Intelligent light control system of machine pet and control method thereof

Abstract

The invention relates to the field of robots, in particular to an intelligent light control system and method for a machine pet, and the system comprises an environment detection device, a processing device, and a plurality of light-emitting devices disposed on the machine pet. The environment detection device comprises a radar sensor, an illuminance sensor, a visual sensor and a voice recognition sensor, the processing device comprises an image processing module, a voice processing module and a control module, and the image processing module obtains environment information collected by the visual sensor and preprocesses the environment information to obtain image information; the voice processing module preprocesses the collected environment sound information to obtain a sound instruction; the control module processes the obstacle distance, speed, illumination intensity, image and sound instructions, generates control instructions and performs corresponding control on the light emitting module; the light-emitting devices comprise an ADB module, a DLP module and a contour lamp module. According to the invention, high-beam and low-beam switching can be carried out, the identification degree is high, the night walking safety is high, and the user experience feeling is good.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of robots, in particular to an intelligent light control system of a robot pet and a control method thereof. BACKGROUND

[0002] In the past decade, intelligent robot systems have rapidly developed, and even commercial products have been introduced into the market. Typical examples include the AIBO entertainment robot pet designed and manufactured by SONY Company and the ASIMO humanoid robot produced by Honda Company. These robots have always existed as robot assistants, including robot pets designed to accompany their owners. The existing robot pets can guide the user when walking at night, such as turning on the light source for illumination, so that the user can also see the road when walking at night. The existing robot pets are equipped with only a single lighting function lamp, and there is no special light system. The existing system has the following disadvantages: (1) Poor lighting adaptability, unable to automatically switch between high and low beams according to the environment in front, easy to cause glare interference to pedestrians or vehicles at night, and potential safety hazards; (2) Lack of environmental interaction, lack of marking function for dangerous areas on the road, and the robot cannot prompt the user or the surrounding target about steps, potholes, etc. when walking at night; (3) Low profile recognition, no clear profile display at night, and it is difficult for surrounding people or vehicles to quickly determine the size and motion trajectory of the robot, which is prone to collision; (4) Single interaction mode, light parameter adjustment depends on remote control or physical buttons, and lacks convenient voice interaction means, which is low in user operation efficiency and poor in user experience. SUMMARY

[0003] The present application aims to at least solve one of the technical problems existing in the prior art.

[0004] To this end, the present application provides an intelligent light control system of a robot pet, which can switch between high and low beams, has high recognition, improves the safety of walking at night, and has good user experience.

[0005] According to an embodiment of the present application, an intelligent light control system of a robot pet, the system comprises: an environment detection device, a processing device, and a plurality of light emitting devices arranged on the robot pet; The environment detection device comprises: a radar sensor configured to collect the distance and speed of an obstacle in front; an illumination sensor configured to collect the light intensity information of the surrounding environment; a vision sensor configured to collect the environmental information of the surrounding environment; A voice recognition sensor is configured to collect voice instruction information of a user; The processing device comprises: An image processing module is connected with the visual sensor, and the image processing module is configured to acquire environmental information collected by the visual sensor and perform preprocessing to acquire image information. A voice processing module is connected with the voice recognition sensor, and the voice processing module is configured to perform preprocessing on collected environmental sound information to acquire sound instructions. A control module is connected with the radar sensor, the illumination sensor, the image processing module, the voice processing module and the light emitting module, and the control module is configured to process distance, speed, illumination intensity information, image information and sound instructions of the obstacle and generate control instructions, and the control module executes corresponding control on the corresponding light emitting module according to the control instructions. The light emitting device is an ADB module, a DLP module and a contour lamp module.

[0006] The present application has the following advantages: 1. The ADB module is connected with the environmental detection device to realize intelligent switching of high beam and low beam, avoid glare interference on pedestrians or vehicles at night, and improve safety. 2. The DLP module is connected with the environmental detection device to project visual markers on the road surface and provide night warning. 3. The contour lamp module is provided to form a complete contour after lighting at night, improving the accuracy of the surrounding target in judging the size and motion trajectory of the machine pet. 4. The voice interaction module is used to realize voice control of light parameters, and the user experience is good.

