Research
A Robust Task-Level Control Architecture for Learned Dynamical Systems
arXiv:2511.09790v2 Announce Type: replace-cross Abstract: Dynamical system (DS)-based learning from demonstration (LfD) is a powerful tool for generating motion plans in the operation ('task') space o
arXiv:2511.09790v2 Announce Type: replace-cross Abstract: Dynamical system (DS)-based learning from demonstration (LfD) is a powerful tool for generating motion plans in the operation ('task') space of robotic systems. However, realizing generated motion plans is often compromised by a "task-execution mismatch", where unmodeled dynamics, persistent disturbances, and system latency cause the robot's task-space state to diverge from the desired state. We propose a novel task-level robust control architecture, L1-augmented Dynamical Systems (L1-DS), that explicitly handles the task-execution mismatch in tracking a nominal motion plan generated by any DS-based LfD scheme. Our framework augments any DS-based LfD model with a nominal stabilizing controller and an L1 adaptive controller. Furthermore, we introduce a windowed Dynamic Time Warping (DTW)-based target selector, which enables the nominal stabilizing controller to handle temporal misalignment for improved phase-consistent tracking. We demonstrate the efficacy of our architecture on the LASA and IROS handwriting datasets.
Related
- Zero-Shot Function Encoder-Based Differentiable Predictive Control
- Goal-oriented learning of stochastic differential equations using error bounds on path-space observables
- Model Predictive Path Integral PID Control for Learning-Based Path Following
- Adversarial dynamical systems characterize when data-driven learning succeeds or fails
- Distributionally Robust Regret Optimal Control Under Moment-Based Ambiguity Sets
Source: arXiv cs.LG | 2026-08-26