An indoor navigation control system for a guide robot based on big data

Abstract

The invention discloses an indoor navigation control system of a guide robot based on big data, comprising: a user information collection module, a navigation data processing center, a memory data retrieval module, a high-altitude obstacle avoidance sensing module, a deviation obstacle avoidance sensing module and a voice prompt guidance module. The user information collection module collects the user's behavior, height and body width data to the navigation data processing center, and the memory data retrieval module retrieves the path matching the current destination in the user behavior data to judge whether there is a new obstacle. The high-altitude obstacle avoidance sensing module compares the high-altitude obstacles on the current path with the user's height to judge whether the user can pass the obstacle smoothly. The deviation obstacle avoidance sensing module predicts the relative position of the user and the robot according to the tension change of the traction rope, and adjusts the detection range to cover the path just in front of the user, thereby improving the navigation detection work efficiency and reducing the power consumption of the robot.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to an indoor navigation control system of a guide robot based on big data. Background Art

[0002] The application of robots has been further expanded, gradually extending from industrial production to medical, family and educational services. The application of robots in the medical industry has always attracted much attention. As a kind of service robot, guide robots have ushered in a period of development. At present, most of the guide equipment on the market is still guide sticks. Although they can detect the road conditions ahead to a certain extent, the detection ability is limited, and the blind need to have a certain degree of independent judgment ability. They cannot actively guide the blind to move, especially when walking in large indoor spaces (such as shopping malls, etc.), there are too many people, and guide robots are needed to guide users to avoid obstacles and navigate the correct path. However, the research and development technology of guide robots needs to be further improved, especially for robots that need to be leashed. The blind follow behind the robot, and the robot can only detect obstacles on the path ahead. The possibility that the blind person behind them may gradually deviate from the original navigation path is ignored. Although some robots are equipped with multiple cameras to detect the walking conditions of the blind, turning on multiple cameras at the same time not only increases the production cost of the robot, but also easily consumes the robot's power and reduces the robot's working time. Since a tension sensor is installed on the traction rope, the relative position of the user and the robot can be judged according to the change in tension. When necessary, other cameras can be turned on to accurately adjust the detection range, thereby reducing the robot's power consumption while ensuring the safety of the blind. In addition, high obstacles with hollow holes in the middle (such as railings, door tops, etc.) are often ignored during the detection process. The safety of the user's passage can be judged based on the user's height, avoiding the situation where the user hits high obstacles and improving safety during indoor activities.

[0003] Therefore, people need a guide robot indoor navigation control system based on big data to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide an indoor navigation control system for a guide robot based on big data to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an indoor navigation control system of a guide robot based on big data, characterized in that: the system comprises: a user information collection module, a navigation data processing center, a memory data retrieval module, a high obstacle avoidance sensing module, a deviation obstacle avoidance sensing module and a voice prompt guidance module;

[0006] The user information collection module collects the user's body width data, height data and historical indoor activity trajectory data, and transmits the collected data to the navigation data processing center. The memory data retrieval module retrieves the historical indoor activity trajectory data of the destination that can be reached in the navigation data processing center to the high-altitude obstacle avoidance sensing module. The high-altitude obstacle avoidance sensing module detects the current environment, compares the current environment with the historical path, and determines whether there are new obstacles on the path corresponding to the current indoor environment. If there are new obstacles, adjust the new navigation path; if there are no new obstacles, navigate along the original route, and perform navigation detection on the new path. If there are high obstacles, the memory data retrieval module retrieves the user's height data in the navigation data processing center to the high-altitude obstacle avoidance sensing module. The high-altitude obstacle avoidance sensing module compares the user's height with the height of the obstacle: if the obstacle is higher than the user , navigate along the original route; if the obstacle is lower than the user, adjust the new navigation path, the deviation obstacle avoidance sensing module predicts the relative position of the user and the guide robot through the change of the tension of the traction rope, and adjusts the detection range: when the user is directly behind the guide robot, retrieve the body width data in the navigation data processing center, and set the original detection range to just cover the path directly in front of the user; when the tension is sensed to be 0, it is predicted that the user is not directly behind the guide robot, turn on the rear camera of the robot for detection, analyze the relative position of the user and the robot, and adjust the detection range until it covers the path directly in front of the user. During the navigation process, the user is guided to avoid obstacles through the voice prompt guidance module. In the initial state, only the front path is detected by default. According to the change of the traction rope tension, it is predicted that the user deviates from the position directly behind the robot, and then the rear camera is turned on for confirmation, which is conducive to reducing the power consumption of the robot while adjusting the corresponding detection range to ensure the safety of the user.

