Contact State Estimation using Multiple Model Estimation and Hidden Markov Models

This paper presents an approach to estimating the contact state between a robot and its environment during task execution. Contact states are modeled by constraint equations parameterized by time-dependent sensor data and timeindependent object properties. At each sampling time, multiple model estimation is used to assess the most likely contact state. The assessment is performed by a Hidden Markov Model, which combines a measure of how well each set of constraint equations fit the sensor data with the probability of specific contact state transitions. The latter is embodied in a task-based contact state network. The approach is illustrated for a three dimensional peg-in-hole insertion using a tabletop manipulator robot. Using only position sensing, the contact state sequence is successfully estimated. Property estimates are obtained for the peg dimensions as well as the hole position and orientation.