Self-adaptive indoor positioning and path optimization method based on artificial intelligence

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a self-adaptive indoor positioning and path optimization method based on artificial intelligence, which comprises the following steps: S1, collecting visual positioning data, deploying the position and angle of a camera to cover an indoor area, then collecting image data of an indoor environment in real time, and setting the position and angle of the camera to be the same as the position and angle of the camera; calculating and determining the positions of the target object and other obstacles in the image through visual artificial intelligence; s2, collecting wireless Bluetooth positioning data; s3, positioning data fusion and evaluation; s4, target setting and path planning; s5, executing and monitoring a path; and S6, dynamic path optimization. Visual artificial intelligence and wireless Bluetooth positioning are combined to improve positioning accuracy and stability, visual positioning provides abundant image information, wireless Bluetooth positioning ensures coverage and stability, and matching of the visual positioning and the wireless Bluetooth positioning can correct errors, adapt to different environment changes and provide an accurate position basis for path optimization. The robot navigation efficiency and the task execution success rate are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an adaptive indoor positioning and path optimization method based on artificial intelligence. Background Art

[0002] With the rapid development of science and technology, robots are used more and more widely in indoor environments, from automated handling robots in industrial production to guided robots in the service industry, to cleaning robots in the home, etc. They play an important role in different scenarios. However, in order for these robots to complete their tasks efficiently, accurate indoor positioning and optimized path planning are the key. Traditional indoor positioning methods usually use infrared or ultrasonic positioning technology in robot applications, which provide relatively rough location information in complex indoor layouts. Therefore, adaptive indoor positioning and path optimization methods based on artificial intelligence are needed to accurately locate indoor positions and plan mobile paths.

[0003] However, with current technology, infrared or ultrasonic positioning technology is easily affected by the environment and obstacles. Its signals may be reflected, absorbed or scattered by obstacles, resulting in inaccurate positioning and lack of adaptability to dynamic environments, affecting the robot's navigation efficiency and task execution success rate. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides an adaptive indoor positioning and path optimization method based on artificial intelligence to solve the problems of inaccurate positioning and lack of adaptability to dynamic environments.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an adaptive indoor positioning and path optimization method based on artificial intelligence, comprising the following steps:

[0006] S1: Visual positioning data acquisition, deploying the position and angle of the camera to cover the indoor area, and then collecting image data of the indoor environment in real time, calculating and determining the position of the target object and other obstacles in the image through visual artificial intelligence, and converting it into actual coordinate values;

[0007] S2: Wireless Bluetooth positioning data collection. Multiple Bluetooth beacons are set up indoors. The robot receives the signal from the Bluetooth beacon, measures the signal strength and arrival time parameters, calculates the distance between the target object and the Bluetooth beacon, and uses the triangulation positioning algorithm to determine the position of the target object and obtain its coordinate value.

[0008] S3: Positioning data fusion and evaluation, visual positioning data and wireless Bluetooth positioning data are matched to determine whether they overlap. If they match, they are bound to locate the precise target position. If not, the visual positioning and wireless Bluetooth positioning data are re-collected until the positions match each other;

[0009] S4: Target setting and path planning, determine the target location, use the path planning algorithm to plan a path from the starting point to the target point, and calculate the optimal path through stationary obstacles;

[0010] S5: Path execution and monitoring, executing the robot's running path to the target, and monitoring the robot's position and obstacles moving in front of it in real time;

[0011] S6: Dynamic path optimization. When the robot is running, if someone is walking and blocking the robot's path, the robot needs to avoid the obstacle and replan the path.

[0012] Preferably, the cameras in S1 are divided into four, covering multiple directions in the room, and the acquisition frequency is 30-60 frames per second. The coordinates of the target object and other obstacle objects in the image are calculated by a computer with the indoor corner as the center, and the coordinates are in a rectangular coordinate system.

