${\tt MORALS}$: Analysis of High-Dimensional Robot Controllers via Topological Tools in a Latent Space

TL;DR

MORALS introduces a novel method combining auto-encoders and Morse Graphs to efficiently estimate regions of attraction for high-dimensional robot controllers in learned latent spaces, enhancing safety analysis.

Contribution

It presents MORALS, a new approach that applies topological tools in latent spaces to analyze high-dimensional robot controllers, overcoming previous scalability limitations.

Findings

Effective RoA estimation for high-dimensional systems

Demonstrated on a 67-dim humanoid robot and a 96-dim manipulator

Data-efficient and capable of predicting attractors

Abstract

Estimating the region of attraction (${\tt RoA}$) for a robot controller is essential for safe application and controller composition. Many existing methods require a closed-form expression that limit applicability to data-driven controllers. Methods that operate only over trajectory rollouts tend to be data-hungry. In prior work, we have demonstrated that topological tools based on ${\it Morse Graphs}$ (directed acyclic graphs that combinatorially represent the underlying nonlinear dynamics) offer data-efficient ${\tt RoA}$ estimation without needing an analytical model. They struggle, however, with high-dimensional systems as they operate over a state-space discretization. This paper presents ${\it Mo}$rse Graph-aided discovery of ${\it R}$egions of ${\it A}$ttraction in a learned ${\it L}$atent ${\it S}$pace (${\tt MORALS}$). The approach combines auto-encoding neural networks with Morse Graphs. ${\tt MORALS}$ shows promising predictive capabilities in estimating attractors and their ${\tt RoA}$s for data-driven controllers operating over high-dimensional systems, including a 67-dim humanoid robot and a 96-dim 3-fingered manipulator. It first projects the dynamics of the controlled system into a learned latent space. Then, it constructs a reduced form of Morse Graphs representing the bistability of the underlying dynamics, i.e., detecting when the controller results in a desired versus an undesired behavior. The evaluation on high-dimensional robotic datasets indicates data efficiency in ${\tt RoA}$ estimation.