KR20050110780A - Voice recognition control ship and voice recogintion rudder control system - Google Patents

Abstract

The present invention provides a voice recognition control ship and a voice recognition steering control system.

The voice recognition control ship of the present invention comprises an automatic steering, a system operation and a hull rotating device, a hydraulic machine system and a voice recognition steering control system, the voice recognition steering control system includes a voice input means and a voice recognition means The steering unit includes a recognition unit, a microcontroller, and a driving relay, and calculates a difference between the steering angle recognized by the voice steering command and the current steering angle and controls the hydraulic valve of the steering system until there is no difference. It works by rotating it.

The voice recognition control ship and the voice recognition steering control system of the present invention can be remotely controlled in a wired or wireless manner, thereby performing other tasks at the same time, thereby improving work efficiency.

Description

VOICE RECOGNITION CONTROL SHIP AND VOICE RECOGINTION RUDDER CONTROL SYSTEM}

The present invention relates to a voice recognition control ship and a voice recognition steering control system. More specifically, the present invention relates to a voice recognition control ship and a voice recognition steering control system capable of remotely controlling a steering system in a ship by voice.

Recently, in order to reduce the labor force on the ship and to improve the safety of operating shipbuilding, the automation system of small vessels and fishing vessels is required. In order to improve productivity and efficient operation of small vessels and fishing vessels, rapid technological advances have been made. However, the operation of ships has become more complicated, and there has been a need to operate various instruments and ship equipment.

In order to solve this problem, a method of using voice technology in a ship steering system is being considered. By the way, although the voice recognition technology is already in the practical use stage, it is impossible to find an example of the practical use of the voice technology in a high-noise environment such as a small ship and a fishing boat or in an environment in which an operator performs several tasks at the same time.

Therefore, the development of a system that can remotely control the steering system of small vessels and fishing vessels using voice recognition technology is required.

The present invention is to provide a voice recognition control ship and a voice recognition steering control system used therein for controlling a ship by voice recognition of an implied instruction command containing the intention of a ship operator.

In addition, the present invention is to provide a system for controlling a ship steering remotely by understanding the intention and instructions of the ship operator by converting the recognition result by the voice command to text.

The present invention relates to a voice recognition control ship and a voice recognition steering control system. More specifically, the present invention relates to a voice recognition control ship and a voice recognition steering control system capable of remotely controlling a steering system in a ship by voice.

The voice recognition control ship of the present invention includes an automatic steering, a system operation and a hull rotating device, a hydraulic machine system and a voice recognition steering control system.

More specifically, in the voice recognition control ship of the present invention, the automatic steering device includes a direction sensor, a rudder reference unit, a control unit, and a processor unit, and operates a system and a hull rotating device. It includes a system control board, a hydraulic pump starter, a selector switch, a manual steering device, a steering angle indicator and a motor driver, and the hydraulic machine system includes a hydraulic tank, a hydraulic pump, a solenoid valve hydraulic cylinder and a rudder.

The basic conceptual view of the ship of the present invention is as shown in FIG.

The speech recognition steering control system of the present invention includes a speech recognition unit for speech recognition and a steering controller for processing the recognition result.

More specifically, the voice recognition steering control system of the present invention, which is used in a voice recognition control vessel, includes a voice recognition unit including a voice input means and a voice recognition means, and a steering controller including a microcontroller and a driving relay.

The voice recognition remote steering control system of the present invention can be understood through the dialogue with the system of the navigation instructions of the ship operator necessary for the operation of the ship through artificial intelligence.

The voice recognition steering control system of the present invention can be remotely controlled in a wired or wireless manner, and accordingly it is possible to perform different tasks at the same time can improve the efficiency of the operation.

The voice recognition steering control system of the present invention includes a database unit for databaseing the steering-related commands mainly used on the ship, and includes a voice or wired / wireless communication module of a grouper type, and includes a wireless transmission control system. It includes an unmanned steering control system and a voice instruction based learning (VIBL) system based on ride manipulation know-how.

An operation example of a ship using the voice recognition steering control system of the present invention is as follows.

When the user issues a specific steering command through the voice recognition unit, the voice recognition unit recognizes the voice, interprets the command from the processor, and sets the appropriate rudder angle to the ship's solenoid valve block through the terminal terminal. It is controlled to move the ship.

In this case, the current bearing and the steering angle are detected by the heading sensor and the rudder reference unit of the model ship and are provided from the autopilot. The processor receives data from the autopilot and outputs the result to an external display device (for example, a liquid crystal display).