[0007] According to an embodiment of the present application, the ADB module is used to irradiate the road in front of the machine pet and switch between high beam and low beam modes. The DLP module is used to project optical patterns. The contour lamp module is used to display the position and shape of the machine pet. The ADB module and the DLP module are arranged at the front end of the main body of the machine pet. The contour lamp module comprises a plurality of LED lamps, and the plurality of LED lamps are uniformly distributed on the main body of the machine pet and the limbs and legs.

[0008] According to an embodiment of the present application, the control module comprises a processor, and the chip model of the processor is ARMCortex-A53.

[0009] According to an embodiment of the present application, a method for intelligent light control of a robotic pet, the method employs the intelligent light control system of the robotic pet as described above, and comprises the following steps: S1: starting the robotic pet and entering initialization, and controlling the environment detection device to automatically enter a detection state; S2: collecting the distance and speed of the front obstacle by the radar sensor, collecting the light intensity information of the surrounding environment by the light intensity sensor, collecting the surrounding environment information by the visual sensor, collecting the voice instruction information of the user by the voice recognition sensor, and transmitting all the information to the processing device; S3: processing the received information by the processing device, and generating control instructions to control the corresponding light emitting module to perform corresponding control, which includes: generating corresponding control instructions by threshold algorithm based on the light intensity information, and controlling the contour light module; generating corresponding high-low beam switching instructions by the set rule decision algorithm based on the distance and speed of the obstacle and in combination with the light intensity information, and controlling the ADB module; processing the environment information, and generating DLP projection instructions to control the DLP module; processing the voice instruction information, outputting corresponding control instructions and performing voice feedback; S4: the system repeatedly performs steps S2 to S4 every 500 ms, and records system information including the light source mode, projection pattern and flickering state information for subsequent information diagnosis in case of failure.

[0010] According to an embodiment of the present application, generating corresponding control instructions by threshold algorithm based on the light intensity information, and controlling the contour light module to perform corresponding actions includes: if the light intensity > 15 Lux, entering daytime mode, and controlling the module to turn on the contour light module and turn off the ADB module and the DLP module.

[0011] According to an embodiment of the present application, generating corresponding high-low beam switching instructions by the set rule decision algorithm based on the distance and speed of the obstacle and in combination with the light intensity information, and controlling the ADB module to perform corresponding actions includes: if the light intensity ≤ 15 Lux, or the distance of the obstacle ≤ 20 m, or the relative speed > 30 km / h, the control module issues a low beam mode control instruction to control the ADB module to switch to low beam mode, and controls the contour light module to flash red; if the light intensity ≤ 15 Lux and the distance of the obstacle > 20 m and the relative speed ≤ 30 km / h, the control module issues a high beam mode control instruction to control the ADB module to switch to high beam mode.

[0012] According to one embodiment of the present application, the environment information is processed, and DLP projection instructions are generated, and the DLP module control specifically includes: The image processing module obtains environment information, performs target recognition through a deep learning algorithm, obtains image information, and sends the target object and the target object region to the control module. An XOY coordinate system of the robot pet is established, wherein the forward direction of the robot pet is the X axis, the left-right direction of the robot pet is the Y axis, and the vertical ground direction is the Z axis. The control module calculates the three-dimensional coordinates of the target object region in the XOY coordinate system of the robot pet. X b , Y b , Z b ); The XOY coordinate system of the robot pet is converted into a world coordinate system, the actual position of the target object region on the road surface is calculated, and the projection range of the marker pattern is calculated according to the actual position of the target object region on the road surface. The control module sends the projection range of the marker pattern to the DLP module, and the DLP module projects based on the projection range of the marker pattern.

[0013] According to one embodiment of the present application, the target object is a step, a stone block or a pit.

[0014] According to one embodiment of the present application, the voice instruction information includes turning on or off high beam, switching high beam and low beam mode, adjusting light brightness and / or changing optical color.

