A method for robot to identify obstacles based on human-computer interaction

Abstract

The present invention discloses a method for robot recognition of obstacles based on human-computer interaction, S1, taking a panoramic photo including the starting point and the end point of the robot, uploading the photo to a remote terminal, uploading the photo to a central processing module in the robot through the remote terminal, S2, after the robot receives the photo, plans a straight path of the starting point and the end point through an original path planning module and sends it to an operation decision unit, the present invention relates to the field of robot technology. The method for robot recognition of obstacles based on human-computer interaction can select the optimal obstacle avoidance direction according to the shape of the obstacle, and after selecting the correct direction, can calculate the optimal path in real time according to the shape of different obstacles, ensuring that the robot can safely, effectively, accurately and efficiently select the optimal path during the obstacle avoidance process, thereby greatly improving the robot's obstacle avoidance efficiency.

Description

Technical Field

[0001] The present invention relates to the field of robot technology, and in particular to a method for a robot to recognize obstacles based on human-machine interaction. Background Art

[0002] A robot is an automated machine. The difference is that this machine has some intelligent capabilities similar to humans or living things, such as perception, planning, movement and coordination. It is an automated machine with high flexibility. Robots are widely used in industry, life and other aspects.

[0003] If robots are used in the fields of mobile transportation and mobile operations, their autonomous obstacle recognition and autonomous obstacle avoidance are inevitable. At present, the recognition function of robots with obstacle recognition is relatively mature, but there are also cases of recognition errors. Images that should not be regarded as obstacles are recognized as obstacles, causing the robot to make too many detours during movement. At the same time, when avoiding obstacles, it cannot avoid them efficiently, that is, it cannot effectively find the optimal path when avoiding them. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method for robot obstacle recognition based on human-computer interaction, which solves the problem that the existing obstacle avoidance robots have inaccurate obstacle recognition and poor avoidance paths.

[0005] To achieve the above objectives, the present invention is implemented by the following technical solutions: A method for robot obstacle recognition based on human-computer interaction, specifically comprising the following steps:

[0006] S1, take a panoramic photo including the starting point and the end point of the robot, upload the photo to the remote terminal, and upload the photo to the central processing module in the robot through the remote terminal;

[0007] S2. After receiving the photo, the robot plans a straight path between the starting point and the end point through the original path planning module and sends it to the operation decision unit. The operation decision unit will issue a command to the drive module, and the robot starts to move along the straight path;

[0008] S3. When the robot is moving, the panoramic camera module will take real-time video and transmit the video images to the obstacle recognition unit. The obstacle recognition unit will intercept all suspected obstacles in the video, compare the suspected obstacle images with the networked image library, and determine the types of suspected obstacles. The obstacle selection module determines the final obstacle based on the determined types.

[0009] S4, the laser sensor uses laser to measure the distance to the robot according to the final obstacle determined in step S3, and sends it to the operation decision unit through the central processing module;

[0010] S5. After receiving the obstacle information, the operation decision unit will obtain the real-time operation speed of the robot through the speed transmission module, and calculate the time required to reach the obstacle position through the distance calculation module. The path execution calculation module will re-formulate the route according to the obstacle position, so as to bypass the obstacle before the time required to reach the obstacle position, and return to the original straight path after the bypass is completed;

[0011] S6. The panoramic camera module will send the real-time video to the remote terminal through the central processing module. The remote terminal can view the video in real time through the real-time video viewing module. The remote terminal control personnel can directly control the movement of the robot through the remote control module.

[0012] Preferably, the obstacle selection module in step S3 includes an obstacle database. After receiving the obstacle type name, the obstacle selection module compares the obstacle type name with the obstacle database, and those that can be successfully compared in the obstacle database are considered as final obstacles.

[0013] Preferably, the execution principle of the path execution calculation module in step S5 is: the straight path in step S2 is recorded as L, and the obstacle is recorded as Z. The straight path L will divide the obstacle Z into two parts, which are recorded as Z1 and Z2 respectively. The distance from the farthest point of the outer edge of Z1 to the straight path L is recorded as A1, and the distance from the farthest point of the outer edge of Z2 to the straight path L is recorded as A2. If A1 is less than A2, the robot chooses to walk on the Z1 side to avoid obstacles. If A1 is greater than A2, the robot chooses to walk on the Z2 side to avoid obstacles. If A1 is equal to A2, the robot walks on the right side to avoid obstacles. The walking path is a straight path between the starting turning point of obstacle avoidance and A1 or A2. If an obstacle is encountered again during walking along this path, the path is replanned according to the above-mentioned execution principles until the robot returns to the original path.

[0014] Preferably, the remote terminal is bidirectionally connected to the central processing module, and the output end of the central processing module is connected to the input end of the original path planning module, the output end of the original path planning module is connected to the input end of the operation decision unit, and the output end of the operation decision unit is connected to the input end of the driving module.