[0007] Furthermore, the user information collection module includes a behavior data collection unit, a body width collection unit and a height data collection unit. The behavior data collection unit collects historical indoor activity paths that have been successfully navigated according to the user's destination, the body width collection unit collects the user's body width data, and the height data collection unit collects the user's height data. All collected data are stored in the navigation data processing center for retrieval by the memory data retrieval module.

[0008] Furthermore, the high-altitude obstacle avoidance sensing module includes an indoor environment detection unit, a change obstacle capture unit, a navigation path adjustment unit, an obstacle height detection unit and a height comparison unit. The indoor environment detection unit detects the current indoor environment, compares the current environment with the environment on the historical path through the change obstacle capture unit, and determines whether there are new obstacles on the path corresponding to the current environment compared with the historical path. If there are new obstacles, a new path navigation is selected through the navigation path adjustment unit, and obstacles at high places on the new path are detected through the obstacle height detection unit. When an obstacle at a high place is detected, the obstacle height is compared with the user's height through the height comparison unit, and the path is readjusted when the obstacle is lower than the user's height.

[0009] Furthermore, the deviation obstacle avoidance sensing module includes an occlusion obstacle detection unit, a traction looseness sensing unit, a relative position analysis unit, a detection range adjustment unit and an out-of-range detection unit. The traction looseness sensing unit determines whether the traction rope is loose based on the sensed tension change. When it is set that the user is just behind the robot when it is not loose, the initial detection range is set by the occlusion obstacle detection unit to just cover the path directly in front of the user. When it is determined that the traction rope is loose, it is predicted that the user has deviated from the position directly behind the robot, and the rear camera of the robot is started. The relative position analysis unit predicts the user's current position based on the image detected by the rear camera. When it is confirmed that the user has deviated from the position directly behind the robot, the initial detection range is adjusted by the detection range adjustment unit until it covers the path directly in front of the user's current position. The out-of-range detection unit performs path navigation detection based on the adjusted range, and guides the user to avoid obstacles during navigation through the voice prompt guidance module.

[0010] Furthermore, the height of the user collected by the height data collection unit is H, and the set of historical indoor activity paths of the user who has successfully navigated collected by the behavior data collection unit is L={L1, L2, ..., L n}, the maximum width of the user's body when the user stretches out his arms is collected by the body width collection unit as a, and all collected data are transmitted to the navigation data processing center S2.

[0011] Furthermore, the memory data retrieval module retrieves the user's height and historical indoor activity path to the variable obstacle capture unit, detects the current indoor environment through the indoor environment detection unit, and obtains the activity path matching the user's destination as L i , compared with path L i The current environment and the historical environment in the navigation path are compared. When there is a new obstacle, the navigation path adjustment unit selects a new navigation path. The obstacle height detection unit detects that the height of the obstacle at a high place on the new path is h. H and h are compared: If , indicating that the obstacle is higher than the user and the user can continue to move forward; if , indicating that the obstacle is shorter than the user. It is analyzed that the user will encounter the obstacle if he goes straight, and the path is adjusted by the navigation path adjustment unit. The high obstacle refers to the hollow object in the middle, such as railings, door tops, etc. The user's height data is retrieved by big data technology and compared with the height of the obstacle in the current environment. This is conducive to accurately judging whether the obstacle can be passed smoothly according to the user's own situation without bypassing the obstacle. The addition of the user's height parameter reduces the workload of the guide robot to find a new path, improves the navigation work efficiency, and speeds up the time for the user to reach the destination.