[0015] According to one embodiment of the present application, the recognition confidence of the target object by the deep learning model is C, C is in the range [0, 1], the trigger threshold is C0, and C0 is in the range [0.7, 0.9], when the confidence C is greater than or equal to C0, the DLP module is started to project the marker.

[0016] Other features and advantages of the present application will be described in the following description, and some will become apparent from the description, or will be understood from the practice of the present application. The purposes and other advantages of the present application are achieved and obtained by the structures specifically pointed out in the specification, claims and drawings.

[0017] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are specifically described below, and the accompanying drawings are described in detail as follows. BRIEF DESCRIPTION OF DRAWINGS

[0018] The present application is further described below in combination with the drawings and embodiments.

[0019] Figure 1is a schematic diagram of the system structure of embodiment one of the present application.

[0020] Figure 2 is a schematic diagram of the environment detection device of embodiment one of the present application installed on a robot dog.

[0021] Figure 3 is a schematic diagram of the method of embodiment two of the present application.

[0022] Fig. 1, environment detection device; 11, radar sensor; 12, illumination sensor; 13, visual sensor; 14, voice recognition sensor; 2, processing device; 21, image processing module; 22, voice processing module; 23, control module; 31, ADB module; 32, DLP module; 33, contour light module. DETAILED DESCRIPTION

[0023] The present application will now be further described in greater detail in connection with the accompanying drawings. These drawings are not necessarily to scale and, in fact, emphasis is placed on the principles of the application rather than on the details thereof. In the drawings:

[0024] In the description of the present application, it needs to be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the features defined as "first" and "second" can be explicitly or implicitly included one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0025] In the description of the present application, it needs to be understood that, unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0026] It should be noted that the robot pet in the present embodiment refers to a machine device that can move on its own in the absence of external instructions based on digital and logical computing devices.

[0027] Embodiment 1 The embodiment of the present application provides a kind of intelligent light control system of machine pet, as shown in Figure 1 System includes: environment detection device 1, processing device 2 and multiple light emitting devices arranged on machine pet, wherein, light emitting device is ADB module 31, DLP module 32 and outline lamp module 33 respectively.

[0028] Environment detection device 1 includes: radar sensor 11, illumination sensor 12, visual sensor 13 and voice recognition sensor 14, radar sensor 11 and visual sensor 13 are arranged at the front end of machine pet trunk main body, and illumination sensor 12 and voice recognition sensor 14 are arranged on machine pet trunk main body.Radar sensor 11 is configured to collect the distance and speed of front obstacles;Illumination sensor 12 is configured to collect the light intensity information of surrounding environment;Visual sensor 13 is configured to collect the environmental information of surrounding;Voice recognition sensor 14 is configured to collect the voice instruction information of user.

[0029] Processing device 2 includes: image processing module 21, voice processing module 22 and control module 23, image processing module 21 is connected with visual sensor 13, voice processing module 22 is connected with voice recognition sensor 14, radar sensor 11, illumination sensor 12, image processing module 21, voice processing module 22 and light emitting module are connected with control module 23, image processing module 21 is configured to obtain the environmental information collected by visual sensor 13, and pre-processes, to obtain image information;Voice processing module 22 is configured to pre-process the collected environmental sound information, to obtain sound instruction;Control module 23 is configured to process the distance, speed, light intensity information, image information and sound instruction of obstacle and generate control instruction, control module 23 executes corresponding control according to control instruction on corresponding light emitting module, and corresponding control includes: ADB module 31 is controlled based on light intensity information through threshold algorithm to generate corresponding control instruction; ADB module 31 is controlled based on the distance and speed of obstacle and combined with light intensity information, through the decision algorithm of set rule, to generate corresponding high-low beam switching instruction; DLP module 32 is controlled based on image information, and DLP projection instruction is generated; Based on voice instruction, output corresponding control instruction and carry out voice feedback.

