A navigation obstacle avoidance intelligent control system and method applied to golf bag car

Abstract

This invention relates to the field of navigation control, specifically to an intelligent navigation and obstacle avoidance control system and method for golf carts. The system includes: a camera platform module, an obstacle recognition module, a traffic planning module, an attitude positioning module, and a speed control module. The camera platform module is used to mount dual-lens cameras. The obstacle recognition module is used to identify obstructions and determine their type. The traffic planning module is used to plan the optimal travel direction for the golf cart. The attitude positioning module is used to output the current coordinates of the golf cart and correct any travel deviations. The speed control module is used to control the speed of the golf cart in real time. This invention can improve the accuracy of obstacle avoidance decisions, achieve more complex collaborative obstacle avoidance and adaptive learning functions, reduce vehicle and golf course facility maintenance costs, improve the overall operational efficiency of the golf course, reduce accident risks, and achieve more intelligent golf course traffic management.

Description

Technical Field

[0001] This invention relates to the field of navigation control, specifically to a navigation and obstacle avoidance intelligent control system and method applied to golf bag carts. Background Technology

[0002] A golf bag cart is a small electric vehicle specifically designed for use on golf courses. It is mainly used to help golfers conveniently carry golf bags, clubs, and other equipment on the course. When a user needs equipment, the golf bag cart can sense and follow the user, delivering the golf equipment to the golfer and providing a better sporting experience.

[0003] Golf courses have complex terrain with numerous obstacles such as people, equipment, balls, and tee pegs, all of which are constantly moving. This frequently obstructs the routes of golf carts, often requiring them to stop and wait until the obstacles disappear, preventing them from reaching users in a timely manner. Some golf carts have radar or infrared capabilities, allowing them to detect obstacles and automatically navigate around them; however, the electronic equipment is expensive and susceptible to interference, making it difficult to navigate in areas with high foot traffic and unsuitable for complex course environments.

[0004] In addition, the braking effect of the ball bag car is different and the driving speed is difficult to control. When the obstacle is moving irregularly, the ball bag car is prone to hitting the obstacle at a high speed, causing equipment damage and personnel injury. Furthermore, when selecting a path, it is necessary to detour a long distance around the moved obstacle, which reduces the ball bag car's addressing efficiency and obstacle avoidance flexibility. Summary of the Invention

[0005] The purpose of this invention is to provide a navigation and obstacle avoidance intelligent control system and method for golf carts, in order to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a navigation and obstacle avoidance intelligent control system for golf carts, comprising: a camera platform module, an obstacle recognition module, a traffic planning module, an attitude positioning module, and a speed control module;

[0007] The camera platform module is used to set up a platform on the top of the ball bag car. The front edge of the platform is perpendicular to the direction of travel of the ball bag car and is equipped with a scale. Two cameras are installed on the platform, so that the main optical axis vector of the two cameras is parallel to the direction of travel of the ball bag car, and the line connecting the two cameras is parallel to the front edge of the platform, so that at least one scale unit of the scale is within the shooting range of the two cameras.

[0008] The obstacle recognition module is used to control the camera to capture images along the direction of travel of the ball bag car. The first image and the second image are obtained from the two cameras respectively. The occlusion part in the image is separated by the image segmentation algorithm. Taking the current position of the ball bag car as the origin, the distance of each point of the occlusion is measured according to the length of the ruler captured in the image and the installation distance of the camera, and the spatial coordinates of the point are given. All the coordinates of the points constitute the occlusion model. The occlusion model is sent to the sample database to identify the type of occlusion.

[0009] The traffic planning module is used to determine the direction of travel of the ball bag vehicle based on the user's satellite positioning coordinates, identify impassable obstacles in a semi-circular area centered on the direction of travel vector, filter the space between the edges of the obstacles whose length is greater than the width of the ball bag vehicle, generate passable sectors, verify all passable sectors, select the sector that is closest to the user's positioning after the ball bag vehicle passes through as the planning sector, and control the ball bag vehicle to move forward with the center of the sector as the direction of travel.

[0010] The attitude positioning module is used to establish a positioning model based on the ball bag car's forward direction and real-time speed, input time parameters into the model, output the ball bag car's current coordinates, correct the travel deviation, and measure the distance again with the current coordinates as the center at fixed intervals to update the distance between the ball bag car and each obstacle.

