Research Of The Wireless Monitoring And Control System For Wheelchair Robot Base On The Speech Recognition

Speech recognition is a kind of technology that is using computer to transfer the voice signal to an associated text or command by identification and understand. The speech recognition technology, the video transmission technology, the short distance control are unify with the robot technology which is becoming the present research the hot spot. The research about wheelchair robot's man-machine short distance interactive question has not only the good theory significance, but the great use value. The paper including the following several partial contents:First, has established based on the speech recognition wheelchair robot wireless monitoring system hardware platform. Uses the superior machine,lower position machine plan construction wireless speech recognition monitoring system .The system may observe wheelchair's state of motion and patient's situation through the video transmission. The operators can send out control command through the microphone to the wheelchair robot execution compulsory exercise.Next, has studied the speech recognition basic principle, and studies hidden Markov model the speech recognition algorithm, has constructed one in the MATLAB platform based on the hidden Markov model isolated word speech recognition system. This system has completed the speech recognition main process, contains the pretreatment, the vertex examination, the characteristic parameter extraction and the pattern matching. Uses Hidden Markov Model has completed Chinese pronunciation "0" to "9" training and recognition. The multi-people test the result indicated that this system's recognition rate achieves about 80%.The introduction based on the image marker recognition process, introduced the image pretreatment, the image division, the picture edge extraction, uses HOUGH to transform again obtains the marker central coordinate, calculates robot's pose, according to the TRT way plan, obtains the robot motion diagram. Control the robot to arrive at the target location. Realizes the robot based on the marker guidance.Finally, has realized on the PC platform and the monolithic integrated circuit system to the wheelchair robot's wireless speech recognition supervisory system. Under the VC++ development environment, has compiled the wheelchair robot wireless speech recognition supervisory system software routine. Uses the modular design on the VC++ platform, realized three main functions for the speech recognition, the wireless data transmission, the video monitor. The experimental result indicated that the wheelchair robot wireless speech recognition supervises control system's recognition performance and the wheelchair movement performance is good.