Research Of The Omni-Directional Mobile Rehabilitation Robots Intelligent Motion Strategy

Medical theory and clinical medicine prove that besides the early surgical and necessary drug treatment, correct and scientific rehabilitation training play a very important role to the rehabilitation and improvement of limb movement function for the patients with limb degradation or injury. With the development of society, people makes higher demands of human life and quality of life, which make rehabilitation medicine and rehabilitation robotics research have more realistic significance and broader space. In this paper, based on the whole lower extremity rehabilitation robot to move the motion characteristics, we try to design a robot movement active training mode, which is driven robot autonomous movement rehabilitation of lower limb disorders a sport mode.First, if rehabilitation robot can move itself, the perception of its external environment is essential. The information collection and processing system are made of single-chip and ultrasonic sensors .According to the characteristics of the robot movement, gathering information about the external obstacles environment, using BP neural network technology, for pattern recognition and classification of the surrounding environment. Software BP neural network classifier for the generation process and the results of simulation using MATLAB. Through the improved BP algorithm, the application makes the LM algorithm of BP neural network classifier results meet the design requirements. With the statistical results of many simulation test data, show that the classifier can be good model for the identification and classification of obstacles, which can be a good identification of the surrounding environment.Secondly, in this paper, though the single-limb movement disorders of gait analysis and the law of the distance between the robot and patient values, which designs a mobile robot gait detection technology. In other words, we use LM algorithm of BP neural network to generate gait pattern recognition device through the MATLAB simulation. This classifier can be trained in the rehabilitation of patients with a good gait for identification.Then, based on fuzzy control theory we design a whole to move lower limb rehabilitation exercise machine intelligence methods. According to the research results and fuzzy control robot obstacle avoidance algorithm applications, we construct a fusion of all the mobile features and patient information robot obstacle avoidance gait fuzzy controller. MATLAB simulation results show the robot in the process of self-motion obstacle avoidance can be very good. Also we take into account the patient's different gait conditions, and produce a Omni-directional mobile rehabilitation robots intelligent motion strategy.