[0011] The speed control module is used to control the car's speed in real time based on the sector radius, the distance between the ball car and the obstacle, and the ball car's braking parameters. When the sector radius is smaller than the passage radius during the forward movement, braking is performed; when the deviation exceeds the braking torque, deceleration is performed; and the forward direction and speed of the ball car are replanned.

[0012] Furthermore, the camera platform module includes: an edge positioning unit and a dual-camera imaging unit;

[0013] The edge positioning unit is used to build a platform on the top of the ball bag car. The platform is parallel to the ground and has a scale on the front edge.

[0014] The dual-camera imaging unit is used to fix two cameras on the platform and use computer vision and binocular moment measurement principles to detect obstacles.

[0015] Furthermore, the obstacle recognition module includes: an image segmentation unit and a sample comparison unit;

[0016] The image segmentation unit is used to determine occlusions in the captured image using image segmentation algorithms, including K-means, FCN, DeepLab, R-CNN, and U-Net algorithms.

[0017] The sample comparison unit is used to compare the features of the obstruction with the features in the sample library. If the comparison result is an impassable or unrecognizable sample, it is judged as an obstacle.

[0018] Furthermore, the traffic planning module includes: a signal transmission unit, a sector identification unit, and a path selection unit;

[0019] The signal transmission unit is used to receive the user's location information and determine the direction and distance between the user and the ball bag vehicle;

[0020] The sector identification unit is used to calculate the distance between the ball bag car and each obstacle, and to identify the passable sectors of the ball bag car;

[0021] The path selection unit is used to filter passable sectors and plan the shortest route.

[0022] Furthermore, the attitude localization module includes: an attitude modeling unit and a parameter localization unit;

[0023] The attitude modeling unit is used to establish the attitude positioning model of the ball bag vehicle and return the current coordinates of the ball bag vehicle in real time.

[0024] The parameter positioning unit is used to correct the travel deviation, update the obstacle position, and determine the distance between the obstacle and the ball bag car.

[0025] Furthermore, the speed control module includes: a speed planning unit, a dynamic feedback unit, and a vehicle braking unit;

[0026] The speed planning unit is used to determine the speed of the golf bag car based on the minimum distance between the golf bag car and the obstacle and the vehicle braking distance.

[0027] The dynamic feedback unit is used to update the obstacle position and re-plan the ball bag car's speed and direction of travel;

[0028] The vehicle braking unit is used to apply emergency braking when the distance between the vehicle and an obstacle is detected to be less than a threshold, and to assist in vehicle deceleration.

[0029] A navigation and obstacle avoidance intelligent control method for golf bag vehicles includes the following steps:

[0030] Step S1. Set up a platform on the top of the ball bag car. The front of the platform is marked with a scale. Two cameras are installed in parallel on the platform so that the main optical axis of the cameras is consistent with the direction of travel of the ball bag car, and at least one scale unit of the scale is within the shooting range of the camera.

[0031] Step S2. Two cameras capture images of the front of the ball bag car respectively. The image segmentation algorithm is used to separate the occluded parts in the image. Based on the length of the ruler in the image and the shooting range of the camera, the distance of the occluded object is measured using the binocular imaging principle to obtain the coordinates of each point on the edge of the occluded object.

[0032] Step S3. Input the coordinates and grayscale of each point on the edge of the obstruction into the sample database. Mark the obstructions that are identified as impassable or unrecognizable as obstacles. Filter the space between adjacent obstacle edges whose length is greater than the width of the ball bag car to generate passable sectors.

[0033] Step S4. Determine the speed of the ball bag car based on the distance between the ball bag car and the nearest obstacle. Select the passable sector with the highest effective speed as the planning sector and take the direction of the angle bisector of the ball in the planning sector as the direction of the ball bag car's movement.

[0034] Step S5. Determine the current coordinates of the ball bag car based on its forward direction and real-time speed, correct the travel deviation in real time, update the obstacle position, and apply emergency braking when the distance between the obstacle and the vehicle is detected to be below a threshold, and replan the forward direction.

[0035] Furthermore, step S1 includes:

[0036] Step S11. Set up a platform on the top of the ball bag car, with the platform surface parallel to the ground. Install two cameras parallel to each other on the platform, so that the main optical axis of the cameras is consistent with the direction of travel of the ball bag car, and the distance between the two cameras and the center line of the platform is equal. The platform is made of transparent material so as not to block the shooting aperture of the cameras.

