Parallel KKT Solver in PIQP for Multistage Optimization footnoteinfo This work was supported as a part of NCCR Automation, a National Centre of Competence in Research, funded by the Swiss National Science Foundation (grant number 51NF40_225155).
Fenglong Song, Roland Schwan, Yuwen Chen, Colin N. Jones · 2025
Read Parallel KKT Solver in PIQP for Multistage Optimization footnoteinfo This work was supported as a part of NCCR Automation, a National Centre of Competence in Research, funded by the Swiss National Science Foundation (grant number 51NF40_225155). by Fenglong Song, Roland Schwan, Yuwen Chen, Colin N. Jones (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-00946.txt.
Description
Fenglong Song, Roland Schwan, Yuwen Chen, and Colin Jones speed up a core computation inside the PIQP optimization solver. Multistage problems, such as model predictive control and racing trajectory planning, repeatedly produce Karush-Kuhn-Tucker systems, the linear equations an interior point method must solve at every step.
Their contribution is a parallel Cholesky factorization and triangular solve designed for the block structured matrices these problems generate, with a banded part plus an arrow shaped border. It is implemented as a new backend for PIQP and released as open source. On chain of masses benchmarks and on a minimum curvature racing line problem, the solver shows substantial speed gains over other state of the art solvers.
Catalog record
| CREATOR | Fenglong Song, Roland Schwan, Yuwen Chen, Colin N. Jones |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-00946