[0013] Preferably, the S2 indoor Bluetooth beacon is set to three, denoted as A, B, and C. The robot is equipped with a Bluetooth receiver, which can receive the signal strength from the surrounding Bluetooth beacons, and establish an indoor coordinate system with the indoor corner as the center. The triangulation positioning algorithm includes trilateration, and the position coordinates of the robot are (x, y) and are solved by the following set of equations:

[0014] (x-x1) 2 +(y-y1) 2 =d1 2

[0015] (x-x2) 2 +(y-y2) 2 =d2 2

[0016] (x-x3) 2 +(y-y3) 2 =d3 2

[0017] The distances of the three Bluetooth beacons A, B, and C are d1, d2, and d3 respectively;

[0018] The coordinates of beacons A, B, and C are (x1, y1), (x2, y2), and (x3, y3), respectively.

[0019] Preferably, the visual positioning data in S3 and the target position coordinates of the wireless Bluetooth positioning are in the same coordinate system, and the position threshold is 0.1-0.2 meters. If it exceeds, there is no match, and the data is re-collected for matching. If the distance deviation is within the range, matching and binding are performed, and a unique identifier is assigned to the target object, and the coordinate values ​​of the visual positioning and wireless Bluetooth positioning are associated with the identifier.

[0020] Preferably, in S4, information is obtained through wireless Bluetooth and visual signals to assist in confirming the target location, and the optimal path is calculated through the A* algorithm. First, two lists are created: an open list and a closed list, and the starting point coordinates are added to the open list, and the g value of the starting point is set to 0, and then the Euclidean distance from the starting point to the target point is calculated as the heuristic function value h, and The two nodes are A(x1,y1) and B(x2,y2). Path search: expand through the adjacent coordinate nodes in the four directions of the target position in the coordinate system, and skip if the adjacent node is an obstacle. Through backtracking, select the node with the smallest path f to the target node as the next node target to be expanded. f is the sum of the heuristic function value d and the g value.

[0021] Preferably, in S5, the optimal path is output in the form of a coordinate sequence, indicating the positions that the robot passes through in sequence, and the moving obstacles are monitored in real time through a camera.

[0022] Preferably, in S6, there is a new obstacle in the moving path of the robot, resulting in a change in the environment, and the A* path planning algorithm is re-run to update the optimal path.

[0023] The present invention provides an adaptive indoor positioning and path optimization method based on artificial intelligence.

[0024] It has the following beneficial effects:

[0025] 1. The present invention combines visual artificial intelligence with wireless Bluetooth positioning to improve positioning accuracy and stability. Visual positioning provides rich image information, and wireless Bluetooth positioning ensures coverage and stability. The matching of the two can correct errors, adapt to different environmental changes, provide a precise position basis for path optimization, and improve robot navigation efficiency and task execution success rate.

[0026] 2. The present invention combines wireless Bluetooth and visual signals with the A* algorithm for path planning. The wireless Bluetooth and visual signals complement each other and accurately determine the position to provide reliable data for the A* algorithm. The A* algorithm can efficiently plan the path and avoid obstacles to ensure that the robot reaches the target quickly and accurately, improve navigation efficiency and task execution success rate, and enhance system stability.

[0027] 3. In the present invention, when the robot encounters a new obstacle while running the optimal path, the A* path planning algorithm can be re-run to quickly adapt to environmental changes, ensuring that the robot adjusts the path in time, effectively avoids new obstacles, improves movement efficiency and safety, and at the same time enhances adaptability in dynamic environments, ensuring the smooth progress of tasks, and improving operational reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The present invention is a flowchart of the adaptive indoor positioning and path optimization method based on artificial intelligence. DETAILED DESCRIPTION

[0029] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings 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.

[0030] Please refer to the attached Figure 1 The embodiment of the present invention provides an adaptive indoor positioning and path optimization method based on artificial intelligence, comprising the following steps:

[0031] S1: Visual positioning data acquisition, deploying the position and angle of the camera to cover the indoor area, and then collecting image data of the indoor environment in real time, calculating and determining the position of the target object and other obstacles in the image through visual artificial intelligence, and converting it into actual coordinate values;

[0032] S2: Wireless Bluetooth positioning data collection. Multiple Bluetooth beacons are set up indoors. The robot receives the signal from the Bluetooth beacon, measures the signal strength and arrival time parameters, calculates the distance between the target object and the Bluetooth beacon, and uses the triangulation positioning algorithm to determine the position of the target object and obtain its coordinate value.

