Rory Hector
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Rory Hector has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Gokarna Sharma
Kent State University, Kent, USA
2 shared publications - Jerry L. Trahan
Louisiana State University,Division of Electrical Computer Engineering,Baton Rouge,LA,USA
2 shared publications - Ramachandran Vaidyanathan
Louisiana State University, Baton Rouge, LA, USA
2 shared publications Asif Mehmood
Division of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, LA, United States
1 shared publicationMuhammad Umar
Division of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, LA, United States
1 shared publicationShuvra Bhattacharyya
Division of Electrical and Computer Engineering, Louisiana State University, Baton Rouge, LA, United States
1 shared publication
Latest Publications
- Open Access
- 10 June 2022
- Published by Louisiana State University Libraries
Practical Considerations and Applications for Autonomous Robot Swarms
In recent years, the study of autonomous entities such as unmanned vehicles has begun to revolutionize both military and civilian devices. One important research focus of autonomous entities has been coordination problems for autonomous robot swarms. Traditionally, robot models are used for algorithms that account for the minimum specifications needed to operate the swarm. However, these theoretical models also gloss over important practical details. Some of these details, such as time, have been considered before (as epochs of execution). In this dissertation, we examine these details in the context of several problems and introduce new performance measures to capture practical details. Specifically, we introduce three new metrics: (1) the distance complexity (reflecting power usage and wear-and-tear of robots), (2) the spatial complexity (reflecting the space needed for the algorithm to work), and (3) local computational complexity (reflecting the computational requirements for each robot in the swarm). We apply these metrics in the study of some well-known and important problems, such as Complete Visibility and Arbitrary Pattern Formation. We also introduce and study a new problem, Doorway Egress, that captures the essence of a swarm’s navigation through restricted spaces. First, we examine the distance and spatial complexity used across a class of Complete Visibility algorithms. Second, we provide algorithms for Complete Visibility on an integer plane, including some that are asymptotically optimal in terms of time, distance complexity, and spatial complexity. Third, we introduce the problem of Doorway Egress and provide algorithms for a variety of robot swarm models with various optimalities. Finally, we provide an optimal algorithm for Arbitrary Pattern Formation on the grid.
- 10 June 2022
- Published by Louisiana State University Libraries
Open AccessPractical Considerations and Applications for Autonomous Robot Swarms
In recent years, the study of autonomous entities such as unmanned vehicles has begun to revolutionize both military and civilian devices. One important research focus of autonomous entities has been coordination problems for autonomous robot swarms. Traditionally, robot models are used for algorithms that account for the minimum specifications needed to operate the swarm. However, these theoretical models also gloss over important practical details. Some of these details, such as time, have been considered before (as epochs of execution). In this dissertation, we examine these details in the context of several problems and introduce new performance measures to capture practical details. Specifically, we introduce three new metrics: (1) the distance complexity (reflecting power usage and wear-and-tear of robots), (2) the spatial complexity (reflecting the space needed for the algorithm to work), and (3) local computational complexity (reflecting the computational requirements for each robot in the swarm). We apply these metrics in the study of some well-known and important problems, such as Complete Visibility and Arbitrary Pattern Formation. We also introduce and study a new problem, Doorway Egress, that captures the essence of a swarm’s navigation through restricted spaces. First, we examine the distance and spatial complexity used across a class of Complete Visibility algorithms. Second, we provide algorithms for Complete Visibility on an integer plane, including some that are asymptotically optimal in terms of time, distance complexity, and spatial complexity. Third, we introduce the problem of Doorway Egress and provide algorithms for a variety of robot swarm models with various optimalities. Finally, we provide an optimal algorithm for Arbitrary Pattern Formation on the grid.
