A Methodology for Benchmarking Force Control Algorithms

Abstract

Force control is nowadays a mature technology, widespread in modern robotics systems and commercialized within several application domains. However, assessing the performance of a force-controlled system is not a trivial task because it may be strongly influenced by the environment dynamics. Exerting a force on a soft environment is different from exerting a force on a rigid environment. Indeed, the same force-controlled robot can have different force responses in different environments and a standardized and comprehensive method to assess the performance of a force-controlled system is not available yet. This paper, as part of the Forecast project, aims at filling this gap by proposing a benchmarking method able to define a comprehensive score for a given force-controlled system accounting for its sensitivity to environment uncertainties and variations. This paper describes such benchmarking methodology which allows to compare force control algorithms on a common ground and to determine the preferable control solutions in a specific application domain.