[0012] Furthermore, the tension of the traction rope when the traction loosening sensing unit senses that the traction rope is not loose is F 初 , the user is right behind the robot, and the longest detection diameter of the robot is set to r by the obstacle detection unit. The initial detection angle is calculated according to the following formula :

[0013] ,

[0014] At the detection angle When the detection range just covers the path in front of the user, when the traction rope tension is sensed to be 0, the traction rope is loose, the rear camera of the guide robot is started, and it is detected that the relative horizontal distance between the user and the robot is d at this time. The detection range is adjusted by the detection range adjustment unit: the detection angle of the robot after the range adjustment is calculated according to the following formula: :

[0015] ,

[0016] At the detection angle When the detection range just covers the path in front of the user's current position, the adjusted angle is transmitted to the detection unit outside the range. When the tension is sensed to be 0, it is predicted that the user has deviated from the rear of the robot. At this time, the rear camera is turned on for detection to confirm whether the prediction result is correct. The relative horizontal distance between the user and the robot and the detection distance of the robot are confirmed according to the relative position, and the detection angle is adjusted in combination with the retrieved user's body width data, thereby reducing the invalid area of ​​the path detection range in the traditional method. The behavioral data of users' indoor activities are uncertain, and the detection area is automatically adjusted according to the different positions of the users, thereby improving the efficiency of navigation detection and reducing the power consumption of the robot.

[0017] Furthermore, the out-of-range detection unit is configured to detect the angle of the The corresponding range-controlled robot detects the path ahead and transmits the images detected during the forward movement to the destination to the voice prompt guidance module. The voice prompt guidance module reminds the user to change direction after advancing a certain number of steps based on the received images and the user's stride data, and to bypass obstacles and successfully reach the destination. The user's stride data is obtained through the height data in the navigation data processing center. The user's stride can be analyzed accordingly based on the retrieved user's height data. Prompting the user with a clear number of steps is clearer than prompting how far an obstacle will be encountered, thereby enhancing the user's sense of security and trust in the guide robot.

[0018] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0019] 1. The present invention uses big data technology to retrieve the user's behavior data to match the historical path corresponding to the destination the current user wants to go to, uses the indoor environment detection unit to detect the surrounding of the path, and uses the change obstacle capture unit to determine whether there are new obstacles in the current path matching the historical path. When encountering a high obstacle, the user's height data is retrieved and compared with the height of the obstacle to determine whether the user can successfully pass the obstacle, thereby reducing the unnecessary work of the robot to find a new path and speeding up the time to navigate to the destination;

[0020] The relative position of the user and the robot is predicted by the change in the tension of the traction rope, and the rear camera is turned on to confirm the prediction result. According to the different positions of the user, the detection range is adjusted in combination with the retrieved user body width data to cover the path directly in front of the user, thereby improving the efficiency of navigation detection and reducing the power consumption of the robot.

[0021] The user's stride can be analyzed based on the retrieved user's height data. Providing the user with a clear number of steps is clearer than indicating the distance at which obstacles will be encountered, which enhances the user's sense of security during indoor activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0023] Figure 1 This is a structural composition diagram of an indoor navigation control system of a guide robot based on big data in the present invention;

[0024] Figure 2 It is a schematic diagram of the detection range adjustment corresponding to the relative position of the user of the present invention. DETAILED DESCRIPTION

[0025] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0026] See also Figure 1-Figure 2 The present invention provides a technical solution: an indoor navigation control system of a guide robot based on big data, characterized in that the system comprises: a user information collection module S1, a navigation data processing center S2, a memory data retrieval module S3, a high obstacle avoidance sensing module S4, a deviation obstacle avoidance sensing module S5 and a voice prompt guidance module S6;

