Person:
Howard,
Ayanna M.
Howard,
Ayanna M.
Permanent Link
Associated Organization(s)
Organizational Unit
Organizational Unit
ORCID
ArchiveSpace Name Record
13 results
Publication Search Results
Now showing
1 - 10 of 13
-
ItemAn Integrated Sensing Approach for Entry, Descent, and Landing of a Robotic Spacecraft(Georgia Institute of Technology, 2011-01) Howard, Ayanna M. ; Jones, Brandon M. ; Serrano, NavidWe present an integrated sensing approach for enabling autonomous landing of a robotic spacecraft on a hazardous terrain surface; this approach is active during the spacecraft descent profile. The methodology incorporates an image transformation algorithm to interpret temporal imagery land data, perform real-time detection and avoidance of terrain hazards that may impede safe landing, and increase the accuracy of landing at a desired site of interest using landmark localization techniques. By integrating a linguistic rule-based engine with linear algebra and computer vision techniques, the approach suitably addresses inherent uncertainty in the hazard assessment process while ensuring computational simplicity for real-time implementation during spacecraft descent. The proposed approach is able to identify new hazards as they emerge and also remember the locations of past hazards that might impede spacecraft landing. We provide details of the methodology in this paper and present simulation results of the approach applied to a representative Mars landing descent profile.
-
ItemDeveloping Monocular Visual Odometry and Pose Estimation for Arctic Environments(Georgia Institute of Technology, 2010-03) Williams, Stephen ; Howard, Ayanna M.Arctic regions present one of the harshest environments on Earth for people or mobile robots, yet many important scientific studies, particularly those involving climate change, require measurements from these areas. For the successful deployment of mobile sensors in the Arctic, a high-quality localization system is required. Although a global positioning system can provide coarse positioning (within several meters), it cannot provide any orientation information. A single-camera-pose-estimation system is presented, based on visual odometry techniques, which is capable of operating in the feature-poor environments of the Arctic. To validate the system, a prototype rover was developed and fielded on a glacier in Alaska. The resulting pose estimates compare favorably to values obtained by hand registering the same video sequence. Although pose errors do accumulate over time, these errors are typical of a standard odometry system but obtained in an environment where standard odometry is not practical.
-
ItemProbabilistic Analysis of Market-Based Algorithms for Initial Robotic Formations(Georgia Institute of Technology, 2009-06) Viguria Jimenez, Luis Antidio ; Howard, Ayanna M.In this paper, we present a probabilistic analysis approach for analyzing market-based algorithms applied to the initial formation problem. These algorithms determine an assignment scheme for associating individual robots with goal positions necessary to achieve a desired formation while minimizing an objective function. The main contribution of this paper is a method that calculates the expected value of the objective function, which allows us to estimate and compare theoretically the performance of two task allocation algorithms. This probabilistic analysis is applied in different runtime scenarios. We validate our approach through both simulations and experiments with real robots.
-
ItemAutomatic Generation of Persistent Formations for Multi-Agent Networks Under Range Constraints(Georgia Institute of Technology, 2009-04) Smith, Brian Stephen ; Howard, Ayanna M. ; Egerstedt, Magnus B.In this paper we present a collection of graphbased methods for determining if a team of mobile robots, subjected to sensor and communication range constraints, can persistently achieve a specified formation. What we mean by this is that the formation, once achieved, will be preserved by the direct maintenance of the smallest subset of all possible pairwise interagent distances. In this context, formations are defined by sets of points separated by distances corresponding to desired inter-agent distances. Further, we provide graph operations to describe agent interactions that implement a given formation, as well as an algorithm that, given a persistent formation, automatically generates a sequence of such operations. Experimental results are presented that illustrate the operation of the proposed methods on real robot platforms.
-
ItemMulti-Robot Deployment and Coordination with Embedded Graph Grammars(Georgia Institute of Technology, 2009-01) Smith, Brian Stephen ; Howard, Ayanna M. ; McNew, John-Michael ; Egerstedt, Magnus B.This paper presents a framework for going from specifications to implementations of decentralized control strategies for multi-robot systems. In particular, we show how the use of Embedded Graph Grammars (EGGs) provides a tool for characterizing local interaction and control laws. This paper highlights some key implementation aspects of the EGG formalism, and develops and discusses experimental results for a hexapod-based multi-robot system, as well as a multi-robot system of wheeled robots.
