Pet nursing robot obstacle avoidance method based on visual image and laser radar

Abstract

The invention discloses a pet nursing robot obstacle avoidance method based on a visual image and a laser radar, and the method has the following beneficial effects: the pet nursing robot is provided with an electric telescopic traction rope device for pulling a pet, and the distance measurement between the pet and an obstacle is realized through the laser radar and the visual image; according to the pet nursing robot, the condition around a pet is detected in real time, the electric telescopic traction rope device is further controlled according to the distance, the problems that the pet collides with an obstacle and the like are well solved, and therefore the safety of the pet and pedestrians is better guaranteed, and the nursing effect of the pet nursing robot is better achieved; in addition, the multi-scene pet care device is provided with the replaceable indoor chassis and the replaceable outdoor chassis, and therefore the requirement for intelligent pet care in multiple indoor and outdoor scenes can be met.

Description

Technical Field

[0001] The present invention relates to the technical field of pet-care robots, and in particular to an obstacle avoidance method for a pet-care robot based on visual images and laser radar. Background Art

[0002] With the development of the social economy, people's living standards have generally improved, and they have more financial resources to cover the various expenses of pets. Furthermore, with the increase in single-person households and DINK (dual-income, no children) families in modern family structures, as well as the arrival of an aging society, more people are choosing pets for emotional comfort and companionship. With the accelerated pace of urban life and increasing work and life pressures, many people are turning to pets to relieve stress. Pets have become an important member of many families, and people are paying more and more attention to their pets' health and well-being. Therefore, the demand for high-quality pet services is growing. In response to this, pet care robots have emerged, which can provide pet services such as feeding and feces cleaning.

[0003] Furthermore, for urban pet owners, the importance of walking pets cannot be underestimated. It impacts multiple aspects of a pet's physical health, psychological state, and social behavior. This is also the paradox of modern urban pet owners lacking the time to walk their pets. Regularly walking pets strengthens their cardiopulmonary function, increases their metabolic rate, and helps maintain a healthy weight and body shape. Exercise also enhances joint flexibility, slowing joint degeneration in elderly pets. Walking pets exposes them to more external stimuli, distracting them and reducing the anxiety and stress that can arise from prolonged confinement in the same environment.

[0004] However, existing pet-care robots cannot meet the outdoor travel needs of pets well, including the difficulty of pet-care robots in controlling the distance between pets and obstacles such as pedestrians and other objects, which may cause collisions between pets and obstacles, affecting the safety of pets and pedestrians. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a pet care robot obstacle avoidance method based on visual images and laser radar, which can solve the above technical problems.

[0007] (2) Technical solution

[0008] To solve the above technical problems, the present invention provides the following technical solutions: a pet-care robot obstacle avoidance method based on visual images and laser radar, the pet-care robot comprising a robot head and a robot chassis, the robot head being provided with a camera device, the robot chassis being provided with a laser radar and an electric retractable traction rope device, wherein the electric retractable traction rope device is used to pull the pet;

[0009] The obstacle avoidance method for a pet care robot based on visual images and lidar includes the following steps:

[0010] S1, build environment map through lidar;

[0011] S2. capturing a visual image of the pet using a camera;

[0012] S3, performing target recognition on the pet visual image to identify the pet;

[0013] S4, obtaining the distance between the pet and obstacles in the environment map;

[0014] S5. Control the electric retractable traction rope device according to the distance.

[0015] Preferably, in step S1, sensor data is acquired by a laser radar, and an environment map is constructed using the sensor data by a Gmapping algorithm.

[0016] Preferably, in step S3, target recognition is performed on the pet visual image using a YOLO algorithm to identify the pet.

[0017] Preferably, in step S5, it is determined whether the distance is greater than a first distance threshold, and if so, the electric retractable traction rope device is controlled to continuously extend and release the traction rope.

[0018] Preferably, in step S5, it is determined whether the distance is greater than a second distance threshold and less than a first distance threshold. If so, the electric retractable traction rope device is controlled to intermittently extend and release the traction rope, wherein the second distance threshold is less than the first distance threshold.

[0019] Preferably, in step S5, it is determined whether the distance is less than a third distance threshold, and if so, the electric retractable traction rope device is controlled to retract the traction rope, wherein the third distance threshold is less than the second distance threshold.

[0020] Preferably, the electric retractable traction rope device includes a motor, a coil spring and a traction rope, the motor is connected to the coil spring, one end of the traction rope is connected to the coil spring, and the other end of the traction rope is used to pull the pet.

[0021] Preferably, after step S3, the method further includes: further performing a target behavior analysis on the pet to obtain a behavior analysis result, and further controlling the electric retractable traction rope device according to the behavior analysis result.

