Hierarchical Policy Blending As Optimal Transport
This work addresses the challenge of efficient and safe policy blending in robot control, offering a novel approach that could enable new applications, though it appears incremental in its use of optimal transport.
The paper tackles the problem of blending low-level reactive expert policies for multi-agent robot control by introducing a hierarchical planning layer based on optimal transport, resulting in improved task success and safety while outperforming state-of-the-art baselines in various scenarios.
We present hierarchical policy blending as optimal transport (HiPBOT). HiPBOT hierarchically adjusts the weights of low-level reactive expert policies of different agents by adding a look-ahead planning layer on the parameter space. The high-level planner renders policy blending as unbalanced optimal transport consolidating the scaling of the underlying Riemannian motion policies. As a result, HiPBOT effectively decides the priorities between expert policies and agents, ensuring the task's success and guaranteeing safety. Experimental results in several application scenarios, from low-dimensional navigation to high-dimensional whole-body control, show the efficacy and efficiency of HiPBOT. Our method outperforms state-of-the-art baselines -- either adopting probabilistic inference or defining a tree structure of experts -- paving the way for new applications of optimal transport to robot control. More material at https://sites.google.com/view/hipobot