[0027] The user information collection module S1 collects the user's body width data, height data and historical indoor activity trajectory data, and transmits the collected data to the navigation data processing center S2. The memory data retrieval module S3 retrieves the historical indoor activity trajectory data that can reach the current destination in the navigation data processing center S2 to the high obstacle avoidance sensing module S4. The high obstacle avoidance sensing module S4 detects the current environment, compares the current environment with the historical path, and determines whether there are new obstacles on the path corresponding to the current indoor environment. If there are new obstacles, adjust the new navigation path; if there are no new obstacles, navigate along the original route, and perform navigation detection on the new path. If there are high obstacles, the memory data retrieval module S3 retrieves the user's height data in the navigation data processing center S2 to the high obstacle avoidance sensing module S4. The high obstacle avoidance sensing module S4 compares the user's height with the height of the obstacle: if the obstacle is higher than the user , navigate along the original route; if the obstacle is lower than the user, adjust the new navigation path, and deviate from the obstacle avoidance sensing module S5. The relative position of the user and the guide robot is predicted by the change of the tension of the traction rope, and the detection range is adjusted: when the user is directly behind the guide robot, the body width data in the navigation data processing center S2 is retrieved, and the original detection range is set to cover the path directly in front of the user; when the tension is sensed to be 0, it is predicted that the user is not directly behind the guide robot, and the rear camera of the robot is turned on for detection, and the relative position of the user and the robot is analyzed. The detection range is adjusted until the path directly in front of the user is covered. During the navigation process, the voice prompt guidance module S6 guides the user to avoid obstacles. In the initial state, only the front path is detected by default. According to the change of the traction rope tension, it is predicted that the user deviates from the position directly behind the robot, and then the rear camera is turned on for confirmation. The corresponding detection range can be adjusted to ensure the safety of the user while reducing the power consumption of the robot.

[0028] The user information collection module S1 includes a behavior data collection unit, a body width collection unit and a height data collection unit. The behavior data collection unit collects historical indoor activity paths that have been successfully navigated according to the user's destination, the body width collection unit collects the user's body width data, and the height data collection unit collects the user's height data. All collected data are stored in the navigation data processing center S2 for retrieval by the memory data retrieval module S3.

[0029] The high-altitude obstacle avoidance sensing module S4 includes an indoor environment detection unit, a change obstacle capture unit, a navigation path adjustment unit, an obstacle height detection unit and a height comparison unit. The indoor environment detection unit detects the current indoor environment, compares the current environment with the environment on the historical path through the change obstacle capture unit, and determines whether there are new obstacles on the path corresponding to the current environment compared with the historical path. If there are new obstacles, the navigation path adjustment unit selects a new path navigation, and the obstacle height detection unit detects the high obstacles on the new path. When an obstacle is detected at a high place, the height of the obstacle is compared with the user's height through the height comparison unit, and the path is readjusted when the obstacle is lower than the user's height.

[0030] The deviation obstacle avoidance sensing module includes an occlusion obstacle detection unit, a traction looseness sensing unit, a relative position analysis unit, a detection range adjustment unit and an out-of-range detection unit. The traction looseness sensing unit determines whether the traction rope is loose based on the sensed tension change. When it is set that the user is just behind the robot when it is not loose, the initial detection range is set by the occlusion obstacle detection unit to cover the path just in front of the user. When it is determined that the traction rope is loose, it is predicted that the user has deviated from the position just behind the robot, and the rear camera of the robot is started. The relative position analysis unit predicts the user's current position based on the image detected by the rear camera. When it is confirmed that the user has deviated from the rear of the robot, the initial detection range is adjusted by the detection range adjustment unit until it covers the path just in front of the user's current position. The out-of-range detection unit performs path navigation detection based on the adjusted range, and guides the user to avoid obstacles during navigation through the voice prompt guidance module.

