Manifold Representations for State Estimation in Contact Manipulation

Abstract

We investigate the problem of using contact sensors to estimate the con- figuration of an object during manipulation. Contact sensing is very discriminative by nature and, therefore, the set of object configurations that a sensor constitutes a lower-dimensional manifold in the state space of the object. This causes conventional state estimation methods, such as particle filters, to perform poorly during periods of contact. The manifold particle filter addresses this problem by sampling particles directly from the contact manifold. When it exists, we can sample these particles from an analytic representation of the contact manifold. We present two alternative sample-based contact manifold representations that make no assumptions about the object-hand geometry: rejection sampling and trajectory rollouts. The rejection sampling representation distributes uniformly in the space surrounding the manifold, while the trajectory rollout representation concentrates samples on the regions of the manifold that we are most likely to encounter during execution. We discuss theoretical considerations behind these three representations and compare their performance in an extensive suite of simulation experiments. We show that all three representations enable the manifold particle filter to outperform the conventional particle filter. Additionally, we show that the trajectory rollout representation performs similarly to the analytic method despite its relative simplicity.