy. The vision-based 6-D sensing system is used on a Universal Robots (UR5) to demonstrate the feasibility of using the system to assess the robot’s accuracy degradation.

Spatial scene-understanding, including dense depth and ego-motion estimation, is an important problem in computer vision for autonomous vehicles and advanced driver assistance systems. Thus, it is beneficial to design perception modules that can utilize crowdsourced videos collected from arbitrary vehicular onboard or dashboard cameras. However, the intrinsic parameters corresponding to such cameras are often unknown or change over time. Typical manual calibration approaches require objects such as a chessboard or additional scene-specific information. On the other hand, automatic camera calibration does not have such requirements. Yet, the automatic calibration of dashboard cameras is challenging as forward and planar navigation results in critical motion sequences with reconstruction ambiguities. Structure reconstruction of complete visual-sequences that may contain tens of thousands of images is also computationally untenable. Here, we propose a system for practical monocular onboard camera auto-calibration from crowdsourced videos. We show the effectiveness of our proposed system on the KITTI raw, Oxford RobotCar, and the crowdsourced D$^2$-City datasets in varying conditions. Finally, we demonstrate its application for accurate monocular dense depth and ego-motion estimation on uncalibrated videos., Accepted at 16th International Conference on Computer Vision Theory and Applications (VISAP, 2021)

Calibration of sensors is fundamental to robust performance for intelligent vehicles. In natural environments, disturbances can easily challenge calibration. One possibility is to use natural objects of known shape to recalibrate sensors. An approach based on recognition of traffic signs, such as stop signs, and use of them for recalibration of cameras is presented. The approach is based on detection, geometry estimation, calibration, and recursive updating. Results from natural environments are presented that clearly show convergence and improved performance., 7 pages, 7 figures, 1 table, Accepted to ICRA 2021

A Functional Electrical stimulation (FES) therapy is a common rehabilitation intervention after stroke, and finite state machine (FSM) has proven to be an effective and intuitive FES control method. The FSM uses the data information generated by the accelerometer to robustly trigger state transitions. In the medical field, it is necessary to obtain highly safe and accurate acceleration data. In order to ensure the accuracy of the acceleration sensor data without affecting the accuracy of the motion analysis, we need to perform acceleration big data calibration. In this context, we propose a method for robustly calculating the auto-calibration gain using redundant acceleration vectors, and then calibrating the data generated by the accelerometer based on the calculated gain. The selection of the acceleration vector involved in the gain calculation is demonstrated by different experiments. The results show that the auto-calibration gain calculated after calibration is very close to 1, and the error is significantly less than before calibration, which indicates that the accelerometer unit is well calibrated.

We present a novel method to auto-calibrate gaze estimators based on gaze patterns obtained from other viewers. Our method is based on the observation that the gaze patterns of humans are indicative of where a new viewer will look at. When a new viewer is looking at a stimulus, we first estimate a topology of gaze points (initial gaze points). Next, these points are transformed so that they match the gaze patterns of other humans to find the correct gaze points. In a flexible uncalibrated setup with a web camera and no chin rest, the proposed method is tested on ten subjects and ten images. The method estimates the gaze points after looking at a stimulus for a few seconds with an average error below \(4.5^{\circ }\). Although the reported performance is lower than what could be achieved with dedicated hardware or calibrated setup, the proposed method still provides sufficient accuracy to trace the viewer attention. This is promising considering the fact that auto-calibration is done in a flexible setup , without the use of a chin rest, and based only on a few seconds of gaze initialization data. To the best of our knowledge, this is the first work to use human gaze patterns in order to auto-calibrate gaze estimators.

An algorithm for auto-calibration of a group of co-located uniform linear array antennas is presented. If the number of signal sources are known and, for at least one array, the ratio of the gains between two consecutive antenna elements is known, the individual unknown antenna gains can be estimated. The method is based on determining the antenna calibration parameters such that a matrix built from the array snapshots has a given rank. A numerical example illustrates the performance of the method. The numerical results suggest that the method is consistent in SNR.

We show that imperfection in an IQ-modulator degrades the accuracy of the auto bias control circuit connected to the modulator's complementary port, and propose an auto calibration technique that can effectively suppress this degradation.

In this paper, we propose to calibrate a vehicle equipped with several 2D-LIDARs and cameras. Our goal is to offer a method which is easy to use with very little manual intervention and which doesn't need a special calibration target. Sensor data is recorded while the vehicle is driven along a trajectory in a man made environment with vertical walls. The first step is a classical autocalibration approach based on visual SLAM and bundle adjustment is used to recover the cameras internal and external parameters. The second step which is the main contribution of the paper, is a novel optimisation algorithm to compute the external parameters of the LIDAR. We present the methodology in its entirety because the user has to do a single data acquisition for camera and LIDAR calibration. Our approach is validated by real experimental results.

