Exercise Fatigue Estimation And The Application In The Control Of The Lower Limb Rehabilitation Robot

The rehabilitation training for limb movement function barrier patients has become a research focus in the modern rehabilitation engineering. Traditional one-to-one treatment mode between healers and patients can not meet the real demand. With the development of the robot technology, the domestic and foreign research institutions will introduce the robot technology into the field of rehabilitation, and design a variety of rehabilitation robot to assist patients in rehabilitation training. Its emergence and application greatly improved the original one to one relationship between doctors and patients. But the existing rehabilitation robot’s control is simple, the control strategy is adopted in the training process which is easy to cause fatigue for patients, and robot cannot estimate the fatigue status of patients by itself. With the rehabilitation training mission, the fatigue level of patients will be deepened, this will extremely easy to cause secondary damage for patients, thus it will affect the rehabilitation effect. Aiming at this situation, this article introduced the biological electrical signal analysis technology and feedback control technique into the rehabilitation training system, researching dynamic feature extraction method based on ECG and s EMG, to obtain the characteristic indexes and change rule which relates to human exercise fatigue. Further, using ECG and s EMG fatigue characteristics of the feedback control of the robot running state. This paper carries out the work as follows:Firstly, the ECG and s EMG of the rehabilitation exercise is introduced, and the signal acquisition experiment is designed. The ECG and s EMG date is obtained under different exercise state, and then discussing the ECG and s EMG relevance to human exercise fatigue, and the ECG and s EMG exercise fatigue characteristics is extracted.Secondly, the fatigue estimation method is researched based on the fusion feature of the ECG and s EMG, then the hidden markov model and support vector machine identification algorithm is introduced, and then human exercise fatigue estimation model is established using the two kind algorithm. Through the actual test data, the applicability of the fatigue estimation model is determined.Thirdly, it introduces the entities for the use of lower limb rehabilitation robot, getting its geometric model, analyzing the kinematics and dynamics model by the geometrical method and lagrangian method, and corresponding kinematics and dynamics model is established. And aiming at the limitations of the existing rehabilitation facility work patterns, it proposes the passive training feedback control strategy based on fatigue characteristics and the active training feedback control strategy based on human-computer interaction.Finally, performing the rehabilitation robot motion control experiment for the research method and control strategy, respectively designing the passive training control and active training control experiment based on the HMM fatigue estimation. The effectiveness of the research methods and the control strategy is verified through the two groups of experiment.