Algorithms for multi-agent robotic assembly planning
Abstract/Contents
- Abstract
- This thesis presents algorithms for multi-agent robotic assembly planning in automated manufacturing contexts. Our work touches on many pieces of the "factory autonomy stack." The first contribution of the thesis is the introduction of a discrete factory problem formulation that includes time-extended multi robot task allocation, inter-task precedence constraints, and collision-avoidance constraints. We propose an efficient method to solve such problems. The two keys to our algorithm's efficiency are its decoupling of task assignment and route planning, and its ability to take advantage of situations where some robots can be delayed in their own schedule without causing any negative impact on the factory's overall performance. The next major contribution of the thesis is a family of replanning algorithms geared toward an online version of our discrete factory problem. In the online setting, the factory command center periodically receives new manufacturing workloads that must be quickly incorporated into the overall schedule. We show through an extensive set of experiments that our replanning methods are applicable across a wide range of problems. Moreover, our proposed methods can be applied in such a way that the factory will never have to freeze while waiting to receive an updated plan. Our final contribution is a proof-of-concept system for large scale multi-robot assembly planning with assemblies and raw materials of arbitrary shapes and sizes. Our system starts with raw materials and a bare-bones set of instructions about how those materials fit together. The planner then synthesizes a construction plan, which defines how each payload will be carried (by one or more robots), where each assembly and subassembly will be built, and which specific robots will be assigned to each solo and collaborative transport task. Finally, a reactive collision-avoidance control policy enables the robots to execute the construction plan in a distributed manner. We demonstrate in simulation that our system can synthesize---in minutes---construction plans for assemblies with hundred of components. Though we do not address all of the relevant ``real-world'' considerations that surround multi robot manufacturing, our work is a small step toward large scale automated construction with mobile robots
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2021; ©2021 |
| Publication date | 2021; 2021 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Brown, Kyle Jordan |
|---|---|
| Degree supervisor | Darve, Eric |
| Degree supervisor | Kochenderfer, Mykel J, 1980- |
| Thesis advisor | Darve, Eric |
| Thesis advisor | Kochenderfer, Mykel J, 1980- |
| Thesis advisor | Pavone, Marco, 1980- |
| Thesis advisor | Schwager, Mac |
| Degree committee member | Pavone, Marco, 1980- |
| Degree committee member | Schwager, Mac |
| Associated with | Stanford University, Department of Mechanical Engineering |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Kyle J. Brown |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering |
| Thesis | Thesis Ph.D. Stanford University 2021 |
| Location | https://purl.stanford.edu/qf931ph2309 |
Access conditions
- Copyright
- © 2021 by Kyle Jordan Brown
- License
- This work is licensed under a Creative Commons Attribution 3.0 Unported license (CC BY).
Also listed in
Loading usage metrics...