[0015] Preferably, the input end of the panoramic camera module is connected to the input end of the central processing module, and the output end of the panoramic camera module is connected to the input end of the obstacle recognition unit, the output end of the obstacle recognition unit is connected to the input end of the laser sensor, the output end of the laser sensor is connected to the input end of the central processing module, and the output end of the central processing module is connected to the input end of the operation decision unit.

[0016] Preferably, the output end of the driving module is connected to the speed data transmission module, and the output end of the speed data transmission module is connected to the input end of the operation decision unit.

[0017] Preferably, the remote terminal includes a remote control module and a real-time video viewing module.

[0018] Preferably, the operation decision unit includes a distance calculation module and a path execution accounting module.

[0019] Beneficial Effects

[0020] The present invention provides a method for robot obstacle recognition based on human-machine interaction. Compared with the existing technology, it has the following beneficial effects:

[0021] (1) The method for robot obstacle recognition based on human-computer interaction, after receiving obstacle information, the running decision unit will obtain the real-time running speed of the robot through the speed transmission module, and calculate the time required to reach the obstacle position through the distance calculation module. The path execution calculation module will re-formulate the route according to the obstacle position, so as to bypass the obstacle before the time required to reach the obstacle position. After bypassing, it returns to the original straight path and can select the optimal obstacle avoidance direction according to the shape of the obstacle. After selecting the correct direction, it can calculate the optimal path in real time according to the shape of different obstacles, ensuring that the robot can safely, effectively, accurately and efficiently select the optimal path during the obstacle avoidance process, thereby greatly improving the robot's obstacle avoidance efficiency.

[0022] (2) The method for robot obstacle recognition based on human-computer interaction includes an obstacle database in the obstacle selection module in step S3. After the obstacle selection module receives the obstacle type name, it will compare the obstacle type name with the obstacle database. The obstacle that can be successfully compared with the obstacle database is considered as the final obstacle. The obstacle can be effectively and accurately identified. When the obstacle is identified, the obstacle can be accurately removed, thereby avoiding unnecessary avoidance of the robot and improving the robot's mobile obstacle avoidance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a schematic diagram of the obstacle avoidance process of the present invention;

[0024] Figure 2 It is a structural principle block diagram of the system of the present invention;

[0025] Figure 3 It is a structural principle block diagram of the operation decision unit of the present invention;

[0026] Figure 4 It is a structural principle block diagram of the remote terminal of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] See also Figure 1-4 The present invention provides a technical solution: a method for robot to identify obstacles based on human-computer interaction, which specifically includes the following steps:

[0029] S1, take a panoramic photo including the starting point and the end point of the robot, upload the photo to the remote terminal, and upload the photo to the central processing module in the robot through the remote terminal;

[0030] S2. After receiving the photo, the robot plans a straight path between the starting point and the end point through the original path planning module and sends it to the operation decision unit. The operation decision unit will issue a command to the drive module, and the robot starts to move along the straight path;

[0031] S3. When the robot is moving, the panoramic camera module will take real-time video and transmit the video images to the obstacle recognition unit. The obstacle recognition unit will intercept all suspected obstacles in the video, compare the suspected obstacle images with the networked image library, and determine the types of suspected obstacles. The obstacle selection module determines the final obstacle based on the determined types.

[0032] S4, the laser sensor uses laser to measure the distance to the robot according to the final obstacle determined in step S3, and sends it to the operation decision unit through the central processing module;

[0033] S5. After receiving the obstacle information, the operation decision unit will obtain the real-time operation speed of the robot through the speed transmission module, and calculate the time required to reach the obstacle position through the distance calculation module. The path execution calculation module will re-formulate the route according to the obstacle position, so as to bypass the obstacle before the time required to reach the obstacle position. After bypassing, it returns to the original straight path. The execution principle of the path execution calculation module is: the straight path in step S2 is recorded as L, and the obstacle is recorded as Z. The straight path L will divide the obstacle Z into two parts, and the two parts are recorded as Z respectively. 1 and Z2, the distance from the farthest point on the outer edge of Z1 to the straight path L is recorded as A1, and the distance from the farthest point on the outer edge of Z2 to the straight path L is recorded as A2. If A1 is less than A2, the robot chooses to walk on the Z1 side to avoid obstacles. If A1 is greater than A2, the robot chooses to walk on the Z2 side to avoid obstacles. If A1 is equal to A2, the robot walks on the right side to avoid obstacles. The walking path is the straight path between the starting turning point of obstacle avoidance and A1 or A2. If an obstacle is encountered again during walking along this path, the path is replanned according to the above execution principle until the robot returns to the original path;

[0034] S6. The panoramic camera module will send the real-time video to the remote terminal through the central processing module. The remote terminal can view the video in real time through the real-time video viewing module. The remote terminal control personnel can directly control the movement of the robot through the remote control module.

[0035] Furthermore, the obstacle selection module in step S3 includes an obstacle database. After receiving the obstacle type name, the obstacle selection module compares the obstacle type name with the obstacle database, and those that can be successfully matched in the obstacle database are considered as final obstacles.

