Space-Time Continuum: Continuous Shape and Time State Estimation for Flexible Robots

Abstract

This extended abstract introduces a novel method for continuous state estimation of continuum robots. We formulate the estimation problem as a factor-graph optimization problem using a novel Gaussian-process prior that is parameterized over both arclength and time. We use this to introduce the first continuous-time-and-space state estimation method for continuum robots.

Figures (3)

• Figure 1: A continuum robot is shown at multiple timesteps with increasing opacity. The prototype has four types of asynchronous sensors: gyroscopes, 6 DOF pose sensors, a strain sensor, and an external motion capture.
• Figure 2: Factor graph for space-time estimation of a continuum robot. With the GP formulation, continuous querying in both arclength and time is available in $O(1)$ time.
• Figure 3: Example state estimation result of the proposed approach using a simulated tendon-driven continuum robot. Sensing includes noisy strain measurements at every estimation node as well as two noisy discrete position measurements. The state estimate includes the smooth pose and strain of the continuum robot over both space (arclength) $s$ and time $t$. The state mean is shown in black and via coloured coordinate frames. The state uncertainty is visualized using blue covariance ellipsoids.