Search for: All records

The paper discusses an intelligent vision-based control solution for autonomous tracking and landing of Vertical Take-Off and Landing (VTOL) capable Unmanned Aerial Vehicles (UAVs) on ships without utilizing GPS signal. The central idea involves automating the Navy helicopter ship landing procedure where the pilot utilizes the ship as the visual reference for long-range tracking; however, refers to a standardized visual cue installed on most Navy ships called the ”horizon bar” for the final approach and landing phases. This idea is implemented using a uniquely designed nonlinear controller integrated with machine vision. The vision system utilizes machine learning based object detection for long-range ship tracking and classical computer vision for the estimation of aircraft relative position and orientation utilizing the horizon bar during the final approach and landing phases. The nonlinear controller operates based on the information estimated by the vision system and has demonstrated robust tracking performance even in the presence of uncertainties. The developed autonomous ship landing system was implemented on a quad-rotor UAV equipped with an onboard camera, and approach and landing were successfully demonstrated on a moving deck, which imitates realistic ship deck motions. Extensive simulations and flight tests were conducted to demonstrate vertical landing safety, tracking capability, and landing accuracy. The video of the real-world experiments and demonstrations is available at this URL. 

Technical details associated with a novel relative motion sensor system are elaborated in the paper. By utilizing the Doppler effect, the optical sensor system estimates the relative motion rates between the sensor and the moving object equipped with modulating light sources and relatively inexpensive electrical components. A transimpedance amplifier (TIA) sensing circuit is employed to measure the Doppler shift exhibited by the amplitude modulated light sources on the moving platform. Implementation details associated with the amplitude modulation and photo-detection processes are discussed using representative hardware elements. A heterodyne mixing process with a reference signal is shown to improve the signal-to-noise ratios of the Doppler shift estimation processing pipeline. Benchtop prototype experiments are used to demonstrate the utility of the proposed technology for relative motion estimation applications. 

This paper presents a navigation system for autonomous rendezvous, proximity operations, and docking (RPOD) with respect to non-cooperative space objects using a novel velocimeter light detection and ranging (LIDAR) sensor. Given only raw position and Doppler velocity measurements, the proposed methodology is capable of estimating the six degree-of-freedom (DOF) relative velocity without any a priori information regarding the body of interest. Further, the raw Doppler velocity measurement field directly exposes the body of interest’s center of rotation (i.e. center of mass) enabling precise 6-DOF pose estimation if the rate estimates are fused within a Kalman filter architecture. These innovative techniques are computationally inexpensive and do not require information from peripheral sensors (i.e. gyroscope, magnetometer, accelerometer etc.). The efficacy of the proposed algorithms were evaluated via emulation robotics experiments at the Land, Air and Space Robotics (LASR) laboratory at Texas A&M University. Although testing was completed with a single body of interest, this approach can be used to online estimate the 6-DOF relative velocity of any amount of non-cooperative bodies within the field-of-view. 

The sensitivity of the initial condition [epoch state] orbit determination program with respect to ranging geometry is studied in this paper. Using the nonlinear least squares orbit determination program with representative dynamics and sensor models, the effects of operational considerations, such as the drop of regular ranging, information are studied. A representation of the linearized error covariance of the epoch state as a function of observation epoch is presented to provide a visual inspection of the relative measures of accuracy of the state. Applications of this analysis are applied to an orbit estimation problem, whereby the orbit of an artificial satellite around the Moon is estimated using range and range rate observations from sites located on the Earth. Information-gain measures are presented to reveal optimal observation epochs and the influence of observational geometry. Comparisons are made between information-gain profiles determined from true measurement data and measurement data generated from a priori estimates. 

An integrated sensing approach that fuses vision and range information to land an autonomous class 1 unmanned aerial system (UAS) controlled by e-modification model reference adaptive control is presented. The navigation system uses a feature detection algorithm to locate features and compute the corresponding range vectors on a coarsely instrumented landing platform. The relative translation and rotation state is estimated and sent to the flight computer for control feedback. A robust adaptive control law that guarantees uniform ultimate boundedness of the adaptive gains in the presence of bounded external disturbances is used to control the flight vehicle. Experimental flight tests are conducted to validate the integration of these systems and measure the quality of result from the navigation solution. Robustness of the control law amidst flight disturbances and hardware failures is demonstrated. The research results demonstrate the utility of low-cost, low-weight navigation solutions for navigation of small, autonomous UAS to carryout littoral proximity operations about unprepared shipdecks. 