[0037] Step S12. Set a scale on the front edge of the platform, with the zero mark of the scale aligned with the center of the line connecting the two cameras, so that at least one scale unit on the scale is within the shooting range of the two cameras. Set a visual processing chip under the platform to receive images captured by the cameras and upload them to the information platform for processing.

[0038] Furthermore, step S2 includes:

[0039] Step S21. Two cameras capture images of the front of the ball bag car, obtaining a first image and a second image respectively. The image segmentation algorithm is used to determine the occluders in the captured images. The image segmentation algorithm includes: K-means, FCN, DeepLab, R-CNN and U-Net algorithms.

[0040] Step S22: Based on the length of the ruler in the image and the camera's shooting range, calculate the distance between each point on the edge of the obstruction and the ball bag workshop using the binocular imaging principle:

[0041]

[0042] Where z1 represents the horizontal distance between the point and the ball bag workshop, x1 and x2 represent the x-coordinates of the point in the first image and the second image respectively, f represents the focal length of the camera, B represents the distance between the two cameras, a represents the compression ratio of the ruler, a = L / A, where L is the scale of the ruler in the image, and A is the pixel width of the ruler.

[0043] Since the two cameras are at the same height, y1 = y2, and the vertical distance between the point and the ball bag workshop is z2 = a·y1-h, where y1 and y2 represent the ordinates of the point in the first and second images, respectively.

[0044] Step S23. Measure the distance to each point on the edge of the obstruction to determine the horizontal and vertical distances from the ball bag workshop. Establish a coordinate system with the ball bag car as the zero point to obtain the coordinates of each point on the edge of the obstruction.

[0045] Furthermore, step S3 includes:

[0046] Step S31. Input the coordinates and grayscale of each point on the edge of the occluder into the sample database. Compare the existing obstacle samples in the sample database. If the comparison result is an impassable or unrecognizable sample, then the occluder is judged as an obstacle.

[0047] Step S32. Based on the user's satellite positioning coordinates, identify impassable obstacles within a semi-circular area centered on the direction of travel, where the direction of travel is from the ball-car coordinates to the user coordinates. Filter out spaces where the width between obstacles is greater than the width of the ball-car. Using the ball-car coordinates as vertices and the lines connecting the edges of the obstacles as chords, generate passable sectors.

[0048] Furthermore, step S4 includes:

[0049] Step S41. Determine the speed of the ball bag car based on the distance between the ball bag car and the nearest obstacle:

[0050]

[0051] Where v represents the speed of the ball-cart, v max The maximum speed of the ball bag car is represented by T, the distance between the ball bag car and the obstacle is represented by u, the braking deceleration of the ball bag car is represented by T0, the reserved deceleration length is represented by v0, and the preset speed reduction is represented by v0.

[0052] Step S42. Calculate the effective travel speed VR of each passable sector, where VR = v·cosθ, and θ represents the angle between the sector angle bisector and the travel direction. Select the passable sector with the largest effective travel speed as the planning sector, and take the direction of the angle bisector of the planning sector as the forward direction of the ball-cart.

[0053] Furthermore, step S5 includes:

[0054] Step S51. Analyze the drive signal of the ball bag car. Based on the coordinates of the ball bag car before its movement, the direction of movement of the ball bag car, and the real-time speed, update the coordinates of the ball bag car and the position of the obstacles in the planned sector in real time, and correct the offset between the direction of movement of the ball bag car and the direction of the angle bisector of the ball in the planned sector.

[0055] Step S52. When the width between obstacles in the planned sector is less than the width of the ball bag car, stop the movement of the ball bag car, the camera takes another picture, identifies the passable sectors around the ball bag car at the current position, and replans the forward direction of the ball bag car. At the same time, when the distance between the obstacle and the vehicle is detected to be lower than the threshold, apply emergency braking.

[0056] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0057] 1. This invention identifies obstacle locations by installing dual cameras on the top of a ball bag vehicle. It employs an image segmentation algorithm to segment occlusions from images, and inputs the occlusion features from the two images into a sample database to identify obstacle types in order to determine whether the ball bag vehicle can pass. This can improve the accuracy of obstacle avoidance decisions and enable more complex collaborative obstacle avoidance and adaptive learning functions.

