Research On Voice Control System Of Biped Robot Based On Android

With the rapid popularization of intelligent terminals and the rapid development of mobile Internet,the goal of connecting the whole world through a mobile phone seems to have reached.After entering the 21 st century,artificial intelligence is a unavoidable topic in the field of science and technology.Although the researches or applications of artificial intelligence are still in the primary stage,the recognition accuracy of speech recognition technology,which is the most successful technology of artificial intelligence landing,is very high.Therefore,it is very important to design and develop the mobile terminal voice control system which meets people's needs.Since there are many inconveniences in the control of robots based on PC,smart phones are selected as control terminals in this paper.The Android platform was chosen because it is a Linux-based open source operating system with a high market share and portability.In this paper,the HBE-Robonova-AI intelligent biped robot is taken as the controlled object,and the application software of the robot voice control system based on Android smart phone is designed and developed.Three different communication modes are used to control the robot from near to far,which improves the convenience of manipulating.The main work and research results of this paper are as follows:1.The overall framework and hierarchical structure of the speech control system based on the whole system are designed and The software and hardware requirements of the system are analyzed..The idea of modular programming was determined,and different hardware modules were extended according to different communication methods,and different communication protocols are designed.2.Two parts of the program of robot's control board and motion board are developed.For the control board,the program is written in C language and cross-compiled and debugged under Linux environment,while for the motion board,the program is written in Robo Basic language under Windows environment and co-debugged with the control board program.3.After analyzing the forming process of the speech signal,the mathematical modeling and characteristic analysis of the speech signal are carried out.The specific process of speech recognition was clarified,and the differences of different recognition modes were compared.The simulation of speech recognition system based on DTW algorithm is designed and realized.Finally,the acoustic model of speech recognition is determined.4.In the environment of Android Studio,the speech recognition program,the Bluetooth/Wi Fi mode information transmission program and the GPRS short message mode information transmission program are developed with the Java language.The program test of each software and hardware module is completed,and the speech recognition program is combined with three communication modes to test.After the successful test,the four parts are combined into a complete voice control App and the compatibility test of different devices is carried out.The problems encountered during the period are analyzed and solutions are given.After the completion of the design and testing of the control system,the full text work is summarized and the prospects for further improvement are put forward.