An artificial intelligence comprehensive experiment box

Abstract

The utility model provides an artificial intelligence comprehensive experiment box, which includes a visual detection module, a voice module, an intelligent sensor module, an edge computing module, a robotic arm, a main controller, peripheral devices and a display module; the edge computing module of the artificial intelligence comprehensive experiment box processes data close to the data source, can respond to and process complex tasks in real time, can effectively alleviate the burden of the main controller, and realize low-latency and high-efficiency data processing; the utility model supports combined experiments of various types of key technologies such as computer vision, voice processing, robotic arms, complex sensors and intelligent control, and can support simultaneous linkage experiments of various different modules, and is suitable for a wide range of teaching, experimental and actual operation scenarios.

Description

Technical Field

[0001] The utility model belongs to the field of intelligent experimental robots, in particular to an artificial intelligence comprehensive experimental box. Background Art

[0002] With the rapid development of artificial intelligence and automation technology, edge computing, computer vision, speech recognition and robotic arm technology are increasingly widely used in many fields such as industry, medical care, and education. However, traditional artificial intelligence and automation systems often require complex computing resources and high-performance processor support, which places high demands on the cost, power consumption and size of the equipment. At the same time, existing systems usually require multiple independent devices to handle tasks such as vision, speech and mechanical operation respectively, which not only increases the complexity of the system, but also poses challenges to the communication and collaboration between devices.

[0003] To solve these problems, edge computing terminals and multifunctional integrated systems have emerged. Edge computing terminals can process data close to the data source, reducing data transmission delays and bandwidth requirements, and improving real-time processing capabilities and data security.

[0004] Although edge computing, computer vision, speech recognition, and robotic arm technologies have made significant progress in their respective fields, there are still many challenges in integrating these technologies into a unified system to achieve efficient and collaborative work. Utility Model Content

[0005] In order to solve the above problems, the utility model provides an artificial intelligence comprehensive experiment box, which constructs a powerful, flexible and efficient automated experiment system by integrating edge computing, computer vision, speech recognition processing, robotic arms and intelligent sensing systems, which can be widely used in teaching experiments and actual operations.

[0006] The utility model provides an artificial intelligence comprehensive experiment box, which includes a visual detection module, a voice module, an intelligent sensor module, an edge computing module, a mechanical arm, a main controller, a peripheral device and a display module; the visual detection module and the voice module are respectively connected to the edge computing module, and the main controller is respectively connected to the intelligent sensor module, the edge computing module, the mechanical arm, the peripheral device and the display module;

[0007] The visual detection module transmits the real-time image or video of the sample collected in the experiment to the edge computing module for processing. The edge computing module recognizes and analyzes the received image and sends the recognition and analysis results to the main controller. The main controller controls the movement of the robot arm according to the analysis and recognition results, and sends the recognition and analysis results to the display module for display;

[0008] The voice module includes a microphone array and an embedded processor. The microphone array picks up the voice signal and sends it to the embedded processor for filtering. The embedded processor transmits the filtered voice signal to the edge computing module for semantic recognition analysis, extracts the voice control command and sends it to the main controller. The main controller controls the action of the robot arm according to the received voice control command.

[0009] The intelligent sensor module collects the environmental parameters inside and outside the intelligent experiment box and the box circuit data in real time and sends them to the main controller. The main controller controls the robot arm and the peripheral equipment to perform corresponding set actions according to the preset logic;

[0010] The main controller stores a preset control logic, and performs comparison and analysis based on the received data from the intelligent sensor module or the edge computing module. When the analysis conditions meet the preset control logic, the main controller outputs corresponding control signals to the robotic arm and peripheral devices to execute corresponding set actions.

[0011] Preferably, the visual inspection module includes a 2D visual inspection module and a 3D visual inspection module. The 2D visual inspection module is used for detecting and collecting images of plane samples inside the box, and the 3D visual inspection module is set up based on a two-degree-of-freedom cloud platform and can perform 360° rotation and pitch movements, and is used for detecting and collecting images of large-volume samples outside the box.

[0012] Preferably, the 3D visual detection module can also be used for face recognition authentication.

[0013] Preferably, the embedded processor in the voice module filters the voice signal picked up by the microphone array, specifically including performing noise reduction, filtering and normalization processing on the voice signal.

[0014] Preferably, the edge computing module uses a deep learning algorithm to perform recognition and analysis on the images received and transmitted by the visual detection module, and uses a machine learning algorithm to perform semantic recognition and analysis on the voice signals transmitted by the voice module.

[0015] Preferably, the edge computing module adopts NVIDIA's Jetson Xavier NX processor.

[0016] Preferably, the intelligent sensing module includes a temperature and humidity sensor, an air pressure sensor, a light sensor, a heart rate sensor, an ultrasonic sensor, a current sensor and a voltage sensor.

[0017] Preferably, the mechanical arm comprises a first-axis steering gear, a second-axis steering gear, a third-axis steering gear, a fourth-axis steering gear, a gripper steering gear and a gripper.

[0018] Preferably, the peripheral device includes a fan.

[0019] Preferably, the display module is a 17-inch high-definition display screen, including USB and HDMI interfaces.