The most important part of auto-calibration is the estimation of the fundamental matrices and the correction of the distortions caused by optical systems. From the fundamental matrices the state-of-the-art method of Lourakis [17] can determine the intrinsic calibration parameters. The prerequisite is that the fundamental matrices are very accurate, so that subsequent methods can converge. State-of-the-art methods minimize the epipolar error to approximate fundamental matrices. If more than two views are given, the trifocal error can theoretically also be used, but it is very noise sensitive and therefore less practical. In this paper, we propose a combination of both error types that leads to consistently improved fundamental matrices compared to the state of the art. The proposed method has been thoroughly evaluated on both synthetic and real data sets. Besides the increased probability that Lourakis' method converges, the resulting intrinsic and extrinsic parameters are of superior quality. The method is quasi parameter-free, easy to implement, and requires only a slightly increased computational effort.

In this paper, a novel time-domain measurement technique on a high parallelism probe card with protection resistors installed is proposed. The measured signal amplitude decreases when the measurement is performed by Needle Auto Calibration (NAC) probing on a high parallelism probe card with installed resistors. Therefore, the original signals must be carefully reconstructed, and the compensation coefficient, which is related to the number of channel branches and the value of protection resistors, must be introduced. The accuracy of the reconstructed signals is analyzed based on the varying number of channel branches and various protection resistances. The results demonstrate that the proposed technique is appropriate for evaluating the overall signal performance of probe cards with Automatic Test Equipment (ATE), which enhances the efficiency of probe card performance test dramatically.

Two auto-calibration methods, namely Nelder-Mead (NM) and Univariate-Gradient (UG) and a manual approach are available for calibrating the HEC-HMS program. However, their being either inefficient or time consuming makes it difficult to work with HEC-HMS especially when using a snowmelt module in a continuous mode. The main objective of this paper is to develop an efficient Genetic Algorithm (GA) based auto-calibration method for HEC-HMS model (HMS-GA) for continuous snowmelt simulation. A general novel procedure is presented in the absence of the HMS source code to link a heuristic algorithm with the HEC program through Jython programming language. The models are developed and evaluated using daily data from basins in Ajichai, northwestern Iran. A comparison of results for a verification period indicates a substantial improvement by applying the HMS-GA over the other available methods. Moreover, it is shown that neither NM nor UG is able to improve the results obtained by either the manual or HMS-GA. Furthermore, the proposed method significantly improves the calibrations of the HMS model found by the three other methods.

In the last few years novel color and depth (RGB-D) sensors have greatly pushed robot perception. To enable a precise pixel-wise fusion of color and depth information good calibration is needed. The calibration determines the intrinsic parameters, the extrinsic parameters, and corrects for depth errors. While classic calibration approaches involve a dedicated calibration target and a trained expert, the autonomous calibration of such camera systems for robots operating in unknown environments is still an open problem. It demands for robust methods that do not need an expert to set up or tune the algorithm. Hence, we present a robust calibration algorithm that utilizes structure from motion (SfM) reconstructions as a calibration target and incorporates plane priors in the optimization to improve the convergence behavior and improve the calibration robustness. We evaluate our method against the state of the art performing over 300 experiments on ten different datasets, and show a significant improvement of the calibration accuracy.

The strong increase of electronically power assisted cycles (EPACs) and the progress in the development of electric bicycles lead to an increasing amount of inertial measurement unit (IMU) -based use-cases for EPACs. One requirement for such use-cases is a calibrated orientation of the IMU on the EPAC. The approach presented in this paper uses bike-specific motions to calibrate a 2D-mounting orientation of a micro-electrical-mechanical-system (MEMS) IMU. The method is independent of sensor biases and requires only a small amount of computations. The results show high accuracy of the calibrated orientation during the experimental testing with an EPAC.

This paper describes novel methods of auto-calibration and auto-registration of 3D imaging and navigation, implemented in a fully integrated device made of a motorized C-arm and a navigation system. Preliminary results are presented. Benefits in terms of saved time, increased accuracy, reasonable x-ray dose, and surgeon satisfaction are reported.

In this paper, we develop an active stereo system for endoscope which requires auto-calibration, because a micro pattern projector is inserted through the instrument channel during an operation and cannot be fixed to the endoscope. For solution, a new auto-calibration technique with full 6-DOF estimation of an active stereo system without any extra devices nor extra pattern projections is proposed. In the technique, the pattern projector itself is simultaneously captured with a target scene by an endoscope camera and the silhouette of the pattern projector is used to conduct 2D-3D matching by using the knowledge of the shape of the projector. In addition, the markers which is included in the projection pattern are extracted and the distances from the closest epipolar lines are calculated as for the cost function. To enhance the robustness of the reconstruction, we also propose a simple high dynamic range (HDR) imaging system for an endoscope by alternating the input power of the pattern projector ON and OFF to blink the pattern so that exposure time will vary with beat frequency, realizing a virtual multi-exposure camera. By applying our auto calibration technique with HDR imaging system, we achieved a robust and accurate reconstruction of tissue in metric 3D under practical operation of the endoscopic system, such as reconstruction of the inside of a real stomach of a pig.