Night obstacle sensing and early warning device for unmanned vehicle
By using an Ethernet switch to connect lidar and infrared cameras in unmanned vehicles, combined with industrial control machines and voice conversion modules, the problem of difficulty in identifying obstacles at night is solved, and the accurate perception and early warning of obstacles at night is achieved, and the non-resistance of obstacles at night is expanded. Application scope of human-driving vehicles.
Patent Information
- Application Number
- CN202420798216.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-16
- Publication Date
- 2024-12-27
- Estimated Expiration
- 2034-04-16
AI Technical Summary
Unmanned vehicles cannot effectively identify obstacle types at night, limiting their application range, especially in dark environments where traditional visible light cameras cannot be accurately perceived.
The Ethernet switch is used to connect the lidar and multiple infrared cameras, combine the industrial control machine and voice conversion module, identify obstacles through infrared cameras and obtain accurate distances by using lidar, and issue early warning signals through indicator lights and speakers to achieve the accuracy of obstacles at night. Perception and early warning.
Accurate obstacle perception and early warning at night, expand the application range of driverless vehicles, and ensure that obstacles can be effectively identified and warned in night environments.
Smart Images
Figure CN222259582U_ABST
Abstract
Description
Technical Field
[0001] The utility model relates to the technical field of perception and warning devices, in particular to a nighttime obstacle perception and warning device for an unmanned vehicle. Background Art
[0002] The safe autonomous driving of driverless vehicles is based on accurate perception of their environment. In actual operation, these vehicles mainly rely on LiDAR and visible light binocular stereo cameras to capture environmental information. LiDAR can accurately detect the direction and distance of obstacles with its high-precision measurement capabilities, but it is unable to identify the type of obstacles. In contrast, although binocular stereo cameras can identify the type of obstacles, they have large errors and limitations in distance measurement.
[0003] At present, most unmanned vehicles use a combination of LiDAR and binocular stereo vision to achieve comprehensive perception of the environment during the day when there is sufficient light. However, once it is night, traditional visible light cameras lose their function, resulting in unmanned vehicles being unable to effectively identify the type of obstacles in the dark. This limitation greatly restricts the application scope of unmanned vehicles, and severely limits their application in scenarios such as night driving. Utility Model Content
[0004] The purpose of the utility model is to provide an obstacle sensing and warning device for unmanned vehicles at night, so as to solve the problem that unmanned vehicles based on visible light cameras to identify obstacles cannot work normally at night and expand the application scope of unmanned vehicles.
[0005] To achieve the above-mentioned purpose, the utility model provides the following technical solutions: a nighttime obstacle perception and warning device for an unmanned vehicle, comprising an Ethernet switch, wherein the outer side of the Ethernet switch is connected to a laser radar through a network port, the outer side of the Ethernet switch is connected to an industrial computer through a network port, the outer side of the Ethernet switch is connected to an infrared camera 1 through a network port, the outer side of the Ethernet switch is connected to an infrared camera 2 through a network port, the outer side of the Ethernet switch is connected to an infrared camera 3 through a network port, the outer side of the Ethernet switch is connected to an infrared camera 4 through a network port, the outer side of the Ethernet switch is connected to a DA conversion module through a wire, the DA conversion module is connected to an indicator light 1 through a network port, the DA conversion module is connected to an indicator light 2 through a network port, the DA conversion module is connected to an indicator light 3 through a network port, and the DA conversion module is connected to an indicator light 4 through a network port.
[0006] Preferably, the industrial computer uses the Linux system, and the Ethernet switch has multiple LAN ports with a rate of 1000Mbps. The industrial computer serves as a core computing unit and is connected to the Ethernet switch through a network port.
[0007] Preferably, the horizontal field of view angles of the infrared camera 1, infrared camera 2, infrared camera 3, and infrared camera 4 are all 90°, and the wavelength is 8-14 μm. The infrared camera 1, infrared camera 2, infrared camera 3, and infrared camera 4 are respectively oriented in different directions, and the angles between them are 90°. By setting the infrared camera 1, infrared camera 2, infrared camera 3, and infrared camera 4, it is used to collect night infrared images in different directions.
