RESUMO
Objective:To develop an automatic adjustment algorithm of bed height of multi-position lower limb rehabilitation robot, to meet the variety of leg lengths and training modes to avoid the collision between robot and ground. Methods:Six mathematical models of robot bed body height were established for six training modes of multi-position lower limb rehabilitation robot, which were described with leg length and bed tilt angle. The influence was analyzed that mechanical clearance and deflection as well as the jitter error of leg bracket during movement. Furthermore, a software related to these models was developed to automatically adjust the bed height for training. Volunteers were recruited to test actually. Results:The test data of bed height are consistent with the theoretical calculation of six mathematical models. Clearance and deflection did not affect the theoretical results of bed height. The end of robot's lower limb was always above the safe height during rehabilitation training. Conclusion:The automatic adjustment algorithm of bed height has been established, which can ensure that the rehabilitation robot runs at a safe height.