This paper presents a Multiplicative Extended Kalman Filter (MEKF) framework using a state-of-the-art velocimeter Light Detection and Ranging (LIDAR) sensor for Terrain Relative Navigation (TRN) applications. The newly developed velocimeter LIDAR is capable of providing simultaneous position, Doppler velocity, and reflectivity measurements for every point in the point cloud. This information, along with pseudo-measurements from point cloud registration techniques, a novel bulk velocity batch state estimation process and inertial measurement data, is fused within a traditional Kalman filter architecture. Results from extensive emulation robotics experiments performed at Texas A&M’s Land, Air, and Space Robotics (LASR) laboratory and Monte Carlo simulations are presented to evaluate the efficacy of the proposed algorithms. 

This paper presents an approach for generating linear time invariant state-space models of a small Unmanned Air System. An instrumentation system using the robot operating system with commercial-off-the-shelf components is implemented to record flight data and inject auto- mated excitation signals. Offline system identification is conducted using the Observer/Kalman Identification algorithm to produce a discrete-time linear time invariant state-space model, which is then converted to a continuous time-model for analysis. Challenges concerning data collection and inverted V-Tail modelling are discussed, and solutions are presented. Longitudiunal, lateral/directional and combined longitudinal lateral/directional models of the test vehicle are generated using both manual and automated excitations, and are presented and compared. The generated longitudinal and lateral/directional results are compared to results for a small Unmanned Air System with a standard empennage. Flight test results presented in the paper show decent matching between the decoupled longitudinal and lateral/directional model and the combined longitudinal/lateral directional model. 

The paper discusses a deep reinforcement learning (RL) control strategy for fully autonomous vision-based approach and landing of vertical take-off and landing (VTOL) capable unmanned aerial vehicles (UAVs) on ships in the presence of disturbances such as wind gusts. The automation closely follows the Navy helicopter ship landing procedure and therefore, it detects a horizon bar that is installed on most Navy ships as a visual aid for pilots by applying uniquely developed computer vision techniques. The vision system utilizes the detected corners of the horizon bar and its known dimensions to estimate the relative position and heading angle of the aircraft. A deep RL-based controller was coupled with the vision system to ensure a safe and robust approach and landing at the proximity of the ship where the airflow is highly turbulent. The vision and RL-based control system was implemented on a quadrotor UAV and flight tests were conducted where the UAV approached and landed on a sub-scale ship platform undergoing 6 degrees of freedom deck motions in the presence of wind gusts. Simulations and flight tests confirmed the superior disturbance rejection capability of the RL controller when subjected to sudden 5 m/s wind gusts in different directions. Specifically, it was observed during flight tests that the deep RL controller demonstrated a 50% reduction in lateral drift from the flight path and 3 times faster disturbance rejection in comparison to a nonlinear proportional-integral-derivative controller. 

Certain dynamic modes of asymmetric quadrotor configurations are difficult to accurately model analytically. This paper synthesizes an analytical nonlinear parametric state-space model of an asymmetric quadrotor, and verifies it using a non-parametric model calculated from experimentally measured inputs and outputs of the actual vehicle. The offline system identification process produces a discrete-time Linear Time Invariant state-space model using the Observer Kalman Identification algorithm. This model is converted to a continuous time model for comparison to the linearized analytical model. Eigenvlaues,modes, and mode metrics are used to compare the parametric and non-parametric linear models. Results presented in the paper demonstrate that the identified linear model compares well to the linearized analytical model and validates the approach. 

The paper discusses a machine learning vision and nonlinear control approach for autonomous ship landing of vertical flight aircraft without utilizing GPS signal. The central idea involves automating the Navy helicopter ship landing procedure where the pilot utilizes the ship as the visual reference for long-range tracking, but refers to a standardized visual cue installed on most Navy ships called the ”horizon bar” for the final approach and landing phases. This idea is implemented using a uniquely designed nonlinear controller integrated with machine vision. The vision system utilizes machine learning based object detection for long-range ship tracking, and classical computer vision for object detection and the estimation of aircraft relative position and orientation during the final approach and landing phases. The nonlinear controller operates based on the information estimated by the vision system and has demonstrated robust tracking performance even in the presence of uncertainties. The developed autonomous ship landing system is implemented on a quad-rotor vertical take-off and landing (VTOL) capable unmanned aerial vehicle (UAV) equipped with an onboard camera and was demonstrated on a moving deck, which imitates realistic ship deck motions using a Stewart platform and a visual cue equivalent to the horizon bar. Extensive simulations and flight tests are conducted to demonstrate vertical landing safety, tracking capability, and landing accuracy while the deck is in motion.