Understanding the mechanism of human balancing is a scientifically challenging task. In order to describe the nature of the underlying control mechanism, first of all, the balancing force has to be discovered. However the kinematics of the motion can be measured directly using a motion capturing system, the calculation of the control forces is more problematic. It seems straightforward to calculate the control forces using an inverse dynamic calculation based on the accelerations. Using

more »

... l motion capturing systems, these accelerations can be derived only by numerical differentiation, which cause large noise in the resulted time signal. Consequently, the reconstructed control force does not reproduce the actual motion. In order to overcome this problem, a new approach is proposed in this paper. First the solution of the linearized system is used, then, an optimization problem is solved to find a control force, which generates the same motion in the numerical simulation as the captured motion. A main advantage of the method is that there is no need for the numerical differentiation of the measured data for the calculation of the control forces. The method is demonstrated for human stick balancing measurements.