[0030] In this embodiment, both the ADB module 31 and the DLP module 32 are located at the front end of the robot pet's torso; the contour light module 33 includes several LEDs, which are evenly distributed on the robot pet's torso and limbs. The ADB module 31 is used to illuminate the road in front of the robot pet and switch between high and low beam modes; the DLP module 32 is used to project optical patterns; and the contour light module 33 is used to display the robot pet's position and shape. Specifically, such as Figure 2 As shown, taking a robot dog as an example, the radar sensor 11 is placed at the robot dog's mouth, the illuminance sensor 12 is placed above the robot dog's body, the vision sensor 13 is placed at the robot dog's nose, the voice recognition sensor 14 is placed at the robot dog's chest, the ADB module 31 and the DLP module 32 are respectively placed at the robot dog's eyes, and LED lights are respectively placed at the robot dog's ears, face, body, tail, legs, joints, and foot corners.

[0031] In this embodiment, the control module 23 includes a processor, which is an ARM Cortex-A53 chip, equipped with 4GB of memory, and runs a Linux operating system.

[0032] The illumination intensity range of the light sensor 12 is 0.01-100,000 Lux, and the maximum detection range of the radar sensor 11 is 30 meters. The radar sensor 11 is either a lidar sensor 11 or a millimeter-wave radar sensor 11.

[0033] Example 2 This application provides an intelligent lighting control method for a robotic pet. The method employs the intelligent lighting control system for robotic pets described above. Figure 3 As shown, the method includes the following steps: S1: Start the robot pet and enter initialization. The control environment detection device 1 will automatically enter the detection state. It should be noted that after entering initialization, the outline light module 33 is set to green constant light and the ADB module 31 is set to low beam mode.

[0034] S2: The radar sensor 11 collects the distance and speed of obstacles in front, the illuminance sensor 12 collects the illuminance information of the surrounding environment, the vision sensor 13 collects the surrounding environment information, and the voice recognition sensor 14 collects the user's voice command information, and transmits all the information to the processing device 2.

[0035] S3: Processing device 2 processes the received information and generates control commands to control the corresponding light-emitting module to perform corresponding controls, including: Based on the light intensity information, a threshold algorithm is used to generate corresponding control commands to control the contour light module 33. Based on the distance and speed of the obstacle and combined with the light intensity information, the corresponding high-low light switching instructions are generated by the decision algorithm of the set rules, and the ADB module 31 is controlled.

[0036] The environmental information is processed, and the DLP projection instruction is generated to control the DLP module 32. The voice instruction information is processed, and the corresponding control instruction is output and voice feedback is performed.

[0037] S4: The system repeats steps S2 to S4 every 500ms, and records system information, including light source mode, projection pattern and flashing state information, for subsequent information diagnosis in case of failure.

[0038] In this embodiment, the corresponding control instruction is generated by the threshold algorithm based on the light intensity information, and the contour lamp module 33 is controlled to perform the corresponding action, including: If the light intensity is > 15Lux, enter the daytime mode, control module 23 opens the contour lamp module 33, and closes the ADB module 31 and DLP module 32.

[0039] In this embodiment, based on the distance and speed of the obstacle and combined with the light intensity information, the corresponding high-low light switching instructions are generated by the decision algorithm of the set rules, and the ADB module 31 is controlled to perform the corresponding action, including: If the light intensity is ≤ 15Lux, or the obstacle distance is ≤ 20m, or the relative speed is > 30km / h, the control module 23 issues a low beam mode control instruction, and the ADB module 31 is switched to a low beam mode, and the contour lamp module 33 is controlled to flash red with a flashing frequency of 1HZ.

[0040] If the light intensity is ≤ 15Lux and the obstacle distance is > 20m and the relative speed is ≤ 30km / h, the control module 23 issues a high beam mode control instruction, and the ADB module 31 is switched to a high beam mode.

[0041] In this embodiment, the environmental information is processed, and the DLP projection instruction is generated to control the DLP module 32, including: The image processing module 21 obtains environmental information, performs target recognition through a deep learning algorithm, obtains image information, and sends the target object and target object area to the control module 23.

