Abstract: A hybrid exoskeleton of knee joint and a motion planning algorithm based on Lie algebra are proposed to solve the problem of “parasitic force” between human and machine.The mapping between the driver and the human motion is established by using the kinematic model，the human motion is sensed，and the joint configuration space is constructed. A parameter selection method based on screw theory and particle swarm optimization algorithm is introduced to optimize the size of knee exoskeleton with the objective of workspace area and global transmission index.The simulated and experimental results show that the workspace of knee exoskeleton covers the motion range of knee joint，and the motion transfer efficiency of limb is greater than 0.7；and the exoskeleton can accurately adjust its own motion axis according to the human motion and eliminate the “parasitic force” between human and machine.

Key words: kneejoint, lower-limbexoskeleton, axis, self-adaptivedesign, dimensionalsynthesis, screwtheory, parameterselection