Unsupervised Contact Learning for Humanoid Estimation and Control
This work addresses contact estimation for humanoid robots in rough, low-friction environments, but it is incremental as it builds on existing kinematics-based methods.
The paper tackles the problem of estimating contact states for humanoid robots using unsupervised learning from proprioceptive sensors, and shows that their method improves base state estimation performance in simulation with realistic sensor noise and challenging terrain.
This work presents a method for contact state estimation using fuzzy clustering to learn contact probability for full, six-dimensional humanoid contacts. The data required for training is solely from proprioceptive sensors - endeffector contact wrench sensors and inertial measurement units (IMUs) - and the method is completely unsupervised. The resulting cluster means are used to efficiently compute the probability of contact in each of the six endeffector degrees of freedom (DoFs) independently. This clustering-based contact probability estimator is validated in a kinematics-based base state estimator in a simulation environment with realistic added sensor noise for locomotion over rough, low-friction terrain on which the robot is subject to foot slip and rotation. The proposed base state estimator which utilizes these six DoF contact probability estimates is shown to perform considerably better than that which determines kinematic contact constraints purely based on measured normal force.