Uses spectral decomposition of inverse dynamics to enable real-time planning of long-horizon robotic manipulation tasks (10+ contact modes).
March 31, 2026
Original Paper
Spectral Decomposition of Inverse Dynamics for Fast Exploration in Model-Based Manipulation
arXiv · 2603.27796
The Takeaway
Solves a major bottleneck in robotics: long-duration planning with complex contact dynamics. It generates 45-second plans in just 15 seconds of computation, where traditional model-based planners typically fail or take orders of magnitude longer.
From the abstract
Planning long duration robotic manipulation sequences is challenging because of the complexity of exploring feasible trajectories through nonlinear contact dynamics and many contact modes. Moreover, this complexity grows with the problem's horizon length. We propose a search tree method that generates trajectories using the spectral decomposition of the inverse dynamics equation. This equation maps actuator displacement to object displacement, and its spectrum is efficient for exploration becaus