CN210582880U

CN210582880U - Novel interactive wheelchair of embedded AI intelligence

Info

Publication number: CN210582880U
Application number: CN201920707229.7U
Authority: CN
Inventors: 邓晓燕; 韦政松; 黄海真; 陈洲楠; 董雷震; 黄敏; 姚望洁; 蔡典仑
Original assignee: South China University of Technology SCUT
Current assignee: South China University of Technology SCUT
Priority date: 2019-05-17
Filing date: 2019-05-17
Publication date: 2020-05-22
Anticipated expiration: 2029-05-17

Abstract

The utility model discloses a novel mutual wheelchair of embedded AI intelligence, this wheelchair include embedded main control board, image acquisition module, sound collection module, display screen and ultrasonic module. The system comprises an image acquisition module, a voice acquisition module and a display module, wherein the image acquisition module is used for acquiring a face image signal, and the voice acquisition module is used for acquiring a voice signal; the embedded main control panel is used for processing the acquired signals and outputting image sound signals and control signals; the ultrasonic wave module comprises a distance measuring unit, a resolving unit and an obstacle avoidance unit which are sequentially connected through circuits. Each device takes an embedded main control board as a core and adopts a star network structure mode to realize mutual connection and communication. The stability of each part connection communication is guaranteed while the anti-interference capability is improved.

Description

Novel interactive wheelchair of embedded AI intelligence

Technical Field

The utility model relates to an artificial intelligence and embedded equipment technical field, concretely relates to novel interactive wheelchair of embedded AI intelligence.

Background

With the continuous development of artificial intelligence, people put higher demands on the intelligence of products. Meanwhile, with the continuous development of the technology of the internet of things and the embedded device, the embedded device is more and more popular in development due to the characteristics of small size, comprehensive functions, low transplanting cost, stable performance and easiness in programming. The utility model provides a novel mutual wheelchair equipment of embedded AI intelligence aims at further satisfying people to the user demand of intelligent equipment through improving the man-machine interaction quality.

SUMMERY OF THE UTILITY MODEL

The utility model aims at solving the defects in the prior art and providing a novel embedded AI intelligent interactive wheelchair.

The purpose of the utility model can be achieved by adopting the following technical scheme:

the utility model provides a novel mutual wheelchair of embedded AI intelligence, this mutual wheelchair of intelligence includes wheelchair, embedded main control board, image acquisition module, sound acquisition module, display screen and ultrasonic module.

The embedded main control board is arranged under the base of the wheelchair; the shortest wiring distance can be obtained, and the transmission time delay and the anti-electromagnetic interference of signals are reduced.

The image acquisition module is arranged on an armrest at one side of the wheelchair, is connected and supported by a long support with better plasticity, and is in circuit connection with the embedded main control board through a USB connecting wire; the plastic long support can fully utilize the advantage that the human-machine position of the wheelchair device is relatively fixed, and the development and use cost is reduced.

The sound acquisition module is embedded in the back of the wheelchair seat and is communicated with the embedded main control board through a USB connecting wire and a GPIO (general purpose input/output) interface; the voice can be acquired most clearly, errors can be reduced through double-channel parallel communication, and the accuracy of recognition is guaranteed.

The three ultrasonic modules are respectively arranged on the front side, the left side and the right side of the wheelchair and are respectively in circuit connection with a general GPIO serial port of the embedded main control board through a DuPont wire; the disposition in three directions can comprehensively detect the distance from the obstacle and make correct calculation of the current position, thereby improving the obstacle avoidance capability.

The display screen is arranged on the other armrest of the wheelchair and is in circuit connection with the embedded main control board through the HDMI line.

Furthermore, the embedded main control board comprises a power supply circuit, an interface conversion circuit, an LED indicator lamp circuit, a main control BCM2839 circuit, an HDMI circuit, a wireless communication module circuit and a GPIO expansion interface circuit; the main control BCM2839 circuit outputs control signals to other peripheral circuits through an internal timer interface; the interface conversion circuit adopts a CAMERA interface ADC conversion circuit based on a UVC architecture protocol; the LED indicating lamp circuit receives a level signal of the main control BCM2839 circuit and outputs a response optical signal, the power supply circuit collects voltage through a Mico USB and provides 1.8V and 3.3V voltage for peripheral equipment through a DC-DC switch voltage stabilizing chip, and the power supply circuit is connected with the main control BCM2839 circuit and provides 5V working voltage; the HDMI circuit transmits the image and sound signals to the display screen through the HDMI interface; the GPIO expansion interface circuit is communicated with the peripheral circuit through a GPIO interface; the wireless communication module circuit is used for carrying out wireless data transmission with a network or a plurality of control terminals; the wireless module circuit can perform wireless networking based on https and Bluetooth 4 protocol standards, and transmits data or control instructions to the main control BCM2839 circuit through the UART interface after receiving the data.

Furthermore, the embedded main control board analyzes and processes the data transmitted by the image acquisition module and the sound acquisition module, sends a network request signal through the built-in wireless communication module, and then obtains a recognition processing signal and returns the recognition processing signal to the embedded main control board.

Furthermore, the embedded main control board receives the returned recognition processing signal, performs operation processing by using a machine learning algorithm to obtain an intention decision result, and performs packaging transmission by fusing voice signals.

Further, the wireless communication module sends the network request again and returns the analysis voice synthesis signal to the embedded main control board. The embedded main control board outputs voice signals, image signals and control signals through the HDMI circuit, the general GPIO interface circuit, the display screen and the external device in a communication mode, and therefore the artificial intelligence technology enables people and wheelchair equipment to be interactively energized.