[0058] 2. This invention can measure the distance to obstacles in the direction of the golf cart's movement. Using the current position of the cart pointing to the destination coordinates as the direction of movement, it determines all passable sectors in the direction of the cart's movement based on the obstacle distance measurement results, selects the passable sector with the smallest deviation angle to proceed, thereby intelligently planning the shortest route, reducing detour time, shortening the time it takes for the golf cart to reach the user's location, reducing the maintenance costs of vehicles and golf course facilities, and improving the overall operational efficiency of the golf course.

[0059] 3. This invention can establish an attitude positioning model based on the forward direction and speed of the ball bag car, and control the car's speed in real time according to the sector radius, the distance between the vehicle and the obstacle, and the ball bag car's braking parameters. During the forward movement, the car brakes according to the position of the obstacle, leaving braking distance for the ball bag car, avoiding collisions with moving obstacles, reducing the risk of accidents, and achieving more intelligent stadium traffic management. Attached Figure Description

[0060] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0061] Figure 1 This is a schematic diagram of the intelligent navigation and obstacle avoidance control system for a golf bag cart according to the present invention;

[0062] Figure 2 This is a schematic diagram illustrating the steps of a navigation and obstacle avoidance intelligent control method for golf bag carts according to the present invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] Please see Figure 1 The present invention provides a technical solution: a navigation and obstacle avoidance intelligent control system for golf carts, comprising: a camera platform module, an obstacle recognition module, a traffic planning module, an attitude positioning module, and a speed control module;

[0065] The camera platform module is used to set up a platform on the top of the ball bag car. The front edge of the platform is perpendicular to the direction of travel of the ball bag car and is equipped with a scale. Two cameras are installed on the platform, so that the main optical axis vector of the two cameras is parallel to the direction of travel of the ball bag car, and the line connecting the two cameras is parallel to the front edge of the platform, so that at least one scale unit of the scale is within the shooting range of the two cameras.

[0066] The camera platform module includes: an edge positioning unit and a dual-camera imaging unit;

[0067] The edge positioning unit is used to build a platform on the top of the ball bag car. The platform is parallel to the ground and has a scale on the front edge.

[0068] The dual-camera imaging unit is used to fix two cameras on the platform and use computer vision and binocular moment measurement principles to detect obstacles.

[0069] The obstacle recognition module is used to control the camera to capture images along the direction of travel of the ball bag car. The first image and the second image are obtained from the two cameras respectively. The occlusion part in the image is separated by the image segmentation algorithm. Taking the current position of the ball bag car as the origin, the distance of each point of the occlusion is measured according to the length of the ruler captured in the image and the installation distance of the camera, and the spatial coordinates of the point are given. All the coordinates of the points constitute the occlusion model. The occlusion model is sent to the sample database to identify the type of occlusion.

[0070] The obstacle recognition module includes: an image segmentation unit and a sample comparison unit;

[0071] The image segmentation unit is used to determine occlusions in the captured image using image segmentation algorithms, including K-means, FCN, DeepLab, R-CNN, and U-Net algorithms.

[0072] The sample comparison unit is used to compare the features of the obstruction with the features in the sample library. If the comparison result is an impassable or unrecognizable sample, it is judged as an obstacle.

[0073] The traffic planning module is used to determine the direction of travel of the ball bag vehicle based on the user's satellite positioning coordinates, identify impassable obstacles in a semi-circular area centered on the direction of travel vector, filter the space between the edges of the obstacles whose length is greater than the width of the ball bag vehicle, generate passable sectors, verify all passable sectors, select the sector that is closest to the user's positioning after the ball bag vehicle passes through as the planning sector, and control the ball bag vehicle to move forward with the center of the sector as the direction of travel.

[0074] The traffic planning module includes: a signal transmission unit, a sector identification unit, and a path selection unit;

[0075] The signal transmission unit is used to receive the user's location information and determine the direction and distance between the user and the ball bag vehicle;

[0076] The sector identification unit is used to calculate the distance between the ball bag car and each obstacle, and to identify the passable sectors of the ball bag car;

[0077] The path selection unit is used to filter passable sectors and plan the shortest route.

[0078] The attitude positioning module is used to establish a positioning model based on the ball bag car's forward direction and real-time speed, input time parameters into the model, output the ball bag car's current coordinates, correct the travel deviation, and measure the distance again with the current coordinates as the center at fixed intervals to update the distance between the ball bag car and each obstacle.