[0042] The machine pet body coordinate system XOY is established, wherein the machine pet forward direction is the X axis, the machine pet left-right direction is the Y axis, and the vertical ground direction is the Z axis.

[0043] The control module 23 calculates the three-dimensional coordinates of the target object area in the machine pet body coordinate systemX b , Y b , Z b );Specifically comprising: The intrinsic matrix of the visual sensor 13 is obtained by Zhang's calibration method K , which describes the mapping relationship between the pixel coordinates and the visual sensor 13 coordinates, and the calculation formula is:

[0044] Among them, , is the focal length of the visual sensor 13 (unit: pixel), , is the principal point coordinates (image center pixel, unit: pixel).

[0045] Let the translation vector of the visual sensor 13 relative to the machine pet body coordinate system be T c→b = [ t x , t y , t z ] T (unit: m), the translation vector of the visual sensor 13 relative to the machine pet body coordinate system is obtained by measuring the installation position of the visual sensor 13 (example: t x =0.15m, t y =0m, t z =0.2m, that is, the visual sensor 13 is installed at the front of the body center 15cm high 20cm).

[0046] Let the installation attitude angle of the visual sensor 13 relative to the machine pet body be the pitch angle α , the roll angle β, and the yaw angle γ (example: α =-15°, β=0°, γ =0°, that is, the pitch angle of the visual sensor 13 is 15°), which is converted into a rotation matrix R c→b (Z-Y-X Euler angle rotation sequence).

[0047] The machine pet body attitude angle is calculated, specifically: the pitch angle θ , the roll angle φ (horizontal driving θ= φ =0°), and the heading angle ψ(forward direction deflection), for calculating the rotation matrix of the computer pet body relative to the world coordinate system R b→w .

[0048] Target recognition is performed by a deep learning algorithm to obtain the pixel bounding box of the target region, denoted as the upper left corner ( u 1, v 1), the lower right corner ( u 2, v 2), i.e. the pixel range of the target region in the image: u ∈[ u 1, u 2] and v∈[ v 1, v 2].

[0049] Given the depth value of any pixel of the target region in the image d (i.e. the coordinates of the point in the vision sensor 13 coordinate system Z c : Z c = d ), the three-dimensional coordinates of the target region in the camera coordinate system are calculated X c , Y c , Z c )

[0050] Based on the three-dimensional coordinates of the target region in the camera coordinate system X c , Y c , Z c ), the three-dimensional coordinates of the target region in the computer pet body coordinate system are calculated by homogeneous transformation X b , Y b , Z b ), specifically including: through the rotation matrix R c→b and the translation vector T c→b , the three-dimensional coordinates of the target region in the computer pet body coordinate system are calculated X b , Y b , Z b ):

[0051] where the rotation matrix (Z-Y-X Euler angle expansion):

[0052] The rotation matrix of each axis (angle→radian):

[0053]

[0054]

[0055] Convert the machine pet body coordinate system to the world coordinate system, calculate the actual position of the target object area on the road surface, and calculate the projection range of the marking pattern according to the actual position of the target object area on the road surface. Specifically, the position of the target object area on the “real road surface” is projected to the ground ( Z=0 ), which needs to be converted to the world coordinate system, including: calculating the ground projection in the machine pet body coordinate system ( , ), assuming that the target object area on the road surface is =0 , the projection formula (ignoring the machine pet body pitch / roll, simplified as horizontal driving) is:

[0056] .

[0057] The conversion of the machine pet body to the world coordinate system is first rotated around the Z axis (heading angle ψ), and then translated (cumulative displacement Δ X,ΔY ): 。

[0058] Iterate through all the pixels in the target object area, and convert each pixel to get the ground projection world coordinate ( ) ( i=1,2,...,N ), take the extreme value:

[0059] 。

[0060] Set the outward expansion distance Δ = 0.05~0.1m (default 0.08m), and the projection range needs to cover the edge of the target object area outside Δ, the calculation formula is:

[0061] .

