CN104076747A - Robot control system based on Arduino control board and voice recognition module - Google Patents

Abstract

The invention belongs to the field of two subjects of voice recognition and robotics. The aim of the invention is to provide a comprehensive control system in which a control common of a human being is recognized via a voice recognition module, the voice recognition module is then connected with a mechanical arm of a robot, and therefore the robot is controlled to complete various planned actions through voice control and mutual communication between the human being and the robot can be realized. The control system is composed of an ASR-M08A module, Arduino MEGA ATmega1280 and Arduino MEGA1280 expansion board modules, a sensor module, a serial TTL 4-way relay control board module, a Bluetooth module, a 32-way servo actuator controller module, an action memory card module and a robot mechanical arm module, wherein the Arduino control board and the sensor module realize collection of sensing data; the relay module mainly realizes functions of starting and stopping of robot mechanical arm actions; the Bluetooth module realizes wireless control; the 32-way servo actuator controller module and the action memory card module realize way planning and storage of the robot actions; and the robot mechanical arm module is the hardware condition for completing the actions and has five degrees of freedom.

Description

Robot control system based on Arduino control panel and sound identification module

Technical field

The present invention mainly belongs to speech recognition and these two ambits of robotics, object is by sound identification module identification people's control command, again sound identification module is connected with the mechanical arm of robot, thereby realize voice control, completes various planning actions and man-machine mutual interchange.

Background technology

The research of robotics relates to numerous subjects such as machine vision, artificial intelligence, digital signal processing, communication theory, electronic technology, computer science, pattern-recognition widely.The development of Robotics will help the mankind to free from heavy manual labor to a great extent.On the one hand service robot by replace more widely people be engaged in various machinery repeatedly, the health that is harmful to health and dangerous production operation.Household service robot will more and more enter into average family and help people on the other hand, look after old man, and complete various tasks.

Enter 21 century robot automtion level and be greatly improved, these intelligent robots, not only apply various feedback transducers, but also use various study in artificial intelligence, reasoning and decision-making technic.Intelligent robot is also applied many up-to-date intellectual technologies, as telepresenc technology, virtual reality technology, multi-agent Technology, artificial neural network technology, genetic algorithm and genetic programming, bionics techniques, multisensor is integrated and integration technology and nanometer technology etc.

The related field of speech recognition technology comprises: signal processing, pattern-recognition, theory of probability and information theory, sound generating mechanism and hearing mechanism, artificial intelligence etc.Speech recognition process mainly comprises pre-service, characteristic parameter extraction and model training etc.Voice signal first after filtering, sampling and Quantifying obtains discrete digital signal, is exactly then pre-emphasis, and the object of pre-emphasis is filtering low-frequency disturbance; Voice signal is a kind of typical non-stationary signal, has time varying characteristic, so divide frame operation by voice signal, after minute frame, can do to each frame the processing of a windowing; The object of end-point detection is the segment signal from comprising voice, to determine starting point and the terminal of voice, and finds out accurately the starting point and ending point of voice segments, and just likely making the data that collect is the voice signals that really will analyze.The method of speech recognition technology on phonic signal character parameter extraction has, linear predictor coefficient (LPC), linear prediction cepstrum coefficient coefficient (LPCC) and Mel frequency cepstrum coefficient (MFCC); The method of using on model training has dynamic time warping (DTW), Hidden Markov Model (HMM) (HMM), and using is at present Hidden Markov Model (HMM) more widely.

It is exactly to allow machine, by identifying, the mankind's voice signal is changed into corresponding text or order that robotics and speech recognition combine, its final purpose is exactly as interpersonal talk exchange of information, realize man-machine conversation, namely give machine with the sense of hearing, make machine can understand people's language, distinguish content and the speaker of speech, further make machine operate according to people's will, the mankind are freed from heavy or dangerous work.

Summary of the invention

The object of system of the present invention has been to provide a kind of practical complex control system that realizes the predetermined planning action of robot based on sound identification module.