[0079] The attitude localization module includes: an attitude modeling unit and a parameter localization unit;

[0080] The attitude modeling unit is used to establish the attitude positioning model of the ball bag vehicle and return the current coordinates of the ball bag vehicle in real time.

[0081] The parameter positioning unit is used to correct the travel deviation, update the obstacle position, and determine the distance between the obstacle and the ball bag car.

[0082] The speed control module is used to control the car's speed in real time based on the sector radius, the distance between the ball car and the obstacle, and the ball car's braking parameters. When the sector radius is smaller than the passage radius during the forward movement, braking is performed; when the deviation exceeds the braking torque, deceleration is performed; and the forward direction and speed of the ball car are replanned.

[0083] The speed control module includes: a speed planning unit, a dynamic feedback unit, and a vehicle braking unit;

[0084] The speed planning unit is used to determine the speed of the golf bag car based on the minimum distance between the golf bag car and the obstacle and the vehicle braking distance.

[0085] The dynamic feedback unit is used to update the obstacle position and re-plan the ball bag car's speed and direction of travel;

[0086] The vehicle braking unit is used to apply emergency braking when the distance between the vehicle and an obstacle is detected to be less than a threshold, and to assist in vehicle deceleration.

[0087] like Figure 2 As shown, a navigation and obstacle avoidance intelligent control method applied to golf carts includes the following steps:

[0088] Step S1. Set up a platform on the top of the ball bag car. The front of the platform is marked with a scale. Two cameras are installed in parallel on the platform so that the main optical axis of the cameras is consistent with the direction of travel of the ball bag car, and at least one scale unit of the scale is within the shooting range of the camera.

[0089] Step S1 includes:

[0090] Step S11. Set up a platform on the top of the ball bag car, with the platform surface parallel to the ground. Install two cameras parallel to each other on the platform, so that the main optical axis of the cameras is consistent with the direction of travel of the ball bag car, and the distance between the two cameras and the center line of the platform is equal. The platform is made of transparent material so as not to block the shooting aperture of the cameras.

[0091] Step S12. Set a scale on the front edge of the platform, with the zero mark of the scale aligned with the center of the line connecting the two cameras, so that at least one scale unit on the scale is within the shooting range of the two cameras. Set a visual processing chip under the platform to receive images captured by the cameras and upload them to the information platform for processing.

[0092] Step S2. Two cameras capture images of the front of the ball bag car respectively. The image segmentation algorithm is used to separate the occluded parts in the image. Based on the length of the ruler in the image and the shooting range of the camera, the distance of the occluded object is measured using the binocular imaging principle to obtain the coordinates of each point on the edge of the occluded object.

[0093] Step S2 includes:

[0094] Step S21. Two cameras capture images of the front of the ball bag car, obtaining a first image and a second image respectively. The image segmentation algorithm is used to determine the occluders in the captured images. The image segmentation algorithm includes: K-means, FCN, DeepLab, R-CNN and U-Net algorithms.

[0095] Step S22: Based on the length of the ruler in the image and the camera's shooting range, calculate the distance between each point on the edge of the obstruction and the ball bag workshop using the binocular imaging principle:

[0096]

[0097] Where z1 represents the horizontal distance between the point and the ball bag workshop, x1 and x2 represent the x-coordinates of the point in the first image and the second image respectively, f represents the focal length of the camera, B represents the distance between the two cameras, a represents the compression ratio of the ruler, a = L / A, where L is the scale of the ruler in the image, and A is the pixel width of the ruler.

[0098] Since the two cameras are at the same height, y1 = y2, and the vertical distance between the point and the ball bag workshop is z2 = a·y1-h, where y1 and y2 represent the ordinates of the point in the first and second images, respectively.

[0099] Step S23. Measure the distance to each point on the edge of the obstruction to determine the horizontal and vertical distances from the ball bag workshop. Establish a coordinate system with the ball bag car as the zero point to obtain the coordinates of each point on the edge of the obstruction.

[0100] Step S3. Input the coordinates and grayscale of each point on the edge of the obstruction into the sample database. Mark the obstructions that are identified as impassable or unrecognizable as obstacles. Filter the space between adjacent obstacle edges whose length is greater than the width of the ball bag car to generate passable sectors.

[0101] Step S3 includes:

[0102] Step S31. Input the coordinates and grayscale of each point on the edge of the occluder into the sample database. Compare the existing obstacle samples in the sample database. If the comparison result is an impassable or unrecognizable sample, then the occluder is judged as an obstacle.