[0062] The control module 23 sends the projection range of the marking pattern to the DLP module 32, and the DLP module 32 projects based on the projection range of the marking pattern.

[0063] In this embodiment, the voice instruction information includes turning on or off the high beam, switching the high beam and low beam mode, adjusting the light brightness and / or changing the optical color, the voice processing module 22 processes the voice instruction information, and matches with the matching of the model, obtains the matching degree M, and when the matching degree M>0.8, controls the corresponding light emitting module to execute the corresponding instruction.

[0064] In this embodiment, the target object is a step, a stone or a pit, the recognition confidence of the target object of the deep learning model is C, the value range of C is [0, 1], the trigger threshold is set as C0, the value range of C0 is [0.7, 0.9], when the confidence C≥C0, the DLP module 32 is started to project the mark.

[0065] In this embodiment, when the target object is a pedestrian or a vehicle, the recognition confidence of the target object of the deep learning model is Z, the value range of Z is [0, 1], the trigger threshold is set as Z1, the value range of Z1 is [0.6, 0.8], when the confidence Z≥Z1, the ADB module 31 is controlled to switch to the low beam mode.

[0066] In summary, the beneficial effects of the intelligent light control system of the machine pet and the control method thereof are: 1. The ADB module 31 and the environment detection device 1 are linked to realize intelligent switching of high beam and low beam, avoid glare interference on pedestrians or vehicles at night, and have high safety; 2. The DLP module 32 and the environment detection device 1 are linked to project visual marks on the road surface for night warning; 3. The contour lamp module 33 is set, and after being lit at night, a complete contour is formed, improving the accuracy of the judgment of the target around the machine pet on the body shape and motion trajectory; 4. The voice interaction module is used to realize voice control of the light parameters, and the user has good use feeling.

[0067] Based on the above ideal embodiments according to the present application, through the above description, relevant personnel can make various changes and modifications without deviating from the technical idea of the present application. The technical scope of the present application is not limited to the contents in the specification, and the technical scope must be determined according to the scope of claims.

Claims

1. An intelligent lighting control system for a robotic pet, characterized in that, The system comprises an environment detection device (1), a processing device (2), and a plurality of light emitting devices arranged on the machine pet; The environment detection device (1) comprises: a radar sensor (11) configured to collect the distance and speed of the front obstacle; an illumination sensor (12) configured to collect the light intensity information of the surrounding environment; a visual sensor (13) configured to collect the surrounding environment information; a voice recognition sensor (14) configured to collect the voice instruction information of the user; The processing device (2) comprises: an image processing module (21) connected with the visual sensor (13), the image processing module (21) is configured to obtain the environment information collected by the visual sensor (13) and pre-process it to obtain image information; a voice processing module (22) connected with the voice recognition sensor (14), the voice processing module (22) is configured to pre-process the collected environmental sound information to obtain sound instructions; a control module (23), the radar sensor (11), the illumination sensor (12), the image processing module (21), the voice processing module (22), and the light emitting module are all connected with the control module (23), the control module (23) is configured to process the distance, speed, light intensity information, image information, and sound instructions of the obstacle and generate control instructions, and the control module (23) executes corresponding control on the corresponding light emitting module according to the control instructions; wherein the light emitting devices are respectively ADB module (31), DLP module (32), and outline lamp module (33).

2. The intelligent light control system of the machine pet according to claim 1, wherein: the ADB module (31) is used for illuminating the road in front of the machine pet and switching between high beam and low beam modes; the DLP module (32) is used for projecting optical patterns; the outline lamp module (33) is used for displaying the position and shape of the machine pet; the ADB module (31) and the DLP module (32) are arranged at the front end of the main body of the machine pet; the outline lamp module (33) comprises a plurality of LED lamps, and the plurality of LED lamps are evenly distributed on the main body and limbs of the machine pet.