The voice recognition unit of the present invention includes a voice input unit including a group speaker type voice or wired / wireless communication module, a database unit of a steering command, and an intelligent voice recognition algorithm unit capable of analyzing a voice instruction type according to the marine environment and the ship operating state. Speech recognition means.

Speech recognition unit of the present invention adopts a speaker independent method, the main characteristics are as follows.

ㆍ Low power, low cost high performance DSP based

ㆍ Independent speaker and variable vocabulary recognition engine

ㆍ Consist of DSP, CODEC, Data ROM

ㆍ Customizable by taking advantage of programmable DSP

ㆍ The number of recognized words varies depending on the length of the recognized command.

ㆍ Expansion of recognition instruction by adding memory

ㆍ Deny function support

ㆍ Microcontroller and easy interface support

ㆍ Recognition time: around 400msec

ㆍ Recognition rate: around 98%

Hereinafter, the algorithm of the speech recognition unit of the present invention will be described.

The following description is only one specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto.

In this embodiment, the voice recognition module VR-33R is adopted as the voice recognition unit.

First, the data format used in the speech recognition unit of the present invention will be described.

In the present embodiment, the output data from the voice recognition unit to the steering gear control unit uses 8-bit data. Accordingly, the recognition result is output as a value of 1 to 255. When the recognition is rejected by pronouncing a word that is not a recognition target, 0 is output.

Input data from the control unit to the voice recognition unit is also 8-bit data, the format of the input data is shown in the following table.

BIT NAME FUNCTION 2-0 REJECT Set the rejection rate. It can be set from 0 to 7, and the higher it is, the higher the probability of rejection if the pronunciation is not recognized. If it is set to 0, it is not rejected at the time of recognition. 5-3 VOLUME Set the microphone input volume. You can set from 0 to 7. 6 LED Set Blink mode of LED 0: LED gives 0.25 seconds when recognizing Toggle1: If no voice is input at the time of recognition, it turns off and LED is on while voice is input. 7 PD Set standby mode 0: Set to Stand By mode 1: Set to Power Down mode

 The rejection rate can be set between 0 and 7, but the default value is set to 4.

Next, transmission of the input data to the voice recognition unit will be described.

Change HST2DSP (Host to DSP Interrupt) from HIGH to LOW (Keep more than 5μsec), and then HIGH (Keep more than 5μsec), and then use 8-bit input data using SCL (Serial Clock) and SDA (Serial Data). To the voice recognition unit.

After outputting 1 bit to SDA, change SCL from HIGH to LOW (more than 5 μsec) and HIGH again (more than 5 μsec). Repeat this 8 times to transmit 8bit input data (MSB First).

Subsequently, the transmission of the output data in the control section will be described.

SDA and SCL are used to read the output data of the recognition result. Switch the I / O pin of the microcomputer of the controller connected to SDA to input. Change SCL from HIGH to LOW (more than 5μsec) and back to HIGH (more than 5μsec) and read SDA to get 1 bit.

SDA remains in effect until SCL is granted again. Repeat this eight times and read the 8-bit result (LSB First). After reading all 8th bit, convert the I / O Pin of the microcomputer connected to the STATUS Pin to the output and apply LOW over 5μsec, then switch the I / O Pin of the microcomputer connected to the STATUS Pin to input again.

Hereinafter, programming of the speech recognition unit will be described.

Set HST2DSP and SDA to HIGH, set the I / O pin of the microcomputer connected to STATUS as input, and apply SLOW to RESET for more than 1μsec while setting SCL to LOW, then set HIGH to VR-33R. Apply Reset.

After applying Reset to VR-33R, check if the status is LOW (if not, wait until LOW) and set SCL to HIGH.

Set the SDA to HIGH (hold at least 5μsec) after maintaining SLOW at least 5μsec.

To complete the initialization of the module, one of two standby modes must be set. In power down mode, the input data is set to binary 10000000b, and in stand-by mode, the input data is set to binary 00vvv000b. Where vvv is the 3-bit microphone input volume setting. Transmit input data to VR-33R, and after transmission is completed, keep SDA LOW for over 5μsec and set HIGH (over 5μsec). When this process is performed, the VR-33R will enter standby mode.

In order to start the recognition, after confirming that the STATUS is HIGH (if it is not HIGH, wait until HIGH), 8-bit input data is transferred to the VR-33R, and when the transfer is complete, the VR-33R waits for voice input (LED bit 0). When set to, LED OUT sends out Toggle output of LOW and HIGH at intervals of about 0.25sec while waiting for voice). While waiting for recognition, if you change HST2DSP from HIGH to LOW and again HIGH, and then set SDA to LOW for 5μsec or more and set HIGH (to keep 5μsec or more), the VR-33R will stop recognition and go back to standby mode. Enter At this time, the standby mode is the standby mode set in the previous input data.

