Anti-collision system of intelligent mobile robot

Abstract

The invention relates to the technical field of anti-collision of intelligent mobile robots, and discloses an anti-collision system of an intelligent mobile robot, which comprises a data acquisition module, an image processing module, a region analysis module, a reflection intensity calculation module and an intelligent avoidance module, the technical problems that in the prior art, when a reflecting surface obstacle such as glass and a mirror surface with high reflectivity exists, reflected signal interference exists, the reflecting surface is difficult to recognize, and consequently a system cannot accurately evaluate the influence of the reflecting surface on robot movement are solved. According to the invention, through multi-source data fusion, reflection surface feature extraction and reflection intensity calculation, the reflection surface and the influence of the reflection surface can be accurately identified in real time, so that collision between the robot and a reflection surface obstacle is effectively avoided. According to the invention, the anti-collision capability of the intelligent mobile robot is improved, and the safety of the robot in a complex environment, especially in a scene with more reflecting surfaces, is ensured.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent mobile robot anti-collision, and in particular relates to an anti-collision system of an intelligent mobile robot. Background Art

[0002] At present, with the rapid development of intelligent mobile robot technology, robots are increasingly used in various complex environments, such as industrial automation, warehousing and logistics, smart home and other fields. In order to ensure that robots can interact with the surrounding environment safely and effectively, anti-collision technology has become one of the core technologies of intelligent mobile robots. Existing anti-collision systems are generally based on sensors such as lasers, ultrasound, and vision, and rely on robots to perceive the distance information or image features of obstacles. However, the existing technology has obvious shortcomings in certain specific scenarios, especially in environments with reflective surfaces such as glass and mirrors. For example, when a laser sensor encounters a reflective surface, the reflected beam may interfere, causing the sensor to misjudge the position of the obstacle; the existing technology cannot fully meet the needs of robots being able to accurately identify reflective surfaces and effectively avoid collisions in complex environments with reflective obstacles. Especially when the robot needs to interact with objects with mirror reflections or transparent glass surfaces, traditional anti-collision technology often cannot correctly evaluate the true shape and position of the obstacle, resulting in obstacle avoidance failure. Therefore, there is an urgent need for an anti-collision system for an intelligent mobile robot that can accurately and real-time identify reflective obstacles and calculate their anti-collision technology for the robot's movement when facing obstacles with many reflective surfaces, so as to improve the safety and stability of the robot in complex environments, ensure that it can operate reliably in complex dynamic environments, and avoid collisions caused by interference from reflective surfaces. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to propose an anti-collision system for an intelligent mobile robot, aiming to solve the technical problems in the prior art that when encountering reflective surface obstacles such as glass, mirrors and other highly reflective surfaces, there is interference with reflected signals and difficulty in identifying the reflective surface, resulting in the system being unable to accurately evaluate the impact of the reflective surface on the robot's movement.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides an anti-collision system for an intelligent mobile robot.

[0005] The anti-collision system of the intelligent mobile robot comprises:

[0006] A data acquisition module is used to collect the robot's environment data and the robot's current state information using multi-source sensors;

[0007] The multi-source sensor includes laser sensor, visual sensor and IMU sensor; among them, the laser sensor is used to collect obstacle position point cloud information; the visual sensor is used to collect obstacle image information; the IMU sensor is used to collect robot posture data;

[0008] An image processing module is used to fuse the obstacle position point cloud information collected by the laser sensor and the obstacle image information collected by the visual sensor to obtain fused image data. After the fused image data is denoised, the Canny operator is used to detect obstacle boundary points in the image, and only the first reflection surface area where the obstacle position point cloud information is located is retained. The findContours function in OpenCV is used to segment the first reflection surface area in the fused image data, and the first reflection surface area is extracted;

[0009] In the first reflection surface area, Gabor filtering is used to extract specular reflection features, and the second reflection surface area is extracted by screening the area in the first reflection surface area where the pixel brightness is greater than a preset brightness threshold;

[0010] The area analysis module is used to calculate the pixel area A of the second reflective surface area by pixel accumulation. pixels , set the calibration parameter τ, according to the pixel area A of the second reflective surface area pixels And set the calibration parameter τ to calculate the actual reflection surface area A reflect ;