[0020] The beneficial effects of the utility model are as follows: the edge computing module of the artificial intelligence comprehensive experiment box is pre-installed with commonly used deep learning frameworks: it supports direct deployment of training framework models such as Caffe, TensorFlow, and Pytorch. The edge computing module processes data close to the data source, can respond and process complex tasks in real time, can effectively alleviate the burden of the main controller, and achieve low-latency and high-efficiency data processing; the artificial intelligence comprehensive experiment box integrates supporting tutorials and source codes such as image processing, background extraction, face detection, and stereo vision, as well as supporting tutorials and source codes such as speech processing, speech recognition, and natural language processing. Users only need to have prerequisite courses in Linux, C or Python programming to carry out related course teaching and practice based on the experiment box; the artificial intelligence comprehensive experiment box contains storage cells, which support the storage of keyboards, mice, power adapters, and experimental teaching aids; the artificial intelligence comprehensive experiment box supports combined experiments of various key technologies such as computer vision, speech processing, robotic arms, complex sensors and intelligent control, and can support simultaneous linkage experiments of multiple different modules, which is suitable for a wide range of teaching, experimental and actual operation scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is the overall structure diagram of the device of the utility model;

[0022] Figure 2 It is a 3D visual detection module of the device of the utility model;

[0023] Figure 3 For the robotic arm.

[0024] In the figure: 1. 17-inch display screen; 2. Edge computing module; 3. 3D visual detection module; 4. Two-degree-of-freedom gimbal; 5. Voice module; 6. Intelligent sensing module; 7. Robotic arm; 8. 2D visual detection module; 9. 3D camera control servo; 10. Two-degree-of-freedom gimbal control servo; 11. Robotic arm first axis servo; 12. Robotic arm second axis servo; 13. Robotic arm third axis servo; 14. Robotic arm fourth axis servo; 15. Robotic arm gripper servo; 16. Robotic arm gripper. DETAILED DESCRIPTION

[0025] The utility model is further described below in conjunction with the accompanying drawings.

[0026] like Figure 1As shown, the utility model provides an artificial intelligence comprehensive experiment box, the main body of which is located inside a box. The artificial intelligence comprehensive experiment box includes a visual detection module, a voice module 5, an intelligent sensor module 6, an edge computing module 2, a mechanical arm 7, a main controller, peripheral devices and a display module; the visual detection module and the voice module are respectively connected to the edge computing module, and the main controller is respectively connected to the intelligent sensor module, the edge computing module, the mechanical arm, the peripheral devices and the display module;

[0027] The visual detection module transmits the real-time image or video of the sample collected in the experiment to the edge computing module for processing. The edge computing module recognizes and analyzes the received image and sends the recognition and analysis results to the main controller. The main controller controls the movement of the robot arm according to the analysis and recognition results, and sends the recognition and analysis results to the display module for display;

[0028] The voice module includes a microphone array and an embedded processor. The microphone array picks up the voice signal and sends it to the embedded processor for filtering. The embedded processor transmits the filtered voice signal to the edge computing module for semantic recognition analysis, extracts the voice control command and sends it to the main controller. The main controller controls the action of the robot arm according to the received voice control command.

[0029] The intelligent sensor module collects the environmental parameters inside and outside the intelligent experiment box and the box circuit data in real time and sends them to the main controller. The main controller controls the robot arm and the peripheral equipment to perform corresponding set actions according to the preset logic;

[0030] The main controller stores a preset control logic, and performs comparison and analysis based on the received data from the intelligent sensor module or the edge computing module. When the analysis conditions meet the preset control logic, the main controller outputs corresponding control signals to the robotic arm and peripheral devices to execute corresponding set actions.

[0031] Preferably, the visual inspection module includes a 2D visual inspection module 8 and a 3D visual inspection module 3. The 2D visual inspection module is used for detecting and collecting images of plane samples inside the box, and the 3D visual inspection module is set up based on a two-degree-of-freedom cloud platform 4, which can perform 360° rotation and pitch movements, and is used for detecting and collecting images of large-volume samples outside the box.

[0032] Preferably, the 3D visual detection module can also be used for face recognition authentication. Figure 2 As shown, the 3D visual detection module includes a 3D camera, a 3D camera control servo 9 and a two-degree-of-freedom gimbal control servo 10. The 3D camera mainly uses a 2-megapixel depth sensing camera. The camera uses a Sony IMX179CMOS chip, which is compatible with the USB3.0 interface protocol and can perform depth information perception, living body recognition, etc.

[0033] The 2D visual detection module includes a 2D camera, which uses a USB interface, has a resolution of 640*480, and a frame rate of 30bps.

[0034] Preferably, the embedded processor in the voice module filters the voice signal picked up by the microphone array, specifically including noise reduction, filtering and normalization of the voice signal. The voice module includes a programmable microphone array module, which supports functions such as sound recognition, sound source localization, and beamforming. The microphone array is a system composed of a certain number of acoustic sensors (microphones) that can be used to sample and process the spatial characteristics of the sound field. Because in a complex acoustic environment, noise and voice signals often overlap in time and spectrum, coupled with the influence of echo and reverberation, it is very difficult to capture relatively pure voice using a single microphone, and the microphone array integrates the space-time information of the voice signal, which can simultaneously extract the sound source and suppress noise.