[0022] Preferably, the robot chassis includes an indoor chassis and an outdoor chassis, the robot head is detachably arranged on the robot chassis, and the outdoor chassis is provided with a laser radar and an electric retractable traction rope device.

[0023] Preferably, the indoor chassis is provided with an airbag, a gyroscope and a control device. The gyroscope detects the body position data of the indoor chassis, and the control device determines whether the body position data exceeds a preset body position critical value. If so, the control device controls the airbag to start.

[0024] (3) Beneficial effects

[0025] Compared with the prior art, the present invention provides an obstacle avoidance method for a pet-care robot based on visual images and laser radar, which has the following beneficial effects: the pet-care robot of the present invention is provided with an electric retractable traction rope device for pulling the pet, and the distance between the pet and the obstacle is measured by laser radar and visual images, that is, the situation around the pet is detected in real time, and the electric retractable traction rope device is further controlled according to the distance, so as to better avoid problems such as collisions between the pet and obstacles, thereby better ensuring the safety of pets and pedestrians, and better achieving the care effect of the pet-care robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flowchart of the steps of a pet care robot obstacle avoidance method based on visual images and laser radar of the present invention;

[0027] Figure 2 This is a functional block diagram of the pet-care robot (outdoor chassis) of the present invention;

[0028] Figure 3 A structural perspective diagram of the pet-care robot (indoor chassis) of the present invention;

[0029] Figure 4 A structural perspective diagram of the feces cleaning mechanism of the present invention;

[0030] Figure 5 This is a principle block diagram of the electric telescopic traction rope device of the present invention.

[0031] The numbers in the figure are: 1 robot head, 2 robot chassis, 11 camera device, 21 laser radar, 22 electric telescopic traction rope device, 221 motor, 222 coil spring, 223 traction rope, 23 feces cleaning mechanism, 231 robot arm pan / tilt platform, 232 clamping device, 233 steering axis, 234 robot arm growth axis. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0033] The present invention provides an obstacle avoidance method for a pet-care robot based on visual images and laser radar. The pet-care robot of the present invention includes a robot head 1 and a robot chassis 2. The robot head 1 is provided with a camera 11, and the robot chassis 2 is provided with a laser radar 21 and an electric retractable traction rope device 22, wherein the electric retractable traction rope device 22 is used to pull a pet.

[0034] The obstacle avoidance method of a pet-care robot based on visual images and laser radar of the present invention comprises the following steps:

[0035] S1. Build an environmental map using LiDAR.

[0036] It is understood that the environmental map is a map of the environment in which the pet-care robot is located. The pet-care robot of the present invention can be used in outdoor environments, etc. The environmental map includes the location information of pets and various obstacles such as pedestrians, buildings, and vehicles. Preferably, in step S1, sensor data is obtained by a laser radar, and the environmental map is constructed using the sensor data through the Gmapping algorithm. The Gmapping algorithm is a SLAM (Simultaneous Localization and Mapping) algorithm based on a particle filter, and the sensor data of the laser radar plays a vital role in the SLAM algorithm system by providing accurate and high-resolution environmental data. The laser radar sensor emits a laser pulse, which bounces off the object and returns to the sensor, enabling it to measure the distance to the object based on the time required for the laser pulse to return. The laser radar sensor can generate a dense point cloud, which is a set of points in 3D space that represents the surface of an object in the environment. The point cloud data generated by the laser radar sensor can be used to identify and track landmarks, such as walls, corners, or other unique features in the environment. These landmarks are crucial to the SLAM algorithm because the information they provide can be used to refine the robot's estimated position and orientation. In addition, lidar data can also be used to identify and avoid obstacles. The above-mentioned Gmapping algorithm is an implementation of the Rao-Blackwell lized particle filter (RBPF) for lidar-based SLAM. The algorithm uses a grid map representation and uses scan matching to align the lidar scan with the map. Gmapping can provide accurate mapping and positioning results in small and large environments. In other embodiments, the environment map can also be constructed by using lidar in conjunction with other map construction and positioning algorithms in the existing technology, and there are no excessive restrictions here.

[0037] S2. Capture the pet's visual image through a camera device.

[0038] Preferably, the camera device is arranged to be rotatable so as to facilitate tracking and photographing the pet; multiple camera devices can be provided.

[0039] S3. Perform target recognition on the pet visual image to identify the pet.

[0040] The pet visual image captured by the camera device may include information about the pet and its surrounding environment. In step S3, objects such as obstacles in the pet visual image may also be identified.

[0041] Preferably, in step S3, object recognition is performed on the pet visual image using a YOLO algorithm, such as YOLOv8, to identify the pet. YOLO (You Only Look Once) is a deep learning-based object detection algorithm whose core concept is to transform object detection into a regression problem, simultaneously predicting bounding box positions and class probabilities through a single forward pass. In other embodiments, other object detection algorithms may also be used to perform object recognition on the pet visual image, without further limitation herein.

