Android Based Wheel Chair Operation using PIC Controller

Android based Wheel Chair operation has been developed to control a mobile robot/wheel chair using android based Commands. The system is divided into two main unit’s Android application unit and robotic/wheel chair control unit. In the Android application, hidden Markov models are used as main method in recognizing android based commands. The robotic control unit is developed under the ARIA environment, which is provide by the robot/wheelchair manufacturer. We are taking the Android data from the microphone using a Bluetooth module. This data is stored in an array. This array is passed on to a function which extracts word from the array (i.e. Transferred data are extracted & quiet periods are dumped). These words are sent to a function which extracts frequency as a function of time. This frequency vector of the Transferred data. This vector is compared with reference vectors.one of the reference vectors would match (i.e. the inner product in this case greater than the other T The command corresponding to this reference vector is fed to the microcontroller. In this project we are getting a user dependent isolated data recognition system with a recognition accuracy of about 85% using the accuracy can be improved further & the system can be used for more number of data if during the training of the system the Bluetooth data are improved.