[0103] Step S32. Based on the user's satellite positioning coordinates, identify impassable obstacles within a semi-circular area centered on the direction of travel, where the direction of travel is from the ball-car coordinates to the user coordinates. Filter out spaces where the width between obstacles is greater than the width of the ball-car. Using the ball-car coordinates as vertices and the lines connecting the edges of the obstacles as chords, generate passable sectors.

[0104] Step S4. Determine the speed of the ball bag car based on the distance between the ball bag car and the nearest obstacle. Select the passable sector with the highest effective speed as the planning sector and take the direction of the angle bisector of the ball in the planning sector as the direction of the ball bag car's movement.

[0105] Step S4 includes:

[0106] Step S41. Determine the speed of the ball bag car based on the distance between the ball bag car and the nearest obstacle:

[0107]

[0108] Where v represents the speed of the ball-cart, v max The maximum speed of the ball bag car is represented by T, the distance between the ball bag car and the obstacle is represented by u, the braking deceleration of the ball bag car is represented by T0, the reserved deceleration length is represented by v0, and the preset speed reduction is represented by v0.

[0109] Step S42. Calculate the effective travel speed VR of each passable sector, where VR = v·cosθ, and θ represents the angle between the sector angle bisector and the travel direction. Select the passable sector with the largest effective travel speed as the planning sector, and take the direction of the angle bisector of the planning sector as the forward direction of the ball-cart.

[0110] Step S5. Determine the current coordinates of the ball bag car based on its forward direction and real-time speed, correct the travel deviation in real time, update the obstacle position, and apply emergency braking when the distance between the obstacle and the vehicle is detected to be below a threshold, and replan the forward direction.

[0111] Step S5 includes:

[0112] Step S51. Analyze the drive signal of the ball bag car. Based on the coordinates of the ball bag car before its movement, the direction of movement of the ball bag car, and the real-time speed, update the coordinates of the ball bag car and the position of the obstacles in the planned sector in real time, and correct the offset between the direction of movement of the ball bag car and the direction of the angle bisector of the ball in the planned sector.

[0113] Step S52. When the width between obstacles in the planned sector is less than the width of the ball bag car, stop the movement of the ball bag car, the camera takes another picture, identifies the passable sectors around the ball bag car at the current position, and replans the forward direction of the ball bag car. At the same time, when the distance between the obstacle and the vehicle is detected to be lower than the threshold, apply emergency braking.

[0114] Example: The pixel coordinates of the obstacle point in front of the ball bag car in camera 1 and camera 2 are (200, 100) and (400, 100) respectively. The focal length of the camera is 2m, the 0.1m mark on the scale occupies 100 pixels, and the distance between the cameras is 0.5m. Then the horizontal distance between the obstacle point and the ball bag car is 25m, and the vertical distance is 0.1m.

