Sample-Based Hybrid Mode Control: Asymptotically Optimal Switching of Algorithmic and Non-Differentiable Control Modes
Yilang Liu, Haoxiang You, Ian Abraham · 2025
Read Sample-Based Hybrid Mode Control: Asymptotically Optimal Switching of Algorithmic and Non-Differentiable Control Modes by Yilang Liu, Haoxiang You, Ian Abraham (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-2510-19074.txt.
Description
Yilang Liu, Haoxiang You, and Ian Abraham present a sample-based method for control problems that switch between very different modes. Some modes are algorithmic or cannot be differentiated, so standard optimization that follows gradients cannot decide which mode to use, when to switch, or how long to stay.
Their method treats mode choice as an integer optimization problem and searches it efficiently with samples, while keeping strong performance guarantees. It can stitch together complex behaviors from simpler policies to complete hard tasks. Tests on real robots show the approach works outside simulation, a key check for switching control schemes.
Catalog record
| CREATOR | Yilang Liu, Haoxiang You, Ian Abraham |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-19074