-
ItemLearning Approaches Applied to Human-Robot Interaction for Space Missions(Georgia Institute of Technology, 2008) Remy, Sekou ; Howard, Ayanna M.Advances in space science and technology have enabled humanity to reach a stage where we are able to send manned and unmanned vehicles to explore nearby planets. However, given key differences between terrestrial and space environments such as differences in atmospheric content and pressure, acceleration due to gravity among many others between our planet and those we wish to explore, it is not always easy or feasible to expect all mission related tasks to be accomplished by astronauts alone. The presence of robots that specialize in different tasks would greatly enhance our capabilities and enable better overall performance. In this paper we discuss a methodology for building a robotic system that can learn to perform tasks via interactive learning. This learning functionality extends the ability for a robot agent to operate with similar competence as their human teacher- whether astronaut, mission designer, or engineer. We provide details on our approach and give representative examples of applying the different methods in relevant task scenarios.
-
ItemA Learning Approach to Enable Locomotion of Multiple Robotic Agents Operating in Natural Terrain Environments(Georgia Institute of Technology, 2008) Howard, Ayanna M. ; Parker, Lonnie T. ; Smith, Brian StephenThis paper presents a methodology that utilizes soft computing approaches to enable locomotion of multiple legged robotic agents operating in natural terrain environments. For individual robotic control, the locomotion strategy consists of a hybrid FSM-GA approach that couples leg orientation states with a genetic algorithm to learn necessary leg movement sequences. To achieve multi-agent formations, locomotion behavior is driven by using a trained neural network to extract relevant distance metrics necessary to realize desired robotic formations while operating in the field. These distance metrics are then fed into local controllers for realizing linear and rotational velocity values for each robotic agent. Details of the methodology are discussed, and experimental results with a team of mobile robots are presented.
-
ItemIntegrating Virtual and Human Instructors in Robotic Learning Environments(Georgia Institute of Technology, 2008) Howard, Ayanna M. ; Remy, SekouThis paper presents two different approaches for utilizing virtual environments to enable learning for both human and robotic students. In the first approach, we showcase a 3D interactive environment that allows a human user to learn how to interact with a virtual robot, before interaction with a physical robot. In the second approach, we present a method that utilizes a simulation environment to provide feedback to a human teacher during a training session in order to concurrently allow adaptation of the learning process for both the teacher and the robotic student. We provide details of the approaches in this paper and provide results of the learning outcomes for the two different scenarios.
-
ItemComponents, Curriculum, and Community: Robots and Robotics in Undergraduate AI Education(Georgia Institute of Technology, 2006) Dodds, Zachary ; Greenwald, Lloyd ; Howard, Ayanna M. ; Tejada, Sheila ; Weinberg, Jerry B.This editorial introduction presents an overview of the robotic resources available to AI educators and provides context for the articles in this special issue. We set the stage by addressing the trade-offs among a number of established and emerging hardware and software platforms, curricular topics, and robot contests used to motivate and teach undergraduate AI.
-
ItemGlobal and Regional Path Planners for Integrated Planning and Navigation(Georgia Institute of Technology, 2005-12) Howard, Ayanna M. ; Seraji, Homayoun ; Werger, BarryThis paper presents a hierarchical strategy for field mobile robots that incorporates path planning at different ranges. At the top layer is a global path planner that utilizes gross terrain characteristics, such as hills and valleys, to determine globally safe paths through the rough terrain. This information is then passed via waypoints to a regional layer that plans appropriate navigation paths using regional terrain characteristics. The global and regional path planners share the same map information, but at different ranges. The motion recommendations from the regional layer are then combined with those of the reactive navigation layer to provide reactive control for the mobile robot. Details of the global and regional path planners are discussed, and simulation and experimental results are presented.