[0115] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0116] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A navigation and obstacle avoidance intelligent control method applied to golf bag carts, characterized in that, The method includes the following steps: Step S1. Set up a platform on the top of the ball bag car. The front of the platform is marked with a scale. Two cameras are installed in parallel on the platform so that the main optical axis of the cameras is consistent with the direction of travel of the ball bag car, and at least one scale unit of the scale is within the shooting range of the camera. Step S2. Two cameras capture images of the front of the ball bag car respectively. The image segmentation algorithm is used to separate the occluded parts in the image. Based on the length of the ruler in the image and the shooting range of the camera, the distance of the occluded object is measured using the binocular imaging principle to obtain the coordinates of each point on the edge of the occluded object. Step S3. Input the coordinates and grayscale of each point on the edge of the obstruction into the sample database. Mark the obstructions that are identified as impassable or unrecognizable as obstacles. Filter the space between adjacent obstacle edges whose length is greater than the width of the ball bag car to generate passable sectors. Step S4. Determine the speed of the ball bag car based on the distance between the ball bag car and the nearest obstacle. Select the passable sector with the highest effective speed as the planning sector and take the direction of the angle bisector of the ball in the planning sector as the direction of the ball bag car's movement. Step S5. Determine the current coordinates of the ball bag car based on its forward direction and real-time speed, correct the travel deviation in real time, update the obstacle position, and apply emergency braking when the distance between the obstacle and the vehicle is detected to be below a threshold, and replan the forward direction. Step S2 includes: Step S21. Two cameras capture images of the front of the ball bag car, obtaining a first image and a second image respectively. The image segmentation algorithm is used to determine the occluders in the captured images. The image segmentation algorithm includes: K-means, FCN, DeepLab, R-CNN and U-Net algorithms. Step S22: Based on the length of the ruler in the image and the camera's shooting range, calculate the distance between the point on the edge of the obstruction and the ball bag workshop using the binocular imaging principle: ； Where z represents the horizontal distance between the point and the ball bag workshop, x1 and x2 represent the x-coordinates of the points in the first and second images respectively, f represents the focal length of the camera, B represents the distance between the two cameras, a represents the compression ratio of the ruler, a=L / A, where L is the scale of the ruler in the image, and A is the pixel width of the ruler. Since the two cameras are at the same height, y1=y2, and the vertical distance between the point and the ball bag workshop is z2=a·y1-h, where h is the height of the camera, and y1 and y2 represent the ordinates of the point in the first and second images, respectively. Step S23. Measure the distance to each point on the edge of the obstruction to determine the horizontal and vertical distances from the ball bag workshop. Establish a coordinate system with the ball bag car as the zero point to obtain the coordinates of each point on the edge of the obstruction. Step S3 includes: Step S31. Input the coordinates and grayscale of each point on the edge of the occluder into the sample database. Compare the existing obstacle samples in the sample database. If the comparison result is an impassable or unrecognizable sample, then the occluder is judged as an obstacle. Step S32. Based on the user's satellite positioning coordinates, identify impassable obstacles in a semi-circular area centered on the direction of travel, where the direction of travel is from the ball car coordinates to the user coordinates. Filter out spaces where the width between obstacles is greater than the width of the ball car. Using the ball car coordinates as vertices and the lines connecting the edges of the obstacles as chords, generate passable sectors. Step S4 includes: Step S41. Determine the speed of the ball bag car based on the distance between the ball bag car and the nearest obstacle: ； Where v represents the speed of the ball-cart, v max The maximum speed of the ball bag car is represented by T, the distance between the ball bag car and the obstacle is represented by u, the braking deceleration of the ball bag car is represented by T0, the reserved deceleration length is represented by v0, and the preset speed reduction is represented by v0. Step S42. Calculate the effective travel speed VR of each passable sector, where VR = v·cosθ, and θ represents the angle between the sector angle bisector and the travel direction. Select the passable sector with the largest effective travel speed as the planning sector, and take the direction of the angle bisector of the planning sector as the forward direction of the ball-cart.

2. The intelligent control method for navigation and obstacle avoidance applied to a golf bag cart according to claim 1, characterized in that: Step S1 includes: Step S11. Set up a platform on the top of the ball bag car, with the platform surface parallel to the ground. Install two cameras parallel to each other on the platform, so that the main optical axis of the cameras is consistent with the direction of travel of the ball bag car, and the distance between the two cameras and the center line of the platform is equal. The platform is made of transparent material so as not to block the shooting aperture of the cameras. Step S12. Set a scale on the front edge of the platform, with the zero mark of the scale aligned with the center of the line connecting the two cameras, so that at least one scale unit on the scale is within the shooting range of the two cameras. Set a visual processing chip under the platform to receive images captured by the cameras and upload them to the information platform for processing.

3. The intelligent control method for navigation and obstacle avoidance applied to a golf bag cart according to claim 2, characterized in that: Step S5 includes: Step S51. Analyze the drive signal of the ball bag car. Based on the coordinates of the ball bag car before its movement, the direction of movement of the ball bag car, and the real-time speed, update the coordinates of the ball bag car and the position of the obstacles in the planned sector in real time, and correct the offset between the direction of movement of the ball bag car and the direction of the angle bisector of the ball in the planned sector. Step S52. When the width between obstacles in the planned sector is less than the width of the ball bag car, stop the movement of the ball bag car, the camera takes another picture, identifies the passable sectors around the ball bag car at the current position, and replans the forward direction of the ball bag car. At the same time, when the distance between the obstacle and the vehicle is detected to be lower than the threshold, apply emergency braking.