Furthermore, when the movement is controlled, all the components (the distance measuring unit, the resolving unit and the obstacle avoidance unit) of the ultrasonic module can monitor the distance in real time and resolve the current position, obstacle avoidance interruption can be sent to the system when an obstacle is encountered, the system responds to the obstacle avoidance interruption and controls the wheelchair motor to execute the retreat action, and therefore the obstacle avoidance effect is achieved.

The utility model discloses for prior art have following advantage and effect:

the utility model discloses a novel mutual wheelchair of embedded AI intelligence can realize that wheelchair equipment and mankind carry out the intelligent interaction and intelligent control of humanoid ground. The floor application of artificial intelligence during mobile interaction is completed, and the requirement of people on rich intelligence is met.

Drawings

FIG. 1 is a diagram of a whole machine implementation model of the present invention;

fig. 2 is a software technology roadmap in the present invention;

fig. 3 is a system connection block diagram of the present invention;

fig. 4 is a schematic diagram of a main control circuit in the present invention;

fig. 5 is a schematic diagram of a digital conversion circuit of the image acquisition module in the present invention;

fig. 6 is a schematic diagram of an HDMI circuit according to the present invention;

fig. 7 is a schematic diagram of a GPIO expansion interface circuit in the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by a person skilled in the art without creative work belong to the protection scope of the present invention.

Examples

As shown in fig. 1, the novel embedded AI intelligent interactive wheelchair comprises a wheelchair, an embedded main control board, an image acquisition module, a sound acquisition module, a display screen and three ultrasonic modules. The image acquisition module can be used for acquiring human faces, and the sound acquisition module can be used for acquiring sound signals; the embedded main control panel is used for processing the acquired signals and then outputting image sound signals to a display screen or control signals; the ultrasonic wave module comprises a distance measuring unit, a resolving unit and an obstacle avoiding unit.

As shown in fig. 2, the artificial intelligence technology applied to the AI intelligent interactive wheelchair includes face micro-expression recognition, speech recognition, decision algorithm, NLP, speech synthesis, and the like. The human face micro-expression recognition converts a human face image signal into an expression emotion signal, the voice recognition model is used for converting a voice signal into a text signal, then a response speech text or an execution instruction signal is obtained through a decision algorithm and NLP processing, and the response speech text can be further converted into a voice signal through voice synthesis.

As shown in fig. 3, the hardware structure of the AI intelligent interactive wheelchair includes an embedded main control board, an image acquisition module, a sound acquisition module, a speaker, a display screen, and a communication interface circuit. The digital microphone camera is used as a sensor, the interface conversion circuit is used for discretizing sensor data into data which can be processed by the embedded main control board, the embedded main control board is used for analyzing the data and processing instructions and communicating with the server, the loudspeaker and other circuits are used as external circuits, and the server is responsible for providing artificial intelligence technical services.

As shown in fig. 4, the embedded main control board circuit adopts a boston BCM2839 chip to provide high-frequency signal processing and code instruction control.

As shown in fig. 5, the interface conversion circuit is a CAMERA interface circuit, and is used for processing signals such as images and sounds and sending the signals to the embedded main control board.

As shown in fig. 6, the HDMI circuit is used to display the image and sound signals output by the embedded main control board on the display screen.

As shown in fig. 7, the GPIO extension circuit is used to communicate with an external circuit to transmit control electrical signals to the devices of the wheelchair.

The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be equivalent replacement modes, and all are included in the scope of the present invention.

Claims

1. A novel embedded AI intelligent interactive wheelchair is characterized in that the intelligent interactive wheelchair comprises a wheelchair body, an embedded main control panel, an image acquisition module, a sound acquisition module, a display screen and three ultrasonic modules, wherein,

the embedded main control board is used as an AI central control core of the wheelchair and is arranged under the base of the wheelchair;

the image acquisition module is arranged on an armrest at one side of the wheelchair, is connected and supported by a long support with plasticity and is fixed in direction, and is in circuit connection and mutual communication with the embedded main control board through a USB connecting line;

the sound acquisition module is embedded in the back of the wheelchair seat, and is in circuit connection and mutual communication with the embedded main control board through the USB connecting line and the GPIO interface;

three ultrasonic modules in the ultrasonic modules are respectively deployed on the front side, the left side and the right side of the wheelchair, and are respectively in circuit connection and mutual communication with a general GPIO (general purpose input/output) interface of the embedded main control board through a DuPont wire;

the display screen is arranged on the armrest on the other side of the wheelchair and is in circuit connection and mutual communication with the embedded main control board through the HDMI line.

2. The novel embedded AI intelligent interactive wheelchair as claimed in claim 1, wherein said embedded main control board adopts NVIDIA/Jetson TX2 development board, adopts 64-bit Quad ARM A57 CPU, and has a dominant frequency of 2.5 GHZ.

3. The novel embedded AI intelligent interactive wheelchair as claimed in claim 1, wherein said image acquisition module is a binocular biopsy camera, the module circuit of which integrates a signal collector, an image splitter and a digital signal processor at the same time, complying with UVC protocol architecture, and being drive-free.

4. The novel embedded AI intelligent interactive wheelchair as claimed in claim 1, wherein said sound collection module employs a Respeaker 6MIC microphone module, wherein 6 MICs in the module are connected in parallel with each other and are connected in parallel with USB and GPIO interfaces on the embedded main control board for communication.

5. The novel embedded AI intelligent interactive wheelchair as claimed in claim 1, wherein said ultrasonic module is model HC-SR04 chip, three ultrasonic modules are connected with GPIO port circuit on the embedded main control board to realize communication, and distance measurement, position calculation and obstacle avoidance are performed by IO interruption mode.

6. The novel embedded AI intelligent interactive wheelchair as claimed in claim 1, wherein said display screen is a high definition capacitive screen with speakers for simultaneously performing voice and image signal processing and human interaction.