[0011] The reflection intensity calculation module is used to obtain the distance between the robot and the obstacle from the obstacle distance information, and calculate the reflection intensity I by combining the actual reflection surface area, the incident angle between the laser and the obstacle surface, and the laser emission power of the laser sensor. reflect ;

[0012] Intelligent avoidance module, used to set dynamic reflection intensity threshold I threshold , the formula is:

[0013] I threshold =k1·d lidar +k2·A reflect ·cosθ

[0014] Among them, k1 is the distance weight coefficient between the robot and the obstacle, and k2 is the actual reflection surface area weight coefficient;

[0015] Set the preset safety distance d safe , when the reflection intensity I reflect Greater than the dynamic reflection intensity threshold I threshold And the distance between the robot and the obstacle is less than the preset safety distance d safe , triggers the collision avoidance command.

[0016] Preferably, in the area analysis module, the actual reflection surface area A reflect The calculation formula is:

[0017] A reflect =A pixels ·τ

[0018]

[0019] Among them, real is the actual distance from the reflective surface to the camera, and camera is the calibrated reference distance.

[0020] Preferably, in the reflection intensity calculation module, the reflection intensity calculation formula is:

[0021]

[0022] Among them, P transmit is the laser emission power of the laser sensor, A reflect is the actual reflection surface area, θ is the incident angle between the laser and the obstacle surface, d lidar is the distance between the robot and the obstacle.

[0023] Preferably, in the data acquisition module, the posture data includes robot position data and movement direction information.

[0024] Preferably, in the image processing module, the denoising process adopts Gaussian filtering process.

[0025] Preferably, in the reflection intensity calculation module, the incident angle between the laser and the obstacle surface is calculated by the laser reflection direction measured by the laser sensor and the normal vector of the second reflection surface area.

[0026] Preferably, the calculation formula for the incident angle between the laser and the obstacle surface is:

[0027]

[0028] Wherein, L is the propagation direction of the laser, N is the normal vector of the obstacle surface, |L| and |N| are the normal vector modulus of the laser propagation direction and the second reflection surface area, respectively.

[0029] The beneficial effect of the present invention is that compared with the prior art, when encountering reflective surface obstacles such as glass, mirrors and other highly reflective surfaces, there are interference with reflected signals and difficulty in identifying the reflective surface, which leads to the technical problem that the system cannot accurately evaluate the impact of the reflective surface on the robot's movement; because the present application can accurately identify the reflective surface and its impact in real time through multi-source data fusion, reflective surface feature extraction and reflection intensity calculation, it can effectively avoid the robot from colliding with the reflective surface obstacle. The present invention improves the anti-collision capability of the intelligent mobile robot and ensures the safety of the robot in complex environments, especially in scenes with many reflective surfaces. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 The figure is a flow chart of a first embodiment of an anti-collision system for an intelligent mobile robot according to the present invention.

[0032] Figure 2 The figure is a schematic diagram of the filtering effect of the anti-collision system of an intelligent mobile robot according to the present invention.

[0033] Figure 3 This is a schematic diagram of the incident angle of the incident laser path of the anti-collision system of an intelligent mobile robot according to the present invention.

[0034] Figure 4 The present invention is a schematic diagram of an anti-collision system for an intelligent mobile robot. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Embodiment 1: Figure 1 2 is a flow chart of the first embodiment of the anti-collision system for an intelligent mobile robot according to the present invention, and provides the first embodiment of the anti-collision system for an intelligent mobile robot according to the present invention.

[0037] In a first embodiment, the anti-collision system of the intelligent mobile robot includes:

[0038] A data acquisition module is used to collect the robot's environment data and the robot's current state information using multi-source sensors;

[0039] The multi-source sensor includes laser sensor, visual sensor and IMU sensor; among them, the laser sensor is used to collect obstacle position point cloud information; the visual sensor is used to collect obstacle image information; the IMU sensor is used to collect robot posture data;

[0040] It should be noted that the laser sensor can provide the distance and position data of obstacles, while the visual sensor can provide more details at the image level, such as the shape and material of the obstacles, and the IMU sensor supplements the robot's posture change data in space. The combination of the three can provide comprehensive information for subsequent reflective surface detection and obstacle avoidance decisions. By integrating these data, the system can have a more comprehensive understanding of the state of the surrounding environment.