4. A navigation and obstacle avoidance intelligent control system for a golf cart, wherein the system executes the navigation and obstacle avoidance intelligent control method for a golf cart as described in claim 1, characterized in that, The system includes the following modules: camera platform module, obstacle recognition module, traffic planning module, attitude positioning module, and speed control module; The camera platform module is used to set up a platform on the top of the ball bag car. The front edge of the platform is perpendicular to the direction of travel of the ball bag car and is equipped with a scale. Two cameras are installed on the platform, so that the main optical axis vector of the two cameras is parallel to the direction of travel of the ball bag car, and the line connecting the two cameras is parallel to the front edge of the platform, so that at least one scale unit of the scale is within the shooting range of the two cameras. The obstacle recognition module is used to control the camera to capture images along the direction of travel of the ball bag car. The first image and the second image are obtained from the two cameras respectively. The occlusion part in the image is separated by the image segmentation algorithm. Taking the current position of the ball bag car as the origin, the distance of each point of the occlusion is measured according to the length of the ruler captured in the image and the installation distance of the camera, and the spatial coordinates of the point are given. All the coordinates of the points constitute the occlusion model. The occlusion model is sent to the sample database to identify the type of occlusion. The traffic planning module is used to determine the direction of travel of the ball bag vehicle based on the user's satellite positioning coordinates, identify impassable obstacles in a semi-circular area centered on the direction of travel vector, filter the space between the edges of the obstacles whose length is greater than the width of the ball bag vehicle, generate passable sectors, verify all passable sectors, select the sector that is closest to the user's positioning after the ball bag vehicle passes through as the planning sector, and control the ball bag vehicle to move forward with the center of the sector as the direction of travel. The attitude positioning module is used to establish a positioning model based on the ball bag car's forward direction and real-time speed, input time parameters into the model, output the ball bag car's current coordinates, correct the travel deviation, and measure the distance again with the current coordinates as the center at fixed intervals to update the distance between the ball bag car and each obstacle. The speed control module is used to control the car's speed in real time based on the sector radius, the distance between the ball car and the obstacle, and the ball car's braking parameters. When the sector radius is smaller than the passage radius during the forward movement, braking is performed; when the deviation exceeds the braking torque, deceleration is performed; and the forward direction and speed of the ball car are replanned.

5. The intelligent navigation and obstacle avoidance control system for a golf bag cart according to claim 4, characterized in that: The camera platform module includes: an edge positioning unit and a dual-camera imaging unit; The edge positioning unit is used to build a platform on the top of the ball bag car. The platform is parallel to the ground and has a scale on the front edge. The dual-camera imaging unit is used to fix two cameras on the platform and detect obstacles using computer vision and binocular moment measurement principles; The obstacle recognition module includes: an image segmentation unit and a sample comparison unit; The image segmentation unit is used to determine occlusions in the captured image using image segmentation algorithms, including K-means, FCN, DeepLab, R-CNN, and U-Net algorithms. The sample comparison unit is used to compare the features of the obstruction with the features in the sample library. If the comparison result is an impassable or unrecognizable sample, it is judged as an obstacle.

6. The intelligent navigation and obstacle avoidance control system for a golf bag cart according to claim 5, characterized in that: The traffic planning module includes: a signal transmission unit, a sector identification unit, and a path selection unit; The signal transmission unit is used to receive the user's location information and determine the direction and distance between the user and the ball bag vehicle; The sector identification unit is used to calculate the distance between the ball bag car and each obstacle, and to identify the passable sectors of the ball bag car; The path selection unit is used to filter passable sectors and plan the shortest route.

7. The intelligent navigation and obstacle avoidance control system for a golf bag cart according to claim 6, characterized in that: The attitude localization module includes: an attitude modeling unit and a parameter localization unit; The attitude modeling unit is used to establish the attitude positioning model of the ball bag vehicle and return the current coordinates of the ball bag vehicle in real time. The parameter positioning unit is used to correct the travel deviation, update the obstacle position, and determine the distance between the obstacle and the ball bag car.

8. The intelligent navigation and obstacle avoidance control system for a golf bag cart according to claim 7, characterized in that: The speed control module includes: a speed planning unit, a dynamic feedback unit, and a vehicle braking unit; The speed planning unit is used to determine the speed of the golf bag car based on the minimum distance between the golf bag car and the obstacle and the vehicle braking distance. The dynamic feedback unit is used to update the obstacle position and re-plan the ball bag car's speed and direction of travel; The vehicle braking unit is used to apply emergency braking when the distance between the vehicle and an obstacle is detected to be less than a threshold, and to assist in vehicle deceleration.

Images