[0036] Furthermore, the remote terminal and the central processing module are bidirectionally connected, and the output end of the central processing module is connected to the input end of the original path planning module, the output end of the original path planning module is connected to the input end of the operation decision unit, and the output end of the operation decision unit is connected to the input end of the driving module. The remote terminal and the central processing module can be remotely connected via a 3G, 4G, 5G or WIFI network.

[0037] Furthermore, the input end of the panoramic camera module is connected to the input end of the central processing module, and the output end of the panoramic camera module is connected to the input end of the obstacle recognition unit, the output end of the obstacle recognition unit is connected to the input end of the laser sensor, the output end of the laser sensor is connected to the input end of the central processing module, the output end of the central processing module is connected to the input end of the operation decision unit, the panoramic camera module is a 1080P360-degree camera, and the laser sensor model is ZM10x.

[0038] Furthermore, the output end of the driving module is connected to the speed data transmission module, and the output end of the speed data transmission module is connected to the input end of the operation decision unit.

[0039] Furthermore, the remote terminal includes a remote control module and a real-time video viewing module, and the remote terminal can be a mobile phone, a computer, or a tablet computer.

[0040] Furthermore, the operation decision unit includes a distance calculation module and a path execution accounting module.

[0041] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0042] Although 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 the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for robot obstacle recognition based on human-computer interaction, characterized in that: The specific steps include: S1, take a panoramic photo including the starting point and the end point of the robot, upload the photo to the remote terminal, and upload the photo to the central processing module in the robot through the remote terminal; S2. After receiving the photo, the robot plans a straight path between the starting point and the end point through the original path planning module and sends it to the operation decision unit. The operation decision unit will issue a command to the drive module, and the robot starts to move along the straight path; S3. When the robot is moving, the panoramic camera module will take real-time video and transmit the video images to the obstacle recognition unit. The obstacle recognition unit will intercept all suspected obstacles in the video, compare the suspected obstacle images with the networked image library, and determine the types of suspected obstacles. The obstacle selection module determines the final obstacle based on the determined types. S4, the laser sensor uses laser to measure the distance to the robot according to the final obstacle determined in step S3, and sends it to the operation decision unit through the central processing module; S5. After receiving the obstacle information, the operation decision unit will obtain the real-time operation speed of the robot through the speed transmission module, and calculate the time required to reach the obstacle position through the distance calculation module. The path execution calculation module will re-formulate the route according to the obstacle position, so as to bypass the obstacle before the time required to reach the obstacle position, and return to the original straight path after the bypass is completed; S6. The panoramic camera module will send the real-time video to the remote terminal through the central processing module. The remote terminal can view the video in real time through the real-time video viewing module. The remote terminal controller can directly control the robot's movement through the remote control module. The obstacle selection module in step S3 includes an obstacle database. After receiving the obstacle type name, the obstacle selection module will compare the obstacle type name with the obstacle database, and those that can be successfully compared in the obstacle database are considered as final obstacles; The execution principle of the path execution calculation module in step S5 is as follows: the straight path in step S2 is recorded as L, and the obstacle is recorded as Z. The straight path L will divide the obstacle Z into two parts, which are recorded as Z1 and Z2 respectively. The distance from the farthest point of the outer edge of Z1 to the straight path L is recorded as A1, and the distance from the farthest point of the outer edge of Z2 to the straight path L is recorded as A2. If A1 is less than A2, the robot chooses to walk on the Z1 side to avoid obstacles. If A1 is greater than A2, the robot chooses to walk on the Z2 side to avoid obstacles. If A1 is equal to A2, the robot walks on the right side to avoid obstacles. The walking path is a straight path between the starting turning point of obstacle avoidance and A1 or A2. If an obstacle is encountered again during walking along the path, the path is replanned according to the above execution principle until the robot returns to the original path.

2. The method for robot obstacle recognition based on human-computer interaction according to claim 1, characterized in that: The remote terminal is bidirectionally connected to the central processing module, and the output end of the central processing module is connected to the input end of the original path planning module, the output end of the original path planning module is connected to the input end of the operation decision unit, and the output end of the operation decision unit is connected to the input end of the driving module.

3. The method for robot obstacle recognition based on human-computer interaction according to claim 1, characterized in that: The input end of the panoramic camera module is connected to the input end of the central processing module, and the output end of the panoramic camera module is connected to the input end of the obstacle recognition unit, the output end of the obstacle recognition unit is connected to the input end of the laser sensor, the output end of the laser sensor is connected to the input end of the central processing module, and the output end of the central processing module is connected to the input end of the operation decision unit.

4. The method for robot obstacle recognition based on human-machine interaction according to claim 1, characterized in that: The output end of the driving module is connected to the speed data transmission module, and the output end of the speed data transmission module is connected to the input end of the operation decision unit.

5. The method for robot obstacle recognition based on human-machine interaction according to claim 1, characterized in that: The remote terminal includes a remote control module and a real-time video viewing module.

6. The method for robot obstacle recognition based on human-machine interaction according to claim 1, characterized in that: The operation decision unit includes a distance calculation module and a path execution accounting module.

Images