[0041] An image processing module is used to fuse the obstacle position point cloud information collected by the laser sensor and the obstacle image information collected by the visual sensor to obtain fused image data. After the fused image data is denoised, the Canny operator is used to detect obstacle boundary points in the image, and only the first reflection surface area where the obstacle position point cloud information is located is retained. The findContours function in OpenCV is used to segment the first reflection surface area in the fused image data, and the first reflection surface area is extracted;

[0042] In the first reflection surface area, Gabor filtering is used to extract specular reflection features, and the second reflection surface area is extracted by screening the area in the first reflection surface area where the pixel brightness is greater than a preset brightness threshold;

[0043] It should be noted that the core goal of the image processing module is to fuse the data of the laser and visual sensors in order to accurately locate and identify obstacles in the image. Boundary point detection through the Canny operator helps to extract the outline of the obstacle and enhance the features of the reflective surface in the image, ensuring that the reflective surface can be effectively identified and processed in subsequent steps; in the first reflective surface area, Gabor filtering is used to extract the mirror reflection features, and then the second reflective surface area is extracted by screening the area where the pixel brightness in the first reflective surface area is greater than the preset brightness threshold; in this way, the image processing module can not only accurately detect the obstacle boundary, but also identify the reflective surface area, further providing effective data support for subsequent reflection intensity calculation and robot collision avoidance.

[0044] For example, if Figure 2As shown in the figure, a simple reflective surface area is simulated to demonstrate the effects of Canny edge detection and Gabor filtering. Canny edge detection extracts obvious edges in the image, especially the contour edges of the reflective surface area, and can accurately locate the boundaries of obstacles. The Gabor filter enhances the texture features in the horizontal direction at 0°, and it can be seen that the edge area of ​​the reflective surface is highlighted. The Gabor filter further enhances the response to the texture in the 45° direction at 45°, clearly showing the edge features and details of the reflective surface area.

[0045] The area analysis module is used to calculate the pixel area A of the second reflective surface area by pixel accumulation. pixels , set the calibration parameter τ, according to the pixel area A of the second reflective surface area pixels And set the calibration parameter τ to calculate the actual reflection surface area A reflect ;

[0046] It should be noted that the second reflective surface area extracted from the image is first statistically analyzed at the pixel level, and the pixel area is converted into the actual reflective surface area through the set calibration parameters. Specifically, the calibration parameters take into account factors such as the camera's viewing angle, resolution, and distance to obstacles to ensure that the calculation result of the reflective surface area is consistent with the reflective area in the actual environment. The effect of this technical step is that it can accurately calculate the reflective surface area and provide reliable data support for subsequent reflection intensity calculations, thereby enhancing the ability of the entire anti-collision system to identify and avoid reflective surface obstacles in complex environments. In addition, the combination of image processing and multi-source data fusion solutions can significantly improve the accuracy and real-time performance of reflective surface detection, solving the problem of the inability to accurately identify reflective surface obstacles in the prior art.

[0047] The reflection intensity calculation module is used to obtain the distance between the robot and the obstacle from the obstacle distance information, and calculate the reflection intensity I by combining the actual reflection surface area, the incident angle between the laser and the obstacle surface, and the laser emission power of the laser sensor. reflect ;

[0048] It should be noted that the calculation of reflection intensity fully considers the effects of laser emission power, reflection surface area, incident angle and distance on reflection intensity. The laser emission power determines the initial intensity of the reflection signal. The reflection surface area directly affects the intensity of the reflected light. Larger reflection surfaces usually reflect more laser signals. The incident angle affects the degree of interaction between the laser and the obstacle surface. Smaller incident angles usually cause more lasers to be reflected back to the sensor. The square of the distance will attenuate the reflection intensity, that is, the farther the distance, the weaker the reflection intensity.

[0049] It can be understood that the formula for calculating reflection intensity accurately simulates the interaction between laser and reflecting surface by comprehensively considering multiple factors, especially the square attenuation effect of distance, so that the system can reasonably evaluate the contribution of distant obstacles to the reflected signal and avoid the false triggering of avoidance instructions due to the excessive reflection intensity of distant obstacles.

