Evaluation of the Performance and Cost of Cloud-Based Robot Motion Planning
Public Deposited- Last Modified
- February 26, 2019
- Creator
-
Lynn, Jonathan
- Affiliation: College of Arts and Sciences, Department of Computer Science
- Abstract
- We evaluate and analyze the empirical performance and financial costs of using cloud computing to accelerate robot motion planning. In prior work, Ichnowski et al. proposed a method that splits the computation of a robot’s motion plan between the robot’s own low-powered processor and a cloud-based high-powered compute-optimized server [1]. We implement this method using the compute service offered by Amazon Elastic Compute Cloud (Amazon EC2) on a Fetch robot [2] using 8 degrees of freedom to complete a task in a dynamic environment, as would be required for a consumer-based home-assistance robot. Computing across different data centers around the world, we investigate the network time elapsed during the plan-transmission process, the time taken to complete a task with the robot, and measures of planner performance including the number of vertices and edges generated for each roadmap created during the planning process. Finally, we use these results to estimate the cost model of such pay-per-use cloud services used in this new planning method to evaluate its economic feasibility and desirability compared to a traditional capital expense model.
- Date of publication
- spring 2017
- Keyword
- DOI
- Resource type
- Rights statement
- In Copyright
- Note
- Funding: None
- Advisor
- Alterovitz, Ron
- Degree
- Bachelor of Science
- Academic concentration
- Computer Science
- Honors level
- Honors
- Degree granting institution
- University of North Carolina at Chapel Hill
- Graduation year
- 2017
- Language
- English
Relations
- Parents:
This work has no parents.
Items
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
2017_Lynn.pdf | 2019-05-18 | Public | Download |