Physically Constrained PCB Placement Using Deep Reinforcement Learning
This thesis provides an in depth exploration of Reinforcement Learning (RL) based PCB component placement with emphasis on physically verified placements. Unlike prior methods that rely on heuristic proxies for placement quality, this work focuses entirely on routing based metrics that result in fun...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2022
|
| Online Access: | https://hdl.handle.net/1721.1/139247 |
| _version_ | 1826207788439699456 |
|---|---|
| author | Crocker, Peter |
| author2 | Chan, Vincent W.S. |
| author_facet | Chan, Vincent W.S. Crocker, Peter |
| author_sort | Crocker, Peter |
| collection | MIT |
| description | This thesis provides an in depth exploration of Reinforcement Learning (RL) based PCB component placement with emphasis on physically verified placements. Unlike prior methods that rely on heuristic proxies for placement quality, this work focuses entirely on routing based metrics that result in functioning placements without the need for fine tuning. Additionally, this exploration considers true use cases of PCB auto-placement where a human-in-the-loop pre-places a set list of components and the auto-placer places the remaining. This is achieved by first restricting the placement domain to only ring placements; a domain where routing calculations become accessible. Within the ring placement domain, an RL agent is trained to place components on a simulated PCB canvas such that there are no component overlaps or wire crossings upon manufacture. Through the use of an unbounded reward system, the agent is trained progressively with PCB complexity gradually increasing as training steps are run. The resulting placements are robust to varying numbers of components as well as component shape and size. Finally, this thesis concludes with a discussion about further work and challenges facing the future of PCB auto-placement. |
| first_indexed | 2024-09-23T13:54:58Z |
| format | Thesis |
| id | mit-1721.1/139247 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T13:54:58Z |
| publishDate | 2022 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1392472022-01-15T03:23:50Z Physically Constrained PCB Placement Using Deep Reinforcement Learning Crocker, Peter Chan, Vincent W.S. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science This thesis provides an in depth exploration of Reinforcement Learning (RL) based PCB component placement with emphasis on physically verified placements. Unlike prior methods that rely on heuristic proxies for placement quality, this work focuses entirely on routing based metrics that result in functioning placements without the need for fine tuning. Additionally, this exploration considers true use cases of PCB auto-placement where a human-in-the-loop pre-places a set list of components and the auto-placer places the remaining. This is achieved by first restricting the placement domain to only ring placements; a domain where routing calculations become accessible. Within the ring placement domain, an RL agent is trained to place components on a simulated PCB canvas such that there are no component overlaps or wire crossings upon manufacture. Through the use of an unbounded reward system, the agent is trained progressively with PCB complexity gradually increasing as training steps are run. The resulting placements are robust to varying numbers of components as well as component shape and size. Finally, this thesis concludes with a discussion about further work and challenges facing the future of PCB auto-placement. M.Eng. 2022-01-14T14:59:15Z 2022-01-14T14:59:15Z 2021-06 2021-06-17T20:13:06.570Z Thesis https://hdl.handle.net/1721.1/139247 In Copyright - Educational Use Permitted Copyright MIT http://rightsstatements.org/page/InC-EDU/1.0/ application/pdf Massachusetts Institute of Technology |
| spellingShingle | Crocker, Peter Physically Constrained PCB Placement Using Deep Reinforcement Learning |
| title | Physically Constrained PCB Placement Using Deep Reinforcement Learning |
| title_full | Physically Constrained PCB Placement Using Deep Reinforcement Learning |
| title_fullStr | Physically Constrained PCB Placement Using Deep Reinforcement Learning |
| title_full_unstemmed | Physically Constrained PCB Placement Using Deep Reinforcement Learning |
| title_short | Physically Constrained PCB Placement Using Deep Reinforcement Learning |
| title_sort | physically constrained pcb placement using deep reinforcement learning |
| url | https://hdl.handle.net/1721.1/139247 |
| work_keys_str_mv | AT crockerpeter physicallyconstrainedpcbplacementusingdeepreinforcementlearning |