Deflection-based force sensing for continuum robots: A probabilistic approach

The inherent flexibility of continuum robots allows them to interact with objects in a safe and compliant way. This flexibility also makes it possible to use robot deflection to estimate external forces applied to the robot. This “intrinsic force sensing” concept is particularly useful for thin continuum robots where application constraints preclude the use of traditional force sensors. This paper describes an Extended Kalman Filter approach to estimate forces applied at the tip of a continuum robot using only uncertain pose measurements and a kinematic-static model of the robot with uncertainty.