The microcomputer of the controller waits for the recognition result from the VR-33R. When recognition is complete, the STATUS changes from HIGH to LOW. Microcomputer should use this signal as an interrupt source or poll it.

The microcomputer reads 8bit output data after confirming that the status changes from HIGH to LOW. The recognized result value corresponds to the steering command table.

The VR-33R will automatically enter standby mode after outputting the result. At this time, the standby mode is the standby mode set in the previous input data. You can repeat the above process to resume recognition.

The conceptual diagram of this programming process is shown in FIG.

The steering control unit of the present invention includes a microcontroller and a drive relay.

The microcontroller used in the present invention is not particularly limited as it is commercially used, but the present invention uses a PB-3B model of PICBASIC, which is an ultra-small single board computer moving in the BASIC language as one embodiment. The hardware configuration of PICBASIC consists of PIC microcomputer and EEPROM.

In general, commercially used steering systems use hydraulic valves driven at 24 V or higher, so a relay must be driven based on the PICBASIC output signal (5 V) to control the hydraulic valve.

However, since the PB-3B handles a small current of 5V, there is a problem in that the large current of 24V or more used for operating the electronic hydraulic valve of the steering system cannot be turned ON / OFF. This can be solved by connecting using a drive relay.

Here's how it works. When the LED in the photocoupler is turned on with the current output from the I / O port of the PB-3B, the relay is driven to control the hydraulic valve of the steering system. Since the photocoupler transmits the signal state with light, it is possible to electrically separate the 5V power supply and the relay side of the PB-3B side, so that no noise is transmitted. Here, the D1 diode attached to the relay side prevents the counter current generated when the coil is turned on and off.

The program of the steering control unit of the present invention is as follows.

When the recognition of the steering command is completed in the voice recognition unit, the ship's steering system is controlled as shown in FIG. 3. This will be described in detail as follows. When the power is supplied to the steering control board, the board goes through the initialization process and waits for the switch input (SW1). When SW1 is pressed, serial input starts, and the input value is the signal output from the autopilot. Only the current azimuth and steering angle are extracted from the input signals and displayed on the display device.

After that, press the switch SW2 or SW3 and wait for the voice to be input.When the steering command is input, the error of the set steering angle and the current steering angle recognized by the voice recognition unit is calculated to control the hydraulic valve of the steering system until the error becomes zero. .

Hereinafter, the configuration of the voice recognition control ship of the present invention will be described.

The configuration of the voice recognition control vessel of the present invention is a heading sensor, a hydraulic tank, a hydraulic pump, a solenoid valve, a hydraulic cylinder, and a rudder reference device. Reference Unit).

The configuration of the voice recognition control vessel of the present invention is roughly divided into an autopilot, a system operation and a hull rotating device, and a hydraulic machine system.

The automatic steering of the ship of the present invention includes a direction sensor, a rudder reference unit, a control unit, and a processor unit.

In the present invention, the orientation sensor refers to a device that transmits the current azimuth angle of the ship to the processor unit, and the rudder reference device refers to a device that transmits a ruder angle to the processor device. On the other hand, the control device includes a display unit of the current azimuth, steering angle, etc., and is an operation unit of the autopilot with the function to set the azimuth value, etc. when sailing the autopilot, the processor device is input from the orientation sensor, leather reference device, control device Each value is calculated and the final output signal is transmitted to the solenoid valve block.

The system operation and hull rotating device of the ship of the present invention includes a system control board, a hydraulic pump starter, a selector switch, a manual steering device, a steering angle indicator and a motor driver.

Hereinafter, each configuration thereof will be described in detail.

(1) System Control PCB

It receives the rudder angle and azimuth data from the control unit of the autopilot and delivers the data to the voice recognition system. The solenoid valve open / close signal actually transmits the voltage signal transmitted to the valve to the voice recognition control board.

The rotation direction is determined by the open collector signal to the DC driver (for driving the DC motor for hull rotation) according to the angle data, and the hull rotation speed can be adjusted by the DC driver with the output of DC 0 ~ 5V according to the angle data. The rotation speed is displayed.

    Rudder Angle Indicator gives ± 10V output to indicate the steering angle data.

(2) hydraulic pump starter

The hydraulic pump starter starts and stops the hydraulic pump and displays the current status.

(3) Selection switch

There are three choices for the steering operating position: manual, automatic steering and computer.

(4) Manuel Steering Unit

Electrical signals are provided to the solenoid valve block for other operations.