- 10 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 33 (12) , 3532-3545
Optimal Convex Hull Formation on a Grid by Asynchronous Robots With Lights
We consider the distributed setting of n autonomous mobile robots that operate in Look-Compute-Move cycles and communicate with other robots using a constant number of colored lights (the {\em robots with lights} model). We assume obstructed visibility where collinear robots do not see each other. In addition, we consider a grid-based terrain embedded in the 2-dimensional Euclidean plane. The {\sc Convex Hull Formation} problem is to relocate the $n$ robots (starting at arbitrary, but distinct, initial positions) so that each robot is positioned on a vertex of a convex hull. In this paper, we provide a framework for solving {\sc Convex Hull Formation}. We then provide four asynchronous algorithms under this framework. Key measures of the algorithms' performance include the time taken and the space occupied. The presented algorithms are randomized and their time bounds hold with high probability. The first $O(\max\{n^2,D\})$ -time, $O({n^2})$ -perimeter, and $O({n^3})$ -area algorithm serves to introduce key ideas, where D is the diameter of the initial configuration. The subsequent algorithms, differing in computational requirements, run in $O(\max\{n^{\frac{3}{2}},D\})$ time with a perimeter of $O(n^{\frac{3}{2}})$ and area of $O(n^{3})$ . We also prove lower bounds of $\Omega(n^{\frac{3}{2}})$ for time and perimeter and $\Omega(n^{3})$ for area, for any {\sc Convex Hull Formation} algorithm; i.e., our $O(\max\{n^{\frac{3}{2}},D\})-$ time algorithm is optimal in time, perimeter, and area.- 10 March 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 33 (12) , 3532-3545
Optimal Convex Hull Formation on a Grid by Asynchronous Robots With Lights
We consider the distributed setting of n autonomous mobile robots that operate in Look-Compute-Move cycles and communicate with other robots using a constant number of colored lights (the {\em robots with lights} model). We assume obstructed visibility where collinear robots do not see each other. In addition, we consider a grid-based terrain embedded in the 2-dimensional Euclidean plane. The {\sc Convex Hull Formation} problem is to relocate the $n$ robots (starting at arbitrary, but distinct, initial positions) so that each robot is positioned on a vertex of a convex hull. In this paper, we provide a framework for solving {\sc Convex Hull Formation}. We then provide four asynchronous algorithms under this framework. Key measures of the algorithms' performance include the time taken and the space occupied. The presented algorithms are randomized and their time bounds hold with high probability. The first $O(\max\{n^2,D\})$ -time, $O({n^2})$ -perimeter, and $O({n^3})$ -area algorithm serves to introduce key ideas, where D is the diameter of the initial configuration. The subsequent algorithms, differing in computational requirements, run in $O(\max\{n^{\frac{3}{2}},D\})$ time with a perimeter of $O(n^{\frac{3}{2}})$ and area of $O(n^{3})$ . We also prove lower bounds of $\Omega(n^{\frac{3}{2}})$ for time and perimeter and $\Omega(n^{3})$ for area, for any {\sc Convex Hull Formation} algorithm; i.e., our $O(\max\{n^{\frac{3}{2}},D\})-$ time algorithm is optimal in time, perimeter, and area.- 9 November 2021
- Published by Springer Nature
- p. 489-494
On Optimal Doorway Egress by Autonomous Robots
- Open Access
- 27 July 2021
- Published by Frontiers Media SA in Frontiers in Sustainable Cities
- Vol. 3
Scalable Object Detection for Edge Cloud Environments
Object detection is an important problem in a wide variety of computer vision applications for sustainable smart cities. Deep neural networks (DNNs) have attracted increasing interest in object detection due to their potential to provide high accuracy detection performance in challenging scenarios. However, DNNs involve high computational complexity and are therefore challenging to deploy under the tighter resource constraints of edge cloud environments compared to more resource-abundant platforms, such as conventional cloud computing platforms. Moreover, the monolithic structure of conventional DNN implementations limits their utility under the dynamically changing operational conditions that are typical in edge cloud computing. In this paper, we address these challenges and limitations of conventional DNN implementation techniques by introducing a new resource-adaptive scheme for DNN-based object detection. This scheme applies the recently-introduced concept of elastic neural networks, which involves the incorporation of multiple outputs within intermediate stages of the neural network backbone. We demonstrate a novel elastic DNN design for object detection, and we show how other methods for streamlining resource requirements, in particular network pruning, can be applied in conjunction with the proposed elastic network approach. Through extensive experiments, we demonstrate the ability of our methods to efficiently trade-off computational complexity and object detection accuracy for scalable deployment.
- 27 July 2021
- Published by Frontiers Media SA in Frontiers in Sustainable Cities
- Vol. 3
Open AccessScalable Object Detection for Edge Cloud Environments
Object detection is an important problem in a wide variety of computer vision applications for sustainable smart cities. Deep neural networks (DNNs) have attracted increasing interest in object detection due to their potential to provide high accuracy detection performance in challenging scenarios. However, DNNs involve high computational complexity and are therefore challenging to deploy under the tighter resource constraints of edge cloud environments compared to more resource-abundant platforms, such as conventional cloud computing platforms. Moreover, the monolithic structure of conventional DNN implementations limits their utility under the dynamically changing operational conditions that are typical in edge cloud computing. In this paper, we address these challenges and limitations of conventional DNN implementation techniques by introducing a new resource-adaptive scheme for DNN-based object detection. This scheme applies the recently-introduced concept of elastic neural networks, which involves the incorporation of multiple outputs within intermediate stages of the neural network backbone. We demonstrate a novel elastic DNN design for object detection, and we show how other methods for streamlining resource requirements, in particular network pruning, can be applied in conjunction with the proposed elastic network approach. Through extensive experiments, we demonstrate the ability of our methods to efficiently trade-off computational complexity and object detection accuracy for scalable deployment.