Robot control system based on Arduino control panel and sound identification module is comprised of ASR-M08A module, Arduino MEGA ATmega1280 and Arduino MEGA1280 expansion board module, sensor assembly, serial ports TTL4 road Control plate module, WE-40C bluetooth module, 32 road Servo-controller controller modules, action memory card module, robot arm module.Wherein said ASR-M08A module mainly realizes transmission control command and completes predetermined planning action and good in interactive function to robot arm; Arduino MEGA ATmega1280, Arduino MEGA1280 expansion board module and sensor assembly mainly complete collection and the Presentation Function of robot sensing's data; Relay module mainly realizes beginning and the hold function of robot arm action; WE-40C bluetooth module is connected between ASR-M08A module and serial ports TTL4 road Control plate module, realizes speech recognition Radio Control Function; 32 road Servo-controller controller modules and action memory card module are realized path planning and the storage of robot motion.Robot arm module is the hardware condition of execution, RB-421 Servo-controller, consists of, and mechanical arm degree of freedom is five.

Described ASR-M08A module is connected with serial ports TTL4 road Control plate module, described Control plate module Yu32 road, serial ports TTL4 road Servo-controller controller module connects, described bluetooth module is connected between ASR-M08A module and serial ports TTL4 road Control plate module, described 32 road Servo-controller controller module is connected with action memory card module, described Arduino MEGA ATmega1280 and Arduino MEGA1280 expansion board module Yu32 road Servo-controller controller module connect, described sensor assembly is connected with Arduino MEGA ATmega1280 and Arduino MEGA1280 expansion board module, described robot arm module Yu32 road Servo-controller controller module connects.

Control system can complete the planning of various predetermined action as required, and the content of sound bank in voice module.

Accompanying drawing explanation

Fig. 1 is system hardware module connection diagram of the present invention

Embodiment

In described ASR-M08A sound identification module, first by news, fly the sound bank that speech synthesis system is set up sound identification module, the size of sound bank and content can be set as required.The recognition principle of ASR-M08A module is by predefined key word and the Speech comparison of inputting from microphone, exports the speech recognition content that we want.

ASR-M08A module is connected with serial ports TTL4 road Control plate module, in the Control plate module of serial ports TTL4 road, embedded algorithm, make the output of ASR-M08A module and the input of relay module corresponding, thereby can control by voice the break-make of serial ports TTL4 road Control plate module.In ASR-M08A sound identification module, containing three kinds of speech detection patterns, is respectively cycle detection pattern, password detection pattern, button detecting pattern.Under cycle detection pattern, module can not stop to detect ambient sound, so right you are you any time, instruction all can be detected and identify, and shortcoming is that this pattern exists larger false recognition rate; Under password detection pattern, be to control an action by two instructions, only passing through in article one instruction situation, and say within a certain period of time second instruction could correctly be identified; Under button mode, often press button and tell instruction in 8 seconds and could identify, 8 seconds later again say that instruction, this pattern are exactly can false triggering, but underaction.

In patent of the present invention, sound identification module adopts password detection pattern.Serial ports TTL4 road Control plate module repeat circuit is output as the connection terminal of three, with 1,2,3, replaces respectively, and wherein the 1, the 2nd, normally closed contact, leads to when relay is cold, after energising, disconnects; And the 2, the 3rd, normally opened contact, disconnects when relay is cold, after energising, is closed.Control plate module Yu32 road, serial ports TTL4 road Servo-controller controller module connects, so while connecting serial ports TTL4 road relay module between 32 road Servo-controller control panel feeder ear and power end, the power supply side of Jiang32 road Servo-controller control panel is connected on 2,3 normally opened contacts.In order to realize wireless speech, control, ASR-M08A module and serial ports TTL4 road Control plate module install respectively bluetooth module additional simultaneously.Wherein ASR-M08A module 3.3V power lead is connected with bluetooth module 3.3 interfaces, ASR-M08A module RX end is connected with the TX end of WE-40C bluetooth module, ASR-M08A module TX end is connected with the RX end of WE-40C bluetooth module, and ASR-M08A module GND end is connected with the GND end of WE-40C bluetooth module; And serial ports TTL4 road Control plate module is when be connected with WE-40C bluetooth module, serial ports TTL4 road relay module 3.3V interface is connected with the 3.3V of WE-40C bluetooth module, relay module RX end in serial ports TTL4 road is connected with WE-40C bluetooth module TX end, relay module TX end in serial ports TTL4 road is connected with WE-40C bluetooth module RX end, and relay module GND end in serial ports TTL4 road is connected with WE-40C bluetooth module GND end.