(5) Rudder Angle Indicator

It receives the output of ± 10V from the control board and displays the other current value.

(6) DC Motor Driver

It receives the direction of rotation and speed signal from the control board and drives the DC motor (Hull Rotator) to enable the hull rotation.

The hydraulic machine system of the ship of the present invention is composed of a hydraulic tank, a hydraulic pump, a solenoid valve hydraulic cylinder and a rudder.

The solenoid valves include hydraulic and flow control valves that operate the cylinders in response to electrical signals from automatic steering, manual steering, and computers to operate the hydraulic cylinders, and the rudder is hydraulically operated at the solenoid valves. The cylinder receives hydraulic pressure to induce rotation of the hull.

The speech recognition control ship of the invention preferably comprises the speech recognition steering control system of the invention of the invention.

Referring to the operation of the voice recognition control ship of the present invention as follows.

When the user using the ship system of the present invention instructs the steering command to the microphone connected to the autopilot, the voice recognition unit recognizes the steering command and recognizes the difference between the steering angle recognized as the steering command and the current steering angle obtained from the automatic steering machine. The ship is then rotated by controlling the hydraulic valves of the steering system until the difference is eliminated. After that, the result is output to the display device, and when control is completed, the process returns to the waiting state for waiting for the next command.

The steering commands used in this research and development were constructed with 13 steering commands as shown in Table 1 and four position commands as shown in Table 2 as a database and mounted on the voice recognition unit. However, these steering commands of Tables 1 and 2 are just examples, and it is natural that various steering commands can be added or changed according to each situation.

Steering command Explanation Midships ★ center Port five Port 5 degrees ★ Port ten Port 10 degrees ★ Port fifteen Port 15 degrees ★ Port twenty Port 20 degrees ★ Port twenty five Port 25 degrees ★ Hard a port Port port hit ★ Starboard five Starboard 5 degrees ★ Starboard ten Starboard 10 degrees ★ Starboard fifteen Starboard 15 degrees ★ Starboard twenty Starboard 20 degrees ★ Starboard twenty five Starboard 25 degrees ★ Hard a starboard Starboard strike ★

Position Command Explanation  Heave in  ★ roll up  Slack  ★ loose  Hold on  ★ Keep  Stop  Stop ★

Specific operation of the voice recognition steering control system of the present invention is as follows.

When the voice recognition steering control system is driven, the bearing and steering angle currently transmitted by the autopilot are displayed on the display screen. When the user presses the input switch, it waits for the voice steering command. When the user instructs the steering command by voice to the microphone attached to the autopilot, the system recognizes the steering command from the database and compares the recognized command with the current direction and steering angle to control the relay appropriately. The ON / OFF of the relay controls the hydraulic valve of the ship to move the rudder.

The recognized steering command controls the steering system of the model ship and waits for the steering command of the user once the control is completed.

In the case of using the voice recognition control ship and the voice recognition steering control system of the present invention, the ship can be remotely controlled in a wired or wireless manner, and accordingly, other tasks can be performed at the same time, thereby improving work efficiency.

Therefore, the voice recognition control ship and the voice recognition steering control system of the present invention has the effect of reducing the labor force for ship control, it is possible to improve productivity and efficient operation in small vessels and fishing vessels.

1 is a conceptual diagram of a voice recognition control ship system.

2 is a programming diagram of a speech recognition unit of the speech recognition steering control system.

3 is a steering control programming diagram of a speech recognition steering control system.

Claims (4)

A voice recognition unit including a voice input unit and a voice recognition unit, and a steering unit control unit including a microcontroller and a driving relay, wherein the difference between the recognized steering angle and the current steering angle recognized by the voice steering command is eliminated. Voice recognition steering control system to rotate the ship by controlling the hydraulic valve of the steering system until. The voice input device of claim 1, wherein the voice input means includes a voice of a group speaker type or a wired / wireless communication module, and the voice recognition means is capable of analyzing a voice command form according to a database part of a steering command and a maritime environment and a ship operating state. Speech recognition steering control system comprising a recognition algorithm unit. A voice recognition control ship comprising an automatic steering, a system operation and a hull rotating device, a hydraulic machine system and a voice recognition steering control system according to claim 1 or 2. The system of claim 3, wherein the autopilot includes an orientation sensor, a rudder reference unit, a control unit, and a processor unit. It includes a hydraulic pump starter, a selector switch, a manual steering device, a steering angle indicator and a motor driver, and the hydraulic machine system includes a hydraulic tank, a hydraulic pump, a solenoid valve hydraulic cylinder and a rudder.