基于力/力矩的机器人控制中的一个问题是末端执行器质量产生的作用力和力矩。描述了一种克服该问题的方法。该方法结合了末端效应器质量特性的模型、末端效应器方向的连续更新和重力矢量。通过检测传感器数据，对末端执行器进行几次重新定向，自动建立质量特性模型。通过减去计算出的质量贡献，对每组传感器数据进行补偿。该方法已实施并静态测试，误差约为末端执行器质量的+/-2%。

A PROBLEM IN FORCE/TORQUE-BASED ROBOT CONTROL IS THE CONTRIBUTING FORCES AND TORQUES FROM THE END EFFECTOR MASS. A METHOD TO OVERCOME THIS PROBLEM IS DESCRIBED. THE METHOD INCORPORATES A MODEL OF THE END EFFECTOR MASS PROPERTIES, CONTINUOUS UPDATE OF END EFFECTOR ORIENTATION, AND THE GRAVITATION VECTOR. THE MASS PROPERTY MODEL IS AUTOMATICALLY BUILT BY PERFORMING A FEW REORIENTATIONS OF THE END EFFECTOR, DETECTING SENSOR DATA. THE COMPENSATION IS DONE FOR EACH SET OF SENSOR DATA, BY SUBTRACTING THE CALCULATED MASS CONTRIBUTION. THE METHOD HAS BEEN IMPLEMENTED AND TESTED STATICALLY WITH AN ERROR OF APPROXIMATELY +/- 2% OF THE END EFFECTOR MASS.