Evaluation of three force-position hybrid control methods for a robot-based biological joint-testing system

Abstract

BackgroundRobot-based joint-testing systems (RJTS) can be used to perform unconstrained laxity tests, measuring the stiffness of a degree of freedom (DOF) of the joint at a fixed flexion angle while allowing the other DOFs unconstrained movement. Previous studies using the force-position hybrid (FPH) control method proposed by Fujie et al. (J Biomech Eng 115(3):211–7, 1993) focused on anterior/posterior tests. Its convergence and applicability on other clinically relevant DOFs such as valgus/varus have not been demonstrated. The current s1tudy aimed to develop a 6-DOF RJTS using an industrial robot, to propose two new force-position hybrid control methods, and to evaluate the performance of the methods and FPH in controlling the RJTS for anterior/posterior and valgus/varus laxity tests of the knee joint.MethodsAn RJTS was developed using an industrial 6-DOF robot with a 6-component load-cell attached at the effector. The performances of FPH and two new control methods, namely force-position alternate control (FPA) and force-position hybrid control with force-moment control (FPHFM), for unconstrained anterior/posterior and valgus/varus laxity tests were evaluated and compared with traditional constrained tests (CT) in terms of the number of control iterations, total time and the constraining forces and moments.ResultsAs opposed to CT, the other three control methods successfully reduced the constraining forces and moments for both anterior/posterior and valgus/varus tests, FPHFM being the best followed in order by FPA and FPH. FPHFM had root-mean-squared constraining forces and moments of less than 2.2 N and 0.09 Nm, respectively at 0° flexion, and 2.3 N and 0.14 Nm at 30° flexion. The corresponding values for FPH were 8.5 N and 0.33 Nm, and 11.5 N and 0.45 Nm, respectively. Given the same control parameters including the compliance matrix, FPHFM and FPA reduced the constraining loads of FPH at the expense of additional control iterations, and thus increased total time, FPA taking about 10 % longer than FPHFM.ConclusionsThe FPHFM would be the best choice among the methods considered when longer total time is acceptable in the intended clinical applications. The current results will be useful for selecting a force-position hybrid control method for unconstrained laxity tests using an RJTS.

Knee Kinematics Estimation Using Multi-Body Optimisation Embedding a Knee Joint Stiffness Matrix: A Feasibility Study

“…The axes of measurements in these experiments were consistent with those used for the joint angles and displacements estimated using MBO ( U = [ θ 1 θ 2 θ 3 d 1 d 2 d 3 ] T ). In particular, the control of the robotic arm was operated around the JCS axes ( e 1 , e 2 , e 3 ) [ 33 , 34 ]. On the tested cadaveric knee, the femur z-axis ( e 1 in the JCS) was aligned with the transepicondylar axis and pointed toward the medial epicondyle.…”

Methodology for Robotic In Vitro Testing of the Knee

“…Iteratively sequencing robot positioning commands with force measurements has been shown to take 78-689 seconds to achieve a single static loading condition for the knee joint. 36 Because of the long duration of this early adopted iterative force control technique, superposition testing became a favored testing protocol. With superposition testing, the position and orientation of each static loading condition could be learned and stored.…”