[0031] The height of the user collected by the height data collection unit is H, and the set of historical indoor activity paths of the user who successfully navigated is collected by the behavior data collection unit as L={L1, L2, ..., L n}, the body width of the user is collected as a by the body width collection unit, and all collected data are transmitted to the navigation data processing center S2.

[0032] The memory data retrieval module S3 retrieves the user's height and historical indoor activity path to the change obstacle capture unit, detects the current indoor environment through the indoor environment detection unit, and obtains the activity path matching the user's destination as L i , compared with path L i The current environment and the historical environment in the navigation path are compared. When there is a new obstacle, the navigation path adjustment unit selects a new path for navigation. The obstacle height detection unit detects that the height of the obstacle at a high place on the new path is h. H and h are compared: If , indicating that the obstacle is higher than the user and the user can continue to move forward; if , indicating that the obstacle is shorter than the user. It is analyzed that the user will encounter the obstacle if he goes straight, and the path is adjusted through the navigation path adjustment unit. The high obstacles refer to the hollow objects in the middle, such as railings, door tops, etc. The user's height data is retrieved by big data technology and compared with the height of the obstacle in the current environment. It is helpful to accurately judge whether the obstacle can be passed smoothly according to the user's own situation without bypassing the obstacle. Adding the user's height parameter is convenient to reduce the workload of the guide robot to find a new path, which can improve the navigation work efficiency and speed up the time for the user to reach the destination.

[0033] The tension when the traction loosening sensing unit senses that the traction rope is not loose is F 初 , the user is right behind the robot, and the longest detection diameter of the robot is set to r by blocking the obstacle detection unit. The initial detection angle is calculated according to the following formula :

[0034] ,

[0035] At the detection angle When the detection range just covers the path in front of the user, when the traction rope tension is sensed to be 0, the traction rope is loose, the rear camera of the guide robot is started, and the relative horizontal distance between the user and the robot is detected to be d. The detection range is adjusted by the detection range adjustment unit: the detection angle of the robot after the range adjustment is calculated according to the following formula: :

[0036] ,

[0037] At the detection angle When the detection range just covers the path in front of the user's current position, the adjusted angle is transmitted to the out-of-range detection unit. When the tension is sensed to be 0, it is predicted that the user has deviated from the rear of the robot. At this time, the rear camera is turned on for detection to confirm whether the prediction result is correct. The relative horizontal distance between the user and the robot and the robot's detection distance are confirmed according to the relative position, and the detection angle is adjusted in combination with the retrieved user's body width data, which reduces the invalid area of ​​the path detection range in the traditional method. The user's behavioral data for indoor activities is uncertain, and the detection area is automatically adjusted according to the different positions of the user, which is convenient for improving the efficiency of navigation detection while reducing the robot's power consumption.

[0038] Out-of-range detection unit according to angle The corresponding range-controlled robot detects the path ahead and transmits the images detected during the forward movement to the destination to the voice prompt guidance module S6. The voice prompt guidance module S6 reminds the user to change direction after advancing a certain number of steps based on the received images and the user's stride data, and to bypass obstacles and successfully reach the destination. The user's stride data is obtained through the height data in the navigation data processing center S2. The user's stride can be analyzed accordingly based on the retrieved user's height data. Prompting the user with a clear number of steps is clearer than prompting the user at a certain distance before encountering obstacles, which can enhance the user's sense of security and trust in the guide robot.

[0039] Embodiment 1: The height data collection unit collects the user's height H=1.7 meters, and the behavior data collection unit collects the user's historical indoor activity path set that has successfully navigated as L={L1, L2, L3, L4, L5}. The memory data retrieval module retrieves the user's height and historical indoor activity path into the variable obstacle capture unit, and detects the current indoor environment through the indoor environment detection unit. According to the user's destination, the activity path that matches it is obtained as L5. By comparing the current environment and the historical environment in path L5, it is found that a new obstacle has been generated. The navigation path adjustment unit selects a new path for navigation, and the obstacle height detection unit detects that the height of the obstacle at a high place on the new path is h=1.5 meters. Compare H and h: , which means that the obstacle is shorter than the user. The user may hit the obstacle if he goes straight, and the path is readjusted through the navigation path adjustment unit;