[0035] Preferably, the edge computing module uses a deep learning algorithm to perform recognition and analysis on the images received and transmitted by the visual detection module, and uses a machine learning algorithm to perform semantic recognition and analysis on the voice signals transmitted by the voice module.

[0036] Preferably, the edge computing module adopts NVIDIA's Jetson Xavier NX processor.

[0037] Preferably, the intelligent sensing module includes a temperature and humidity sensor, an air pressure sensor, a light sensor, a heart rate sensor, an ultrasonic sensor, a current sensor and a voltage sensor.

[0038] Preferably, if Figure 3 As shown, the robotic arm includes a first axis steering gear 11, a second axis steering gear 12, a third axis steering gear 13, a fourth axis steering gear 14, a gripper steering gear 15 and a gripper 16. The visual detection module transmits the real-time image or video of the sample collected in the experiment to the edge computing module for processing. The edge computing module recognizes and analyzes the received image, obtains the world coordinates of the target and sends them to the main controller. The main controller solves the received world coordinates, calls the robot kinematics forward and inverse solution class function, obtains the joint coordinates of the target point of the robotic arm, and then calls the robot control class function, inputs the joint parameters, and controls the robotic arm to grasp the target.

[0039] Preferably, the peripheral device includes a fan.

[0040] Preferably, the display module is a 17-inch high-definition display screen 1, including USB and HDMI interfaces, and supporting operating systems such as NVIDIA, Windows, and Raspberry Pi.

[0041] The above is a detailed introduction to an artificial intelligence comprehensive experiment box provided by the utility model. This article uses specific examples to illustrate the principles and implementation methods of the utility model. The description of the above embodiments is only used to help understand the core idea of ​​the utility model. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the utility model, the utility model can also be improved and modified, and these improvements and modifications also fall within the scope of protection of the claims of the utility model.

Claims

1. An artificial intelligence comprehensive experiment box, characterized in that: The artificial intelligence comprehensive experiment box includes a visual detection module, a voice module, an intelligent sensor module, an edge computing module, a robotic arm, a main controller, a peripheral device and a display module; the visual detection module and the voice module are respectively connected to the edge computing module, and the main controller is respectively connected to the intelligent sensor module, the edge computing module, the robotic arm, the peripheral device and the display module; The visual detection module transmits the real-time image or video of the sample collected in the experiment to the edge computing module for processing. The edge computing module recognizes and analyzes the received image and sends the recognition and analysis results to the main controller. The main controller controls the movement of the robot arm according to the analysis and recognition results, and sends the recognition and analysis results to the display module for display; The voice module includes a microphone array and an embedded processor. The microphone array picks up the voice signal and sends it to the embedded processor for filtering. The embedded processor transmits the filtered voice signal to the edge computing module for semantic recognition analysis, extracts the voice control command and sends it to the main controller. The main controller controls the action of the robot arm according to the received voice control command. The intelligent sensor module collects the environmental parameters inside and outside the intelligent experiment box and the box circuit data in real time and sends them to the main controller. The main controller controls the robot arm and the peripheral equipment to perform corresponding set actions according to the preset logic; The main controller stores a preset control logic, and performs comparison and analysis based on the received data from the intelligent sensor module or the edge computing module. When the analysis conditions meet the preset control logic, the main controller outputs corresponding control signals to the robotic arm and peripheral devices to execute corresponding set actions.

2. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The visual inspection module includes a 2D visual inspection module and a 3D visual inspection module. The 2D visual inspection module is used for detecting and collecting images of plane samples inside the box. The 3D visual inspection module is set up based on a two-degree-of-freedom cloud platform and can perform 360° rotation and pitch movements, and is used for detecting and collecting images of large-volume samples outside the box.

3. The artificial intelligence comprehensive experiment box as claimed in claim 2, characterized in that: The 3D visual detection module can also be used for face recognition authentication.

4. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The embedded processor in the voice module filters the voice signal picked up by the microphone array, specifically including performing noise reduction, filtering and normalization processing on the voice signal.

5. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The edge computing module uses a deep learning algorithm to identify and analyze the images received from the visual detection module, and uses a machine learning algorithm to perform semantic recognition and analysis on the voice signals transmitted by the voice module.

6. The artificial intelligence comprehensive experiment box as claimed in claim 5, characterized in that: The edge computing module uses NVIDIA's Jetson Xavier NX processor.

7. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The intelligent sensing module includes a temperature and humidity sensor, an air pressure sensor, a light sensor, a heart rate sensor, an ultrasonic sensor, a current sensor and a voltage sensor.

8. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The mechanical arm comprises a first-axis steering gear, a second-axis steering gear, a third-axis steering gear, a fourth-axis steering gear, a gripper steering gear and a gripper.

9. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The peripheral device includes a fan.

10. The artificial intelligence comprehensive experiment box according to claim 1, characterized in that: The display module is a 17-inch high-definition display screen, including USB and HDMI interfaces.

Images