Organizational Unit:

Unmanned Aerial Vehicle Research Facility

Permanent Link

https://hdl.handle.net/1853/70748

Parent Organization

Organizational Unit

Daniel Guggenheim School of Aerospace Engineering

ArchiveSpace Name Record

https://finding-aids.library.gatech.edu/agents/corporate_entities/1139

Full item page

Publication Search Results

Now showing 1 - 10 of 41

Flight Test Validation of a Neural Network based Long Term Learning Adaptive Flight Controller

(Georgia Institute of Technology, 2009-08) Chowdhary, Girish ; Johnson, Eric N.

The purpose of this paper is to present and analyze flight test results of a Long Term Learning Adaptive Flight Controller implemented on a rotorcraft and a fixed wing Unmanned Aerial Vehicle. The adaptive control architecture used is based on a proven Model Reference Adaptive Control (MRAC) architecture employing a Neural Network as the adaptive element. The method employed for training the Neural Network for these flight tests is unique since it uses current (online) as well as stored (background) information concurrently for adaptation. This ability allows the adaptive element to simulate long term memory by retaining specifically stored input output data pairs and using them for concurrent adaptation. Furthermore, the structure of the adaptive law ensures that concurrent training on past data does not affect the responsiveness of the adaptive element to current data. The results show that the concurrent use of current and background data does not affect the practical stability properties of the MRAC control architecture. The results also confirm expected improvements in performance.
Indoor Navigation for Unmanned Aerial Vehicles

(Georgia Institute of Technology, 2009-08) Sobers, D. Michael Jr. ; Chowdhary, Girish ; Johnson, Eric N.

The ability for vehicles to navigate unknown environments is critical for autonomous operation. Mapping of a vehicle's environment and self-localization within that environment are especially difficult for an Unmanned Aerial Vehicle (UAV) due to the complexity of UAV attitude and motion dynamics, as well as interference from external influences such as wind. By using a stable vehicle platform and taking advantage of the geometric structure typical of most indoor environments, the complexity of the localization and mapping problem can be reduced. Interior wall and obstacle location can be measured using low-cost range sensors. Relative vehicle location within the mapped environment can then be determined. By alternating between mapping and localization, a vehicle can explore its environment autonomously. This paper examines available low-cost range sensors for suitability in solving the mapping and localization problem. A control system and navigation algorithm are developed to perform mapping of indoor environments and localization. Simulation and experimental results are provided to determine feasibility of the proposed approach to indoor navigation.
Real-Time System Identification of a Small Multi-Engine Aircraft

(Georgia Institute of Technology, 2009-08) DeBusk, Wesley M. ; Chowdhary, Girish ; Johnson, Eric N.

In-flight identification of an aircraft's dynamic model can benefit adaptive control schemes by providing estimates of aerodynamic stability derivatives in real time. This information is useful when the dynamic model changes severely in flight such as when faults and failures occur. Moreover a continuously updating model of the aircraft dynamics can be used to monitor the performance of onboard controllers. Flight test data was collected using a sum of sines input implemented in closed loop on a twin engine, fixed wing, Unmanned Aerial Vehicle. This data has been used to estimate a complete six degree of freedom aircraft linear model using the recursive Fourier Transform Regression method in frequency domain. The methods presented in this paper have been successfully validated using computer simulation and real flight data. This paper shows the feasibility of using the frequency domain Fourier Transform Regression method for real time parameter identification.
A Language for Describing Agent Behavior as Hybrid Models

(Georgia Institute of Technology, 2009-08) Kannan, Suresh K. ; Christmann, Hans Claus ; Lee, Seungman ; Johnson, Eric N. ; Kim, So Y. K

Air Traffic conflict detection and resolution (CDR) involves multiple domains, the modeling of physical systems such as aircraft, encoding conflict detection algorithms as well as the procedures(tasks) for conflict resolution. Depending on the analysis being conducted, an implementation language is usually chosen to cater for easy rendering of algorithms in the primary domain of interest. The more specialized the choice of implementation language, the greater the difficulty in expanding the fidelity of models in other domains. This paper takes a unified view of continuous equations, algorithms and procedures. Events that occur in sequence as well as in parallel are represented in a unified manner by interpreting them as hierarchical state-charts at a low-level and as procedures or task trees at a higher level. The relationship between the two levels are recognized and utilized in decomposing task trees in to hierarchical state charts and eventually into C++ code for implementation.
A Process to Obtain Robustness Metrics for Adaptive Flight Controllers

(Georgia Institute of Technology, 2009-08) Kimbrell, Scott ; Johnson, Eric N. ; Chowdhary, Girish ; Calise, Anthony J. ; Chandramohan, Rajeev

This research effort seeks a process to draw parallels between the classical stability metrics of gain and phase margins for classical linear control systems with stability margins for adaptive controllers. The method uses a Monte Carlo simulation to yield stability threshold results for the adaptive controller based on problem-specific performance metrics. By fitting a linear controller's analytical robustness results to the adaptive stability data, the gain and phase margin for the performance-fitting linear system are considered to be the worst case equivalent gain and phase margin for the adaptive controller. This paper also discusses some experiences successfully obtaining time delay margin in a flight test setting.
3D Obstacle Detection Using a Single Camera

(Georgia Institute of Technology, 2009-08) Shah, Syed Irtiza Ali ; Johnson, Eric N.