[0040] Embodiment 2: The maximum body width a of the user when stretching his arms is collected by the body width collection unit is 1.68 meters, and the tension of the traction rope when the traction loosening sensing unit is sensed to be F 初 =150N, the user is right behind the robot, and the longest detection diameter of the robot is set to r=2 meters by blocking the obstacle detection unit. According to the formula Calculate the initial detection angle , at the detection angle of When the detection range just covers the path in front of the user, the tension of the traction rope is sensed to be 0 and the traction rope is loose. The rear camera of the guide robot is started and it is detected that the relative horizontal distance between the user and the robot is d = 0.5 meters. The detection range is adjusted by the detection range adjustment unit: According to the formula Calculate the robot's detection angle after range adjustment , at the detection angle of When the detection range just covers the path in front of the user's current position, the adjusted angle is transmitted to the out-of-range detection unit, and the out-of-range detection unit adjusts the angle according to the angle. The corresponding range controls the robot to detect the path ahead.

[0041] Finally, it should be noted that the above description is only a preferred example of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A guide robot indoor navigation control system based on big data, characterized by: The system includes: a user information collection module S1, a navigation data processing center S2, a memory data retrieval module S3, a high-altitude obstacle avoidance sensing module S4, a deviation obstacle avoidance sensing module S5 and a voice prompt guidance module S6; The user information collection module S1 collects the user's body width data, height data and historical indoor activity trajectory data, and transmits the collected data to the navigation data processing center S2. The memory data retrieval module S3 retrieves the historical indoor activity trajectory data that can reach the current destination in the navigation data processing center S2 to the high-altitude obstacle avoidance sensing module S4. The high-altitude obstacle avoidance sensing module S4 detects the current environment, compares the current environment with the historical path, and determines whether there are new obstacles on the path corresponding to the current indoor environment. If there are new obstacles, adjust the new navigation path; if there are no new obstacles, navigate along the original route, perform navigation detection on the new path, and encounter high obstacles. The memory data retrieval module S3 retrieves the user's height data in the navigation data processing center S2 To the high-altitude obstacle avoidance sensing module S4, the high-altitude obstacle avoidance sensing module S4 compares the height of the user and the height of the obstacle: if the obstacle is higher than the user, navigate along the original route; if the obstacle is lower than the user, adjust the new navigation path, the deviation obstacle avoidance sensing module S5 predicts the relative position of the user and the guide robot through the tension change of the traction rope, and adjusts the detection range: when the user is directly behind the guide robot, the body width data in the navigation data processing center S2 is retrieved, and the original detection range is set to cover the path directly in front of the user; when the tension is sensed to be 0, it is predicted that the user is not directly behind the guide robot, the robot's rear camera is turned on for detection, the relative position of the user and the robot is analyzed, and the detection range is adjusted until it covers the path directly in front of the user. During the navigation process, the user is guided to avoid obstacles through the voice prompt guidance module S6.

2. According to the big data-based indoor navigation control system of a guide robot according to claim 1, it is characterized by: The user information collection module S1 includes a behavior data collection unit, a body width collection unit and a height data collection unit. The behavior data collection unit collects historical indoor activity paths that have been successfully navigated according to the user's destination, the body width collection unit collects the user's body width data, and the height data collection unit collects the user's height data. All collected data are stored in the navigation data processing center S2 for retrieval by the memory data retrieval module S3.