[0042] S4. Obtain the distance between the pet and obstacles in the environment map.

[0043] It can be understood that after the pet is identified, the distance between the pet and various obstacles such as pedestrians, buildings, and vehicles can be calculated based on the position information of the pet and various obstacles such as pedestrians, buildings, and vehicles in the environmental map; preferably, after the pet is identified, the pet can be marked accordingly in the environmental map to facilitate the acquisition of the above distance.

[0044] S5. Control the electric retractable traction rope device according to the distance.

[0045] Specifically, in step S5, it can be determined whether the distance between the pet and the obstacle obtained in the above step S4 is greater than the first distance threshold. If so, the electric retractable traction rope device is controlled to continuously extend and release the traction rope, which indicates that the pet is in an obstacle-free area at this time, so the electric retractable traction rope device is controlled to freely and continuously extend and release the traction rope.

[0046] In step S5, it is also possible to determine whether the distance between the pet and the obstacle obtained in the above step S4 is greater than the second distance threshold and less than the first distance threshold. If so, the electric retractable traction rope device is controlled to intermittently extend and release the traction rope, which indicates that the pet is at a relatively close distance to the obstacle at this time. Therefore, the electric retractable traction rope device is controlled to intermittently extend and release the traction rope, that is, to adopt a traction method of intermittent pauses and extensions to avoid the pet suddenly getting too close to the obstacle; wherein, the second distance threshold is less than the above first distance threshold.

[0047] In addition, in step S5, it is also possible to determine whether the distance between the pet and the obstacle obtained in the above step S4 is less than a third distance threshold. If so, the electric retractable traction rope device is controlled to retract the traction rope, which indicates that the pet is at a very close distance to the obstacle at this time and is prone to collision. Therefore, the electric retractable traction rope device is controlled to retract to avoid collision between the pet and the obstacle; wherein the third distance threshold is less than the above second distance threshold.

[0048] Preferably, the electric retractable traction rope device 22 includes a motor 221, a coil spring 222 and a traction rope 223. The motor 221 is connected to the coil spring 222, one end of the traction rope 223 is connected to the coil spring 222, and the other end of the traction rope 223 is used to pull the pet. It can be understood that the traction rope 223 can be extended or retracted through the coil spring 222 by controlling the motor 221 to rotate forward or reverse; the motor 221 can specifically be a stepper motor, the coil spring 222 can specifically be elastic steel, and the traction rope 223 can specifically be made of thermoplastic rubber material (TPR) or polyester, etc.

[0049] Preferably, after the above step S3, the method may further include: further performing a target behavior analysis on the pet to obtain a behavior analysis result; specifically, the pet's action behavior in the pet's visual image may be analyzed using the existing convolutional network, HRNet and other action posture recognition algorithms, which are not restricted here; and further controlling the electric retractable traction rope device according to the behavior analysis result: for example, when the behavior analysis result is that the pet is in a head-down posture, indicating that the pet may be in a relatively tired state at this time, the electric retractable traction rope device is correspondingly controlled to stop extending and releasing the traction rope; conversely, when the behavior analysis result is that the pet is in a head-up posture, indicating that the pet may be in a relatively active state at this time, the electric retractable traction rope device is correspondingly controlled to speed up the extension and release of the traction rope.

[0050] In addition, the robot chassis of the present invention may include an indoor chassis and an outdoor chassis. The indoor chassis is suitable for indoor environments, while the outdoor chassis is suitable for outdoor environments. The robot head is detachably mounted on the robot chassis to facilitate replacement of different robot chassis. The outdoor chassis is equipped with the aforementioned laser radar and an electrically retractable traction rope device. Preferably, the outdoor chassis's running wheel system includes planetary wheels and Mecanum wheels. Planetary wheels are highly adaptable obstacle-crossing mechanisms that can provide stable mobility support in complex terrain. The use of Mecanum wheels, based on the planetary wheel system, further greatly improves the maneuverability and flexibility of the outdoor mode. In this way, the outdoor chassis of the present invention can better adapt to walking in different environmental terrains.

[0051] Further preferably, the above-mentioned indoor chassis is provided with an airbag, a gyroscope and a control device. The posture data of the indoor chassis is detected by the gyroscope, and the control device determines whether the posture data exceeds a preset posture critical value. If so, it indicates that the pet-care robot has fallen over due to impact by external factors such as pets. At this time, the control device controls the airbag to start accordingly to protect the pet-care robot.

