Obtaining reliable wind information is critical for efficiently managing air traffic and airport operations. Wind forecasting has been considered one of the most challenging tasks in the aviation industry. Recently, with the advent of artificial intelligence, many machine learning techniques have been widely used to address a variety of complex phenomena in wind predictions. In this paper, we propose a hybrid framework that combines a machine learning model with Kalman filtering for a wind nowcasting problem in the aviation industry. More specifically, this study has three objectives as follows: (1) compare the performance of the machine learning models (i.e., Gaussian process, multi-layer perceptron, and long short-term memory (LSTM) network) to identify the most appropriate model for wind predictions, (2) combine the machine learning model selected in step (1) with an unscented Kalman filter (UKF) to improve the fidelity of the model, and (3) perform Monte Carlo simulations to quantify uncertainties arising from the modeling process. Results show that short-term time-series wind datasets are best predicted by the LSTM network compared to the other machine learning models and the UKF-aided LSTM (UKF-LSTM) approach outperforms the LSTM network only, especially when long-term wind forecasting needs to be considered.
The goal of this study is to use Gaussian process (GP) regression models to estimate the state of colored noise systems. The derivation of a Kalman filter assumes that the process noise and measurement noise are uncorrelated and both white. In relaxing those assumptions, the Kalman filter equations were modified to deal with the non-whiteness of each noise source. The standard Kalman filter ran on an augmented system that had white noises and other approaches were also introduced depending on the forms of the noises. Those existing methods can only work when the characteristics of the colored noise are perfectly known. However, it is usually difficult to model a noise without additional knowledge of the noise statistics. When the parameters of colored noise models are totally unknown and the functions of each underlying model (nonlinear dynamic and measurement functions) are uncertain or partially known, filtering using GP-Color models can perform regardless of whatever forms of colored noise. The GPs can learn the residual outputs between the GP models and the approximate parametric models (or between actual sensor readings and predicted measurement readings), as a member of a distribution over functions, typically with a mean and covariance function. Lastly, a series of simulations, including Monte Carlo results, will be run to compare the GP based filtering techniques with the existing methods to handle the sequentially correlated noise.
Unmanned aircraft systems (UAS) can be used to collect visual assets such as 2D geo-referenced and infrared images. These 2D images can be converted into 3D point clouds through the UAS photogrammetry process. Such point clouds can then be utilized to detect obstructions around airports. This research describes the process of data collection, using UAS at airports to analyze point cloud data and document the potential implications of UAS-based airport inspections. Two field tests were conducted and three different UAS platforms were deployed. The Point Cloud Library (PCL) was employed to detect and measure trees around runways. The main contribution of this study is the provision of an overview of how UAS and visual assets can be used for airport inspections, in order to improve work efficiency and reduce human error.
With the advent of unmanned aerial vehicles (UAVs), a major area of interest in the research field of UAVs has been vision-aided inertial navigation systems (V-INS). In the front-end of V-INS, image processing extracts information about the surrounding environment and determines features or points of interest. With the extracted vision data and inertial measurement unit (IMU) dead reckoning, the most widely used algorithm for estimating vehicle and feature states in the back-end of V-INS is an extended Kalman filter (EKF). An important assumption of the EKF is Gaussian white noise. In fact, measurement outliers that arise in various realistic conditions are often non-Gaussian. A lack of compensation for unknown noise parameters often leads to a serious impact on the reliability and robustness of these navigation systems. To compensate for uncertainties of the outliers, we require modified versions of the estimator or the incorporation of other techniques into the filter. The main purpose of this paper is to develop accurate and robust V-INS for UAVs, in particular, those for situations pertaining to such unknown outliers. Feature correspondence in image processing front-end rejects vision outliers, and then a statistic test in filtering back-end detects the remaining outliers of the vision data. For frequent outliers occurrence, variational approximation for Bayesian inference derives a way to compute the optimal noise precision matrices of the measurement outliers. The overall process of outlier removal and adaptation is referred to here as “outlier-adaptive filtering”. Even though almost all approaches of V-INS remove outliers by some method, few researchers have treated outlier adaptation in V-INS in much detail. Here, results from flight datasets validate the improved accuracy of V-INS employing the proposed outlier-adaptive filtering framework.
In visual-inertial odometry (VIO), inertial measurement unit (IMU) dead reckoning acts as the dynamic model for flight vehicles while camera vision extracts information about the surrounding environment and determines features or points of interest. With these sensors, the most widely used algorithm for estimating vehicle and feature states for VIO is an extended Kalman filter (EKF). The design of the standard EKF does not inherently allow for time offsets between the timestamps of the IMU and vision data. In fact, sensor-related delays that arise in various realistic conditions are at least partially unknown parameters. A lack of compensation for unknown parameters often leads to a serious impact on the accuracy of VIO systems and systems like them. To compensate for the uncertainties of the unknown time delays, this study incorporates parameter estimation into feature initialization and state estimation. Moreover, computing cross-covariance and estimating delays in online temporal calibration correct residual, Jacobian, and covariance. Results from flight dataset testing validate the improved accuracy of VIO employing latency compensated filtering frameworks. The insights and methods proposed here are ultimately useful in any estimation problem (e.g., multi-sensor fusion scenarios) where compensation for partially unknown time delays can enhance performance.
Collision avoidance (CA) using the artificial potential field (APF) usually faces several known issues such as local minima and dynamically infeasible problems, so unmanned aerial vehicles’ (UAVs) paths planned based on the APF are safe only in a certain environment. This research proposes a CA approach that combines the APF and motion primitives (MPs) to tackle the known problems associated with the APF. Since MPs solve for a locally optimal trajectory with respect to allocated time, the trajectory obtained by the MPs is verified as dynamically feasible. When a collision checker based on the k-d tree search algorithm detects collision risk on extracted sample points from the planned trajectory, generating re-planned path candidates to avoid obstacles is performed. After rejecting unsafe route candidates, one applies the APF to select the best route among the remaining safe-path candidates. To validate the proposed approach, we simulated two meaningful scenario cases—the presence of static obstacles situation with local minima and dynamic environments with multiple UAVs present. The simulation results show that the proposed approach provides smooth, efficient, and dynamically feasible pathing compared to the APF.