3. According to the big data-based indoor navigation control system of a guide robot according to claim 2, it is characterized by: The high-altitude obstacle avoidance sensing module S4 includes an indoor environment detection unit, a change obstacle capture unit, a navigation path adjustment unit, an obstacle height detection unit and a height comparison unit. The indoor environment detection unit detects the current indoor environment, compares the current environment with the environment on the historical path through the change obstacle capture unit, and determines whether there are new obstacles on the path corresponding to the current environment compared with the historical path. If there are new obstacles, a new path navigation is selected through the navigation path adjustment unit, and obstacles at high places on the new path are detected through the obstacle height detection unit. When an obstacle at a high place is detected, the obstacle height is compared with the user's height through the height comparison unit, and the path is readjusted when the obstacle is lower than the user's height.

4. The indoor navigation control system of a guide robot based on big data according to claim 2 is characterized in that: The deviation obstacle avoidance sensing module S5 includes an obstruction obstacle detection unit, a traction looseness sensing unit, a relative position analysis unit, a detection range adjustment unit and an out-of-range detection unit. The traction looseness sensing unit determines whether the traction rope is loose based on the sensed tension change. It is set that the user is just behind the robot when it is not loose. The initial detection range is set by the obstruction obstacle detection unit to just cover the path in front of the user. When it is determined that the traction rope is loose, it is predicted that the user has deviated from the position directly behind the robot, and the rear camera of the robot is started. The relative position analysis unit predicts the user's current position based on the image detected by the rear camera. When it is confirmed that the user has deviated from the position directly behind the robot, the initial detection range is adjusted by the detection range adjustment unit until it covers the path in front of the user's current position. The out-of-range detection unit performs path navigation detection based on the adjusted range, and guides the user to avoid obstacles during navigation through the voice prompt guidance module.

5. According to the big data-based indoor navigation control system of a guide robot according to claim 3, it is characterized by: The height of the user collected by the height data collection unit is H, and the set of historical indoor activity paths of the user who successfully navigated collected by the behavior data collection unit is L={L1, L2, ..., L n }, the maximum width of the user's body when the user stretches out his arms is collected by the body width collection unit as a, and all collected data are transmitted to the navigation data processing center S2.

6. The indoor navigation control system of a guide robot based on big data according to claim 3 is characterized by: The memory data retrieval module S3 retrieves the user's height and historical indoor activity path to the variable obstacle capture unit, detects the current indoor environment through the indoor environment detection unit, and obtains the activity path matching the user's destination as L i , compared with path L i The current environment and the historical environment in the navigation path are compared. When there is a new obstacle, the navigation path adjustment unit selects a new navigation path. The obstacle height detection unit detects that the height of the obstacle at a high place on the new path is h. H and h are compared: If , indicating that the obstacle is higher than the user and the user can continue to move forward; if , indicating that the obstacle is shorter than the user. It is analyzed that the user will encounter the obstacle if he goes straight, and the path is adjusted by the navigation path adjustment unit.

7. The indoor navigation control system of a guide robot based on big data according to claim 4 is characterized by: The tension when the traction loosening sensing unit senses that the traction rope is not loose is F 初 , the user is right behind the robot, and the longest detection diameter of the robot is set to r by the obstacle detection unit. The initial detection angle is calculated according to the following formula : ， a represents the maximum width of the user's body when the user stretches out his arms. When the detection range just covers the path in front of the user, when the traction rope tension is sensed to be 0, the traction rope is loose, the rear camera of the guide robot is started, and it is detected that the relative horizontal distance between the user and the robot is d at this time. The detection range is adjusted by the detection range adjustment unit: the detection angle of the robot after the range adjustment is calculated according to the following formula: : ， At the detection angle When the detection range just covers the path in front of the user's current position, the adjusted angle is transmitted to the detection unit outside the range.

8. The indoor navigation control system of a guide robot based on big data according to claim 7, characterized in that: The out-of-range detection unit is configured according to the angle The corresponding range control robot detects the path ahead and transmits the images detected during the forward movement to the destination to the voice prompt guidance module S6. The voice prompt guidance module S6 reminds the user to change direction after advancing a certain number of steps based on the received images and the user's stride data, bypassing obstacles and successfully reaching the destination. The user's stride data is obtained through the height data in the navigation data processing center S2.

Images