论文/arXiv 2609.27463·1 个月前

安全过滤的分布式 Koopman-MPC

针对DMPC丢包会同时丢失预测与避碰约束的问题，提出Koopman-MPC与局部感知QP安全投影分离的方法，并在仓储多机器人仿真等测试中验证。

AI 整理

• 本体 SDK 与控制
• 仿真与训练
• 端侧部署

原文链接：Safety-Filtered Distributed Koopman-MPC（arXiv 2609.27463 · 发布日期：2026-09-23）

Safety-Filtered Distributed Koopman-MPC

针对DMPC丢包会同时丢失预测与避碰约束的问题，提出Koopman-MPC与局部感知QP安全投影分离的方法，并在仓储多机器人仿真等测试中验证。

核心要点

• DMPC常由邻居轨迹构造预测和碰撞约束，丢包可能同时移除两者。
• 方法将两种角色分离：接收轨迹驱动Koopman-MPC，局部感知与货架几何定义硬约束QP投影输入。
• 给出采样间隔分离定理、同步局部可行性的精确max-min测试，以及切换交互图的感知半径条件。
• 八机器人仓储仿真中完整控制器20/20试验无碰并达到160/160目标，无最终投影的预测Koopman-MPC仅1/20无碰。

验证范围与局限：验证范围包括八机器人仓储仿真、五流车队扫描，以及有界感知和差速驱动压力测试。

英文摘要 (Abstract)

Distributed model predictive control (DMPC) often constructs both predictions and collision constraints from neighbor trajectories, so packet loss can remove both. We separate these roles: received trajectories drive Koopman-MPC, while local sensing and shelf geometry define a hard-constrained quadratic program (QP) that projects the applied input. Its radial demand is the least constant acceleration that keeps a supporting-plane clearance nonnegative throughout one zero-order-hold interval. Complementary pair rows recover the coupled demand without exchanging safety decisions. We give an intersample separation theorem under bounded snapshot and directional plant errors, an exact max-min test for simultaneous local feasibility, and a sensing-radius condition for switching interaction graphs. Anticipatory high-order rows may be relaxed for performance, but the finite-hold rows contain no safety slack. Matched eight-robot warehouse simulations use a frozen Koopman model, nonlinear drift, bounded inputs and speed, shelf constraints, a 120 ms control period, and packet dropout. The full controller is collision-free in 20/20 matched trials and reaches 160/160 robot goals; predictive Koopman-MPC without the final projection is collision-free in 1/20 trials. All 38,400 full-method hard-row sets pass the online feasibility test, and every local QP solves. Five-stream fleet sweeps are collision-free and hard-row feasible through 16 robots; the 20-robot boundary fails only after the online margin turns negative, while the reconstructed per-agent critical path remains below the sampling period. Bounded-sensing and differential-drive tests provide additional deployment stress.

作者：Shengjun Zhang, Wenhao Li, Zhenxin Lin, Zhenglong Sun

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2609.27463：Safety-Filtered Distributed Koopman-MPC访问 2026-10-21