[0050] Intelligent avoidance module, used to set dynamic reflection intensity threshold I threshold , the formula is:

[0051] I thresholf =k1·d lidar +k2·A reflect ·cosθ

[0052] Among them, k1 is the distance weight coefficient between the robot and the obstacle, and k2 is the actual reflection surface area weight coefficient;

[0053] Set the preset safety distance d safe , when the reflection intensity I reflect Greater than the dynamic reflection intensity threshold I threshold And the distance between the robot and the obstacle is less than the preset safety distance d safe , triggers the collision avoidance command.

[0054] It should be noted that the calculation formula for the incident angle between the laser and the obstacle surface is:

[0055]

[0056] Where L is the propagation direction of the laser, N is the normal vector of the obstacle surface, |L| and |N| are the normal vector modulus of the laser propagation direction and the second reflection surface area respectively;

[0057] like Figure 3 As shown, L is the propagation direction of the laser, N is the normal vector of the obstacle surface, and the incident angle is 30°; combined with the reflection intensity I reflect and safety distance d safe It can comprehensively judge the position and type of obstacles in multiple dimensions. For example, at a small incident angle (θ→0), the reflected light is strong and the threshold will be automatically adjusted to avoid false triggering. At a large incident angle (θ→90°), the reflected light is weak and the system dynamically lowers the threshold to ensure that smaller reflected signals can be detected. The setting of the dynamic reflection intensity threshold takes into account two main factors: distance and reflection surface area. The distance weight coefficient k1 reflects the impact of distance on the reflection intensity threshold, that is, the closer the robot is to the obstacle, the threshold will be adjusted accordingly to increase the system's sensitivity to close obstacles. The reflection surface area weight coefficient k2 reflects the impact of the reflection surface area on the threshold. The larger the area, the stronger the reflection intensity, and the system will use a higher threshold to determine whether there is a collision risk.

[0058] It is understandable that the dynamic reflection intensity threshold not only depends on the reflection intensity, but also adjusts according to the distance between the robot and the obstacle and the change in the reflective surface area. This enables the system to respond more accurately to real-time changing environmental conditions, especially when the robot approaches an obstacle. The threshold adjustment will enable the system to identify possible collision risks in a timely manner.

[0059] It should be understood that by setting the dynamic reflection intensity threshold, the system can compare the actual reflection intensity of the obstacle with the safety distance. When the reflection intensity is higher than the set threshold and the distance between the robot and the obstacle is less than the preset safety distance, the system will determine that there is a collision risk and trigger a collision avoidance command, thereby effectively avoiding a collision.

[0060] For example, assuming that the distance between the robot and the obstacle is 1 meter, the reflective surface area is large, and the calculated dynamic reflection intensity threshold is 0.8, when the reflection intensity is greater than the threshold and the actual distance between the robot and the obstacle is less than the preset safety distance of 2 meters, the system will trigger a collision avoidance command to avoid collision with the obstacle; on the contrary, if the distance is far or the reflective surface area is small, and the reflection intensity does not exceed the threshold, the system will not trigger an avoidance command.

[0061] In addition, the anti-collision system of an intelligent mobile robot provided by the present invention adopts the anti-collision system of an intelligent mobile robot in the above embodiment, which can solve the technical problem of anti-collision of an intelligent mobile robot. Compared with the prior art, the beneficial effects of the anti-collision system of an intelligent mobile robot provided by the present invention are the same as the beneficial effects of the anti-collision system of an intelligent mobile robot provided by the above embodiment, and the other technical features of the anti-collision system of the intelligent mobile robot are the same as the features disclosed in the above embodiment system, which will not be repeated here.

[0062] The present invention provides an anti-collision device for an intelligent mobile robot. Figure 4, an anti-collision device for an intelligent mobile robot comprises: at least one processor; and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the anti-collision system for an intelligent mobile robot in the above-mentioned embodiment 1. An anti-collision device for an intelligent mobile robot in an embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. An anti-collision device for an intelligent mobile robot is merely an example and should not impose any limitations on the functions and scope of use of the embodiments of the present invention. An anti-collision device for an intelligent mobile robot may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of an anti-collision device for an intelligent mobile robot are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow an anti-collision device of an intelligent mobile robot to communicate wirelessly or wired with other devices to exchange data. Although the figure shows an anti-collision device of an intelligent mobile robot with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0063] The present invention also provides a computer program product, including a computer program, which implements the steps of the anti-collision system of an intelligent mobile robot as described above when executed by a processor. The computer program product provided by the present invention can solve the technical problem of anti-collision of an intelligent mobile robot. Compared with the prior art, the beneficial effects of the computer program product provided by the present invention are the same as the beneficial effects of the anti-collision system of an intelligent mobile robot provided by the above embodiment, which will not be repeated here.

