Parallel Force-Position control mediated by tactile maps for robot contact tasks

This article introduces an extension of the original Parallel Force-Position control framework based on the use of tactile maps. Whole body skin systems for humanoid robots are considered a fundamental feature to improve contact interaction tasks by means of large scale tactile feedback. Stemming from previous work [1], the article describes how it is possible to extend the Parallel Force-Position control paradigm with the use of tactile maps, i.e., computational representation structures encoding the position of tactile elements located on the robot body surface. Furthermore, tactile maps are used to encode specific local features of contact trajectories, such as the desired exerted force at the contact point and the desired velocity. Results in simulation validate the approach.