# 快速思考，选择性规划：面向高效数据驱动 MPC 的自适应审慎推理

> 原文链接：https://robotics.hellobike.com/papers/2026-09-26-2609-32591
> 频道：papers | 更新日期：2026-09-26 | 标签：arxiv, cs.RO
> 核心摘要：针对数据驱动MPC逐步规划开销高，Fast-TD-MPC自适应切换策略执行与测试时规划，并在103个连续控制任务验证。

# 快速思考，选择性规划：面向高效数据驱动 MPC 的自适应审慎推理

::: v-pre
> 原文链接：[Think Fast, Plan Selectively: Adaptive Deliberation for Efficient Data-Driven MPC](https://arxiv.org/abs/2609.32591)（arXiv 2609.32591 · 发布日期：2026-09-26）

**Think Fast, Plan Selectively: Adaptive Deliberation for Efficient Data-Driven MPC**

针对数据驱动MPC逐步规划开销高，Fast-TD-MPC自适应切换策略执行与测试时规划，并在103个连续控制任务验证。

## 核心要点

- 数据驱动MPC性能强，但逐步采样评估大量轨迹限制实时控制频率。
- 受双过程认知理论启发，方法判断哪些状态需要更多计算审慎。
- Fast-TD-MPC在快速策略执行与测试时规划之间自适应路由。
- 在103个任务中性能具竞争力、推理最高约快4倍，扰动下可回退规划。

**验证范围与局限**：验证范围为103个连续控制任务及外部扰动条件。

## 英文摘要 (Abstract)

Data-driven model predictive control (MPC) combines learned world models with online trajectory optimization, achieving strong performance in continuous control. However, the per-step cost of sampling and evaluating hundreds of candidate trajectories restricts deployment to control frequencies well below what real-time robotics demands. Motivated by the dual-process theory of human cognition, which distinguishes between fast, intuitive processing (System 1) and slower, deliberative reasoning (System 2), we ask whether every decision requires the same degree of computational deliberation. We propose Fast-TD-MPC, a lightweight framework that adaptively routes between fast policy execution and test-time planning, reserving costly deliberation for states where it is most needed. Fast-TD-MPC delivers competitive task performance across 103 continuous control tasks while achieving up to ~4x faster inference. Under external disturbances, Fast-TD-MPC selectively falls back to planning, maintaining robustness comparable to the original planner.

作者：Yi Xian Goh, Sze Jue Yang, Hao Luan
:::

## 原始参考来源 (Sources)

- [arXiv 2609.32591：Think Fast, Plan Selectively: Adaptive Deliberation for Efficient Data-Driven MPC](https://arxiv.org/abs/2609.32591)