[0052] In addition, the indoor chassis can also be provided with a feces cleaning mechanism 23 for grabbing pet excrement. Preferably, the feces cleaning mechanism 23 adopts a humanoid flexible robotic arm, which can be moved up and down, left and right on the indoor chassis. The humanoid flexible robotic arm includes a robotic arm platform 231, a clamping device 232, a number of motors, a number of steering shafts 233 and a number of robotic arm growth shafts 234. The robotic arm platform is provided on the indoor chassis, the robotic arm platform 231 is connected to the steering shaft 233 through a motor, the robotic arm growth shaft 234 is connected to the steering shaft 233 through a motor, and the clamping device 232 is provided on the indoor chassis. 32 is set at the tail end of the humanoid flexible robotic arm, and the clamping device 232 preferably adopts a shovel-shaped clamper. The motor is used to drive the steering shaft 233 or the robotic arm growth shaft 234 to rotate to drive the clamping device 232 to rotate; because human manual handling can achieve a good effect on cleaning pet feces, through the above method, the humanoid flexible robotic arm can better imitate human arm movements and realize multi-degree-of-freedom movement, so that the humanoid flexible robotic arm is more precise in performing tasks, that is, it can better enable the clamping device 232 to move to the location of feces to be cleaned for clamping.

[0053] Compared with the prior art, the present invention provides an obstacle avoidance method for a pet-care robot based on visual images and laser radar, which has the following beneficial effects: the pet-care robot of the present invention is provided with an electric retractable traction rope device for pulling the pet, and the distance between the pet and the obstacle is achieved through the laser radar and visual images, that is, the situation around the pet is detected in real time, and the electric retractable traction rope device is further controlled according to the distance, so as to better avoid problems such as collision between the pet and the obstacle, thereby better protecting the safety of the pet and pedestrians, and better achieving the care effect of the pet-care robot; in addition, the present invention is provided with a replaceable indoor chassis and outdoor chassis dual-scene chassis, which can meet the intelligent care needs of pets in multiple indoor and outdoor scenes.

[0054] It should be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not preclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0055] While 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 these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A pet care robot obstacle avoidance method based on visual images and laser radar, characterized by: The pet care robot includes a robot head and a robot chassis, wherein the robot head is provided with a camera device, and the robot chassis is provided with a laser radar and an electric retractable traction rope device, wherein the electric retractable traction rope device is used to pull the pet; The pet care robot obstacle avoidance method based on visual images and laser radar includes the following steps: S1, constructing an environment map using the laser radar; S2, capturing a visual image of the pet by the camera device; S3, performing target recognition on the pet visual image to identify the pet; S4. Obtaining the distance between the pet and the obstacle in the environment map; S5. Control the electric retractable traction rope device according to the distance.

2. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 1, characterized in that: In step S1, sensor data is acquired by the laser radar, and the environment map is constructed using the sensor data through the Gmapping algorithm.

3. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 1, characterized in that: In step S3, target recognition is performed on the pet visual image using the YOLO algorithm to identify the pet.

4. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 1, characterized in that: In step S5, it is determined whether the distance is greater than a first distance threshold, and if so, the electric retractable traction rope device is controlled to continuously extend and release the traction rope.

5. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 4, characterized in that: In step S5, it is determined whether the distance is greater than a second distance threshold and less than the first distance threshold. If so, the electric retractable traction rope device is controlled to intermittently extend and release the traction rope, wherein the second distance threshold is less than the first distance threshold.

6. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 5, characterized in that: In step S5, it is determined whether the distance is less than a third distance threshold, and if so, the electric retractable traction rope device is controlled to retract the traction rope, wherein the third distance threshold is less than the second distance threshold.

7. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 1, characterized in that: The electric retractable traction rope device includes a motor, a coil spring and a traction rope, the motor is connected to the coil spring, one end of the traction rope is connected to the coil spring, and the other end of the traction rope is used for pulling a pet.

8. The obstacle avoidance method for a pet-care robot based on visual images and laser radar according to claim 1, characterized in that: After step S3, the method further includes: further performing a target behavior analysis on the pet to obtain a behavior analysis result, and further controlling the electric retractable traction rope device according to the behavior analysis result.

9. The pet-care robot obstacle avoidance method based on visual images and laser radar according to claim 1, characterized in that: The robot chassis includes an indoor chassis and an outdoor chassis. The robot head is detachably arranged on the robot chassis. The outdoor chassis is provided with the laser radar and the electric retractable traction rope device.

10. The pet-care robot obstacle avoidance method based on visual images and laser radar according to claim 9, characterized in that: The indoor chassis is provided with an airbag, a gyroscope and a control device. The gyroscope detects the body position data of the indoor chassis, and the control device determines whether the body position data exceeds a preset body position critical value. If so, the control device controls the airbag to be activated.

Images