Real time human motion compensation based on joint DoF constraints for upper limb soft tissue artefacts

To correct the human motion tracking errors caused by the soft tissue artefacts (STA) in humerus axial rotation and forearm pronation/supination (PR-SU) rotation, a compensation method based on joint degree of freedom (DoF) constraints is presented. A seven DoFs upper limb model with four segments is used to describe the kinematic information which is obtained by retro-reflective skin-mounted marker clusters. The relative attitude angles of the connected limbs are calculated by matrix decomposition. The angles which are not satisfied joint DoF constraints of elbow and wrist are used to compensate the motion of humerus axial rotation and forearm PR-SU rotation, respectively. The feasibility of the proposed method was tested in vivo in terms of different flexion - extension angle of elbow and wrist joints. The results demonstrate that the method can effectively reduce the motion tracking error caused by STA for large range of motion, especially when the limb's rotational angle closes to the limit.