[0008] Preferably, the laser radar has a measurement distance of not less than 50m and an accuracy of not less than 1cm, and the laser radar is located at the center of infrared camera 1, infrared camera 2, infrared camera 3, and infrared camera 4 and is higher than the installation plane of infrared camera 1, infrared camera 2, infrared camera 3, and infrared camera 4. By setting up the laser radar, the precise distance of obstacles in all directions can be obtained.
[0009] Preferably, the outer side of the Ethernet switch is connected to a voice conversion module via a wire, and the voice conversion module is connected to a speaker via a wire. A text instruction is sent to the voice control module via the network to control the speaker to send a corresponding voice signal.
[0010] The beneficial effects of the utility model are as follows:
[0011] The utility model identifies obstacles through infrared cameras one, two, three and four, senses the distance and direction of obstacles through laser radar, and sends out light and sound warning signals through indicator lights one, two, three and four and a speaker. The system is connected through Ethernet, has reliable data communication, is easy to operate, and can realize accurate obstacle perception and warning at night. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 It is a schematic diagram of the overall structure of the utility model.
[0013] In the figure: 1. Ethernet switch; 2. LiDAR; 3. Industrial computer; 4. Infrared camera 1; 5. Infrared camera 2; 6. Infrared camera 3; 7. Infrared camera 4; 8. Voice conversion module; 9. Speaker; 10. DA conversion module; 11. Indicator light 1; 12. Indicator light 2; 13. Indicator light 3; 14. Indicator light 4. DETAILED DESCRIPTION
[0014] The following will be combined with the drawings in the embodiments of the utility model to clearly and completely describe the technical solutions in the embodiments of the utility model. Obviously, the described embodiments are only part of the embodiments of the utility model, not all of the embodiments. Based on the embodiments in the utility model, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the utility model.
[0015] See also Figure 1 A nighttime obstacle perception and warning device for an unmanned vehicle comprises an Ethernet switch 1, a laser radar 2 is connected to the outside of the Ethernet switch 1 through a network port, an industrial computer 3 is connected to the outside of the Ethernet switch 1 through a network port, the industrial computer 3 adopts a Linux system, the Ethernet switch 1 has multiple LAN ports with a rate of 1000Mbps, the industrial computer 3 is used as a core computing unit and is connected to the Ethernet switch 1 through a network port, an infrared camera 1 4 is connected to the outside of the Ethernet switch 1 through a network port, an infrared camera 2 5 is connected to the outside of the Ethernet switch 1 through a network port, an infrared camera 3 6 is connected to the outside of the Ethernet switch 1 through a network port, an infrared camera 4 7 is connected to the outside of the Ethernet switch 1 through a network port, infrared camera 1 4 and infrared camera 2 5 The horizontal field of view angles of infrared cameras 1, 2, 5, 3, 6 and 4 are all 90°, and the wavelength is 8-14μm. Infrared cameras 1, 2, 5, 3, 6 and 4 are facing different directions, and the angle between them is 90°. Infrared cameras 1, 2, 5, 3, 6 and 4 are set to collect nighttime infrared images in different directions. The measuring distance of laser radar 2 is not less than 50m, and the accuracy is not less than 1cm. Laser radar 2 is located at the center of infrared cameras 1, 2, 5, 3, 6 and 4 and is higher than the installation plane of infrared cameras 1, 2, 5, 3, 6 and 4. By setting laser radar 2, the precise distance of obstacles in all directions can be obtained.
[0016] See also Figure 1 A voice conversion module 8 is connected to the outside of the Ethernet switch 1 through a wire, and a speaker 9 is connected to the voice conversion module 8 through a wire. A text instruction is sent to the voice control module 8 through the network to control the speaker 9 to send a corresponding voice signal. A DA conversion module 10 is connected to the outside of the Ethernet switch 1 through a wire, and an indicator light 11 is connected to the DA conversion module 10 through a network port. An indicator light 2 12 is connected to the DA conversion module 10 through a network port. An indicator light 3 13 is connected to the DA conversion module 10 through a network port. An indicator light 4 14 is connected to the DA conversion module 10 through a network port.