3. The intelligent light control system of a robotic pet according to claim 1, wherein, The control module (23) comprises a processor, and the chip model of the processor is ARM Cortex-A53. 4.A method for intelligent light control of a robotic pet, characterized by, The method adopts the intelligent light control system of the machine pet according to any one of claims 1 to 3, and the method comprises the following steps: S1: starting the machine pet and entering initialization, and controlling the environment detection device (1) to automatically enter the detection state; S2: collecting the distance and speed of the front obstacle by the radar sensor (11), collecting the light intensity information of the surrounding environment by the illumination sensor (12), collecting the surrounding environment information by the visual sensor (13), collecting the voice instruction information of the user by the voice recognition sensor (14), and transmitting all the information to the processing device (2); S3: The processing device (2) processes the received information and generates control instructions to control the corresponding light-emitting module to perform corresponding control, which includes: Based on the light intensity information, the corresponding control instructions are generated by a threshold algorithm to control the contour lamp module (33); Based on the distance and speed of the obstacle and combined with the light intensity information, the corresponding high-low beam switching instructions are generated by a set of rule-based decision algorithms to control the ADB module (31); Processing environmental information and generating DLP projection instructions to control the DLP module (32); Processing voice instruction information, outputting corresponding control instructions and voice feedback; S4: The system repeats steps S2 to S4 every 500ms and records system information, including light source mode, projection pattern and flashing state information, for subsequent information diagnosis in case of failure.

5. The intelligent light control method of a robotic pet according to claim 4, wherein, Based on the light intensity information, the corresponding control instructions are generated by a threshold algorithm to control the contour lamp module (33) to perform corresponding actions, which include: If the light intensity is > 15Lux, enter daytime mode, control module (23) turns on contour lamp module (33), and turns off ADB module (31) and DLP module (32). 6.The intelligent light control method of a robotic pet according to claim 4, wherein, Based on the distance and speed of the obstacle and combined with the light intensity information, the corresponding high-low beam switching instructions are generated by a set of rule-based decision algorithms to control the ADB module (31) to perform corresponding actions, which include: If the light intensity is ≤ 15Lux, or the obstacle distance is ≤ 20m, or the relative speed is > 30km / h, the control module (23) issues a low beam mode control instruction to control the ADB module (31) to switch to low beam mode, and the contour lamp module (33) flashes red; If the light intensity is ≤ 15Lux and the obstacle distance is > 20m and the relative speed is ≤ 30km / h, the control module (23) issues a high beam mode control instruction to control the ADB module (31) to switch to high beam mode.

7. The intelligent light control method of a robotic pet according to claim 4, wherein, Processing environmental information and generating DLP projection instructions to control the DLP module (32) specifically include: The image processing module (21) acquires environmental information, performs target recognition through a deep learning algorithm, obtains image information, and sends the target object and target object area to the control module (23); An XOY coordinate system of the machine pet is established, wherein the forward direction of the machine pet is the X-axis, the left-right direction of the machine pet is the Y-axis, and the vertical ground direction is the Z-axis; The control module (23) calculates three-dimensional coordinates of the target object region in the machine pet body coordinate system X b , Y b , Z b ) The XOY coordinate system of the machine pet is converted into a world coordinate system, the actual position of the target object area on the road surface is calculated, and the projection range of the marker pattern is calculated according to the actual position of the target object area on the road surface; The control module (23) sends the projection range of the marker pattern to the DLP module (32), and the DLP module (32) projects based on the projection range of the marker pattern.

8. The intelligent light control method of a robotic pet according to claim 7, wherein, The target object is a step, a stone or a pit.

9. The intelligent light control method of a robotic pet according to claim 4, wherein, The voice instruction information includes turning on or off the high beam, switching the high-low beam mode, adjusting the light brightness and / or changing the optical color. 10.The intelligent light control method of a robotic pet of claim 7, wherein, The confidence of the deep learning model in recognizing the target object is C, C is in the range [0, 1], the trigger threshold is C0, in the range [0.7, 0.9], when the confidence C is greater than or equal to C0, the DLP module (32) is started to project the mark.

Images