32 road Servo-controller controllers are connected with robot arm module, 32 road Servo-controller controllers are by the motion path of upper computer software planning robot mechanical arm, and trajectory path planning is to realize by controlling respectively the anglec of rotation of each RB-421 Servo-controller.In the present invention, the motion path of planning is to salute, put down.Then deposit the code of planning motion path in action memory card module.Action memory card module Shi Yu 32 road Servo-controller controller modules connect.

Arduino MEGA1280 expansion board module is the expansion of Arduino MEGA ATmega1280 external interface, can directly Arduino MEGA1280 expansion board module be inserted on Arduino MEGA ATmega1280 by draw-in groove, enrich the IO interface of Arduino MEGA ATmega1280.In order to make the functional diversities of robot, in patent of the present invention, return robot and increased sensor assembly, sensor assembly is connected in Arduino MEGA1280 expansion board, and the sensor containing in sensor assembly has DHT11 Temperature Humidity Sensor, flame sensor, hummer, RB URF v1.1 ultrasonic sensor etc.

Use Arduino MEGA ATmega1280, Arduino MEGA1280 expansion board and flame sensor, hummer module can build Yi Ge robot fire protection alarm system; Flame sensor is connected with analog interface in Arduino MEGA1280 expansion board, hummer is connected with digital interface, then define the analog output value of flame sensor as the input value of hummer, when the analogue value that reads input when hummer is greater than certain parameter, hummer rings, and sends alarm of fire.

Use Arduino MEGA ATmega1280, Arduino MEGA1280 expansion board and DHT11 can obtain robot temperature and humidity sensing data around;

Use Arduino MEGA ATmega1280, Arduino MEGA1280 expansion board, keep away barrier sensor and ultrasonic sensor can be realized robot range finding and keep away barrier.First RB URF v1.1 ultrasonic sensor is connected to the ultrasound wave interface in Arduino MEGA1280 expansion board, its correspondence transmits and receives mouth and is respectively numeral 14 and digital 15 interfaces, therefore first we define the receiving interface that numeral 14 and digital 15 interfaces are ultrasonic sensor and send interface, then ultrasonic sensor sends pulse signal, read the burst length, to the burst length be converted into distance, we define a distance parameter as the critical value of robot obstacle-avoiding simultaneously, when the distance of burst length conversion is less than critical value, robot obstacle-avoiding turns to, when the distance of burst length conversion is greater than critical value, robot need not keep away barrier and turn to.

Claims (1)

1. the robot control system based on Arduino control panel and sound identification module, is characterized in that comprising the steps:

Step (1): ASR-M08A module, by predefined key word and the Speech comparison of inputting from microphone, is exported the speech recognition content that we want;

Step (2) ASR-M08A module is connected by bluetooth module with serial ports TTL4 road Control plate module, in the Control plate module of serial ports TTL4 road, embedded algorithm, make the output of ASR-M08A module and the input of relay module corresponding, thereby can control by voice the break-make of serial ports TTL4 road Control plate module;

Step (3): Control plate module Yu32 road, serial ports TTL4 road Servo-controller controller module connects, so while connecting serial ports TTL4 road relay module between 32 road Servo-controller control panel feeder ear and power end, the power supply side of Jiang32 road Servo-controller control panel is connected on normally opened contact;

Step (4): 32 road Servo-controller controllers are connected with robot arm module, 32 road Servo-controller controllers are by the motion path of upper computer software planning robot mechanical arm, and trajectory path planning is to realize by controlling respectively the anglec of rotation of each RB-421 Servo-controller;

Step (5): deposit the code of planning motion path in action memory card module.Action memory card module Shi Yu 32 road Servo-controller controller modules connect;

Step (6): Arduino control panel module is connected with sensor assembly, enriches the collection of robot to the sensing data of environment.