← 研踪 PaperTrailarXiv:2611.19466 · v1
Successive Convexification in Learned World Models for Safe Offline Planning
首次提交 2026-11-14 23:40
版本更新 2026-11-14 23:40
雷达观测 2026-11-17 12:06
摘要
World models learned from offline data allow an agent to plan without further interaction with the environment, but the sampling-based planners used with them handle safety constraints only by scoring or rejecting sampled action sequences. We instead pose safe planning in the latent space of a learned world model as a successive convex program (SCP), in which a learned discrete-time control barrier function is imposed as a constraint at every step of the horizon. SCP requires smooth dynamics, so we build the networks from smoothed MaxMin activations and spectrally normalised layers, and bound the resulting linearisation error in terms of the Lipschitz bound and the smoothing width. As SCP is a local method, we start it from a sampled plan. On two Safety-Gymnasium navigation tasks with offline data, this planner has fewer violations than MPPI and CEM drawing 128 times as many samples at equal wall-clock time in our main comparison, though not significantly, and earns greater return at longer horizons, and is much safer than SCP without a sampled start. SCP converged more often on Lipschitz-bounded models. Unlike sampling, it also reports at each step whether its plan is a feasible, first-order optimal KKT point of the planning problem in the model, or that no feasible plan was found nearby. Our implementation is publicly available at https://github.com/youngers2006/scwm.
作者备注:28 pages, 13 figures. Under review at L4DC 2027
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-14 23:40 · 最新观测版本
Successive Convexification in Learned World Models for Safe Offline Planning
World models learned from offline data allow an agent to plan without further interaction with the environment, but the sampling-based planners used with them handle safety constraints only by scoring or rejecting sampled action sequences. We instead pose safe planning in the latent space of a learned world model as a successive convex program (SCP), in which a learned discrete-time control barrier function is imposed as a constraint at every step of the horizon. SCP requires smooth dynamics, so we build the networks from smoothed MaxMin activations and spectrally normalised layers, and bound the resulting linearisation error in terms of the Lipschitz bound and the smoothing width. As SCP is a local method, we start it from a sampled plan. On two Safety-Gymnasium navigation tasks with offline data, this planner has fewer violations than MPPI and CEM drawing 128 times as many samples at equal wall-clock time in our main comparison, though not significantly, and earns greater return at longer horizons, and is much safer than SCP without a sampled start. SCP converged more often on Lipschitz-bounded models. Unlike sampling, it also reports at each step whether its plan is a feasible, first-order optimal KKT point of the planning problem in the model, or that no feasible plan was found nearby. Our implementation is publicly available at https://github.com/youngers2006/scwm.
查看此版本解读 →查看此版本原文 ↗