[0017] The specific implementation process of the utility model is as follows: when the device is working, the industrial computer 3 collects images from infrared camera 1 4, infrared camera 2 5, infrared camera 3 6, and infrared camera 4 7 through the network, pre-processes the images respectively, and determines whether there are obstacles in the images through the image recognition algorithm. If there are obstacles, the position is preliminarily estimated through the two-dimensional coordinates of the obstacles. The industrial computer 3 then collects data from the laser radar 2 through the network to obtain the precise distance of the obstacles in all directions. The industrial computer 3 sends control information to the DA conversion module 10 through the network to control the indicator light 1 11, the indicator light 2 12, and the indicator light 3 1 3. The intensity of the light of indicator light four 14 is proportional to the distance of the obstacle. Indicator light one 11, indicator light two 12, indicator light three 13, and indicator light four 14 correspond to one of infrared camera one 4, infrared camera two 5, infrared camera three 6, and infrared camera four 7 respectively, reflecting the distance of the obstacle in that direction. The closer the distance, the brighter the light, and the farther the distance, the dimmer the light. Obstacle warning is carried out in the form of light. The industrial computer 3 sends text instructions to the voice control module 8 through the network to control the speaker 9 to send out corresponding voice signals. The function of the voice control module 8 is to convert text into voice to realize voice warning.
[0018] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A nighttime obstacle sensing and warning device for an unmanned vehicle, comprising an Ethernet switch (1), characterized in that: The outer side of the Ethernet switch (1) is connected to a laser radar (2) through a network port, the outer side of the Ethernet switch (1) is connected to an industrial computer (3) through a network port, the outer side of the Ethernet switch (1) is connected to an infrared camera 1 (4) through a network port, the outer side of the Ethernet switch (1) is connected to an infrared camera 2 (5) through a network port, the outer side of the Ethernet switch (1) is connected to an infrared camera 3 (6) through a network port, the outer side of the Ethernet switch (1) is connected to an infrared camera 4 (7) through a network port, the outer side of the Ethernet switch (1) is connected to a DA conversion module (10) through a wire, the DA conversion module (10) is connected to an indicator light 1 (11) through a network port, the DA conversion module (10) is connected to an indicator light 2 (12) through a network port, the DA conversion module (10) is connected to an indicator light 3 (13) through a network port, and the DA conversion module (10) is connected to an indicator light 4 (14) through a network port.
2. The nighttime obstacle sensing and warning device for an unmanned vehicle according to claim 1, characterized in that: The industrial computer (3) adopts the Linux system, and the Ethernet switch (1) has multiple LAN ports with a rate of 1000 Mbps.
3. The nighttime obstacle sensing and warning device for an unmanned vehicle according to claim 1, characterized in that: The horizontal field of view angles of the infrared camera 1 (4), the infrared camera 2 (5), the infrared camera 3 (6), and the infrared camera 4 (7) are all 90°, and the wavelength is 8-14 μm. The infrared camera 1 (4), the infrared camera 2 (5), the infrared camera 3 (6), and the infrared camera 4 (7) are respectively facing different directions, and the angle between them is 90°.
4. The nighttime obstacle sensing and warning device for an unmanned vehicle according to claim 1, characterized in that: The laser radar (2) has a measurement distance of not less than 50 m and an accuracy of not less than 1 cm. The laser radar (2) is located at the center of the infrared camera 1 (4), the infrared camera 2 (5), the infrared camera 3 (6), and the infrared camera 4 (7) and is higher than the installation plane of the infrared camera 1 (4), the infrared camera 2 (5), the infrared camera 3 (6), and the infrared camera 4 (7).
5. The nighttime obstacle sensing and warning device for an unmanned vehicle according to claim 1, Features: The outer side of the Ethernet switch (1) is connected to a voice conversion module (8) via a wire. The voice conversion module (8) is connected to a speaker (9) via a wire.