[0064] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes the program code of the system shown. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the system of the embodiment disclosed in the present invention are executed.

[0065] It should be understood that the various parts disclosed in the present invention can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An anti-collision system for an intelligent mobile robot, characterized in that: The system includes: A data acquisition module is used to collect the robot's environment data and the robot's current state information using multi-source sensors; The multi-source sensor includes a laser sensor, a visual sensor and an IMU sensor; the laser sensor is used to collect point cloud information of obstacle positions; the visual sensor is used to collect image information of obstacles; IMU sensors are used to collect robot posture data; An image processing module is used to fuse the obstacle position point cloud information collected by the laser sensor and the obstacle image information collected by the visual sensor to obtain fused image data. After the fused image data is denoised, the Canny operator is used to detect obstacle boundary points in the image, and only the first reflection surface area where the obstacle position point cloud information is located is retained. The findContours function in OpenCV is used to segment the first reflection surface area in the fused image data, and the first reflection surface area is extracted; In the first reflection surface area, Gabor filtering is used to extract specular reflection features, and the second reflection surface area is extracted by screening the area in the first reflection surface area where the pixel brightness is greater than a preset brightness threshold; The area analysis module is used to calculate the pixel area A of the second reflective surface area by pixel accumulation. pixels , set the calibration parameter τ, according to the pixel area A of the second reflective surface area pixels And set the calibration parameter τ to calculate the actual reflection surface area A reflect ; The reflection intensity calculation module is used to obtain the distance between the robot and the obstacle from the obstacle distance information, and calculate the reflection intensity I by combining the actual reflection surface area, the incident angle between the laser and the obstacle surface, and the laser emission power of the laser sensor. reflect ; Intelligent avoidance module, used to set dynamic reflection intensity threshold I threshold , the formula is: I threshold =k1·d lidar +k2·A reflect ·cosθ Among them, k1 is the distance weight coefficient between the robot and the obstacle, and k2 is the actual reflection surface area weight coefficient; Set the preset safety distance d safe , when the reflection intensity I reflect Greater than the dynamic reflection intensity threshold I threshold And the distance between the robot and the obstacle is less than the preset safety distance d safe , triggers the collision avoidance command.

2. The anti-collision system for an intelligent mobile robot according to claim 1, characterized in that: In the regional analysis module, the actual reflective surface area A reflect The calculation formula is: A reflect =A pixels ·t Among them, real is the actual distance from the reflective surface to the camera, and camera is the calibrated reference distance.

3. The anti-collision system for an intelligent mobile robot according to claim 1, characterized in that: In the reflection intensity calculation module, the reflection intensity calculation formula is: Among them, P transmit is the laser emission power of the laser sensor, A reflect is the actual reflection surface area, θ is the incident angle between the laser and the obstacle surface, d lidar is the distance between the robot and the obstacle.

4. The anti-collision system for an intelligent mobile robot according to claim 1, characterized in that: In the data acquisition module, the posture data includes the robot position data and movement direction information.

5. The anti-collision system for an intelligent mobile robot according to claim 1, characterized in that: In the image processing module, Gaussian filtering is used for noise removal.

6. The anti-collision system for an intelligent mobile robot according to claim 1, characterized in that: In the reflection intensity calculation module, the incident angle between the laser and the obstacle surface is calculated by the laser reflection direction measured by the laser sensor and the normal vector of the second reflection surface area.

7. The anti-collision system for an intelligent mobile robot as claimed in claim 6, characterized in that: The formula for calculating the incident angle between the laser and the obstacle surface is: Wherein, L is the propagation direction of the laser, N is the normal vector of the obstacle surface, |L| and |N| are the normal vector modulus of the laser propagation direction and the second reflection surface area, respectively.

Images