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

I. INTRODUCTION State estimation for continuum robotics is a comparatively underdeveloped area. Several dynamic estimators have been proposed. In [3], [4] the authors make use of Kalman filtering techniques. A typical assumption these dynamic estimators make is that of a constant curvature shape, further limiting the accuracy of the method in pursuit of model simplicity. In mobile robotics, [5] introduces the use of Gaussian-process (GP) regression for Figure 1: A continuum robot is shown at multiple timesteps performing continuous-time state estimation of a mobile with increasing opacity.