This paper aims at detecting obstacles using a single camera in an unknown three dimensional world, for 3D motion of an unmanned air vehicle. Obstacle detection is a pre-requisite for collision-free motion of a UAV through 3D space. Most research towards vision based obstacle detection and avoidance has been done for 2D planar motion of ground robots and using active sensors like laser range finders, sonar, radar etc. Passive camera based research has mostly been done, either using stereo vision (multiple cameras) or, by developing a prior expectation map of the world and its comparison with the new image data. In this paper, an attempt has been made to find a 3D solution of the obstacle detection problem using a single camera in an unknown world. The equations developed and the simulations results presented here, show that a 3D model of the scene can be generated from 2D image information from a single camera flying through a very small arc of lateral flight around the object, without the need of capturing images from all sides as in a typical 'structures from motion' problem. The forward flight simulation results show that the depth extracted from forward motion is in fact usable for large part of the image, which is a significant contribution of this work.
An ILS Inspired Approach and Departure System Utilizing Monocular Vision

(Georgia Institute of Technology, 2009-07) Christmann, Hans Claus ; Johnson, Eric N.

This paper introduces a simple system to provide relative position between a base unit and an active unit. The proposed system is directional and allows the active unit to approach or depart from the base unit along a linear path, determined by the orientation of the base unit. The system does not require a data link between the base and the active unit, just a clear line of sight. The proposed system utilizes monocular vision on the active unit and requires the availability of enough computational power to perform simple computer vision algorithms. Part I describes the physical characteristics of the beacon utilized on the base unit, Part II describes the algorithms utilized to compute the relative position of the active unit to the base, utilizing the vision data. Part III presents simulation results. Part IV discusses the results and findings and proposes future work.
Georgia Tech Aerial Robotics Team: 2009 International Aerial Robotics Competition Entry

(Georgia Institute of Technology, 2009-07) Chowdhary, Girish ; Christmann, Hans Claus ; Johnson, Eric N. ; Salaün, Erwan ; Sobers, D. Michael Jr.

This paper examines the use of low-cost range and target identification sensors on a stable flying vehicle for suitability in solving the 5th Mission proposed for the 2009 International Aerial Robotics Competition. The ability for vehicles to navigate unknown environments is critical for autonomous operation. Mapping of a vehicle's environment and self-localization within that environment are especially difficult for an Unmanned Aerial Vehicle (UAV) due to the complexity of UAV attitude and motion dynamics. Using a stable vehicle platform and taking advantage of the geometric structure typical of most indoor environments reduces the complexity of the localization and mapping problem to the point that wall and obstacle location can be determined using low-cost range sensors. Target identification is accomplished remotely using an onboard video camera with a radio transmitter. Thus complex and time-consuming image processing routines are run on a more powerful computer, enabling further miniaturization of the flight vehicle.
On the Generation of Nearly Optimal, Planar Paths of Bounded Curvature and Curvature Gradient

(Georgia Institute of Technology, 2009) Bakolas, Efstathios ; Tsiotras, Panagiotis

We present a numerically efficient scheme to generate a family of path primitives that can be used to construct paths that take into consideration point-wise constraints on both the curvature and its derivative. The statement of the problem is a generalization of the Dubins problem to account for more realistic vehicle dynamics. The problem is solved by appropriate concatenations of line segments, circular arcs and pieces of clothoids, which are the path primitives in our scheme. Our analysis reveals that the use of clothoid segments, in addition to line segments and circular arcs, for path generation introduces significant changes on issues such as path admissibility and length minimality, when compared with the standard Dubins problem.
Real-time Implementation and Validation of a New Hierarchical Path Planning Scheme for UAVs via Hardware-in-the-Loop Simulation

(Georgia Institute of Technology, 2009) Jung, Dongwon ; Ratti, Jayant ; Tsiotras, Panagiotis

We develop a hierarchical path planning and control algorithm for a small fixed-wing UAV. Incorporating the hardware-in-the-loop (HIL) simulation environment, the hierarchical path planning and control algorithm has been validated through on-board, real-time implementation on a small autopilot. We present two distinct real-time software framework for implementation of the overall control algorithms including path planning, path smoothing, and path following. We especially emphasize the use of a real-time kernel, which shows effectiveness and robustness in accomplishing non-trivial real-time software environment. By a seamless integration of the control algorithms with a help of real-time kernel, it has been demonstrated that the UAV equipped with a small autopilot having limited computational resources manages to autonomously accomplish the mission control objective of reaching the goal while avoiding obstacles without human intervention.