[0033] S3: Positioning data fusion and evaluation, visual positioning data and wireless Bluetooth positioning data are matched to determine whether they overlap. If they match, they are bound to locate the precise target position. If not, the visual positioning and wireless Bluetooth positioning data are re-collected until the positions match each other;

[0034] S4: Target setting and path planning, determine the target location, use the path planning algorithm to plan a path from the starting point to the target point, and calculate the optimal path through stationary obstacles;

[0035] S5: Path execution and monitoring, executing the robot's running path to the target, and monitoring the robot's position and obstacles moving in front of it in real time;

[0036] S6: Dynamic path optimization. When the robot is running, if someone is walking and blocking the robot's path, the robot needs to avoid the obstacle and replan the path.

[0037] There are four cameras in S1, covering multiple directions in the room, and the acquisition frequency is 30-60 frames per second. The computer calculates the coordinates of the target object and other obstacles in the image with the indoor corner as the center. The coordinates are in a rectangular coordinate system.

[0038] Specifically, four cameras cover the room from different directions, which can reduce blind spots, observe various areas of the room more comprehensively, and improve the accuracy and completeness of target object and obstacle detection. At the same time, the acquisition frequency of 30-60 frames per second can quickly capture changes in the indoor environment, and can detect and track moving target objects and obstacles in a timely manner to ensure real-time positioning. Then, a rectangular coordinate system centered on the indoor corner is established to provide a unified reference frame for the image data collected by all cameras, which is convenient for the calculation and fusion of coordinate values, as well as for path planning and navigation.

[0039] S2 sets three indoor Bluetooth beacons as A, B, and C. The robot is equipped with a Bluetooth receiver to receive the signal strength from the surrounding Bluetooth beacons. An indoor coordinate system is established with the indoor corner as the center. The triangulation positioning algorithm includes trilateral measurement. The position coordinates of the robot are (x, y) and solved by the following equations:

[0040] (x-x1) 2 +(y-y1) 2 =d1 2

[0041] (x-x2) 2 +(y-y2) 2 =d2 2

[0042] (x-x3) 2 +(y-y3) 2 =d3 2

[0043] The distances of the three Bluetooth beacons A, B, and C are d1, d2, and d3 respectively;

[0044] The coordinates of beacons A, B, and C are (x1, y1), (x2, y2), and (x3, y3), respectively.

[0045] Specifically, three Bluetooth beacons send signals from different positions to provide multi-angle positioning information, allowing the robot to receive signal strength information from different directions, increasing the accuracy and reliability of positioning. At the same time, the triangulation positioning algorithm is used to determine the robot's position coordinates, and commands can be issued to the robot through wireless Bluetooth. Establishing an indoor coordinate system can provide a unified reference system for the robot's position and the Bluetooth beacon's position, facilitating coordinate calculation and positioning.

[0046] The visual positioning data in S3 and the target position coordinates of wireless Bluetooth positioning are in the same coordinate system, and the position threshold is 0.1-0.2 meters. If it exceeds the threshold, it will not match, and the data will be collected again for matching. If the distance deviation is within the range, the matching and binding will be performed. A unique identifier will be assigned to the target object, and the coordinate values ​​of the visual positioning and wireless Bluetooth positioning will be associated with the identifier.

[0047] Specifically, ensuring that the visual positioning data and the wireless Bluetooth positioning data are in the same coordinate system facilitates direct comparison and fusion of the two positioning results, avoiding conversion errors and complexity caused by different coordinate systems, and by setting a position threshold of 0.1-0.2 meters, the judgment criteria for whether the visual positioning and wireless Bluetooth positioning data match are clarified, which helps to screen out reliable positioning data to improve positioning accuracy. After binding, a unique identifier is assigned to the target object, and the coordinate values ​​of the two positioning are associated with the identifier, which facilitates subsequent data management and use. The mutual verification of the two positioning methods improves the stability and reliability of the system and facilitates the tracking and analysis of the target object.

[0048] In S4, information is obtained through wireless Bluetooth and visual signals to assist in confirming the target location, and the optimal path is calculated through the A* algorithm. First, two lists are created: an open list and a closed list, and the starting point coordinates are added to the open list, the g value of the starting point is set to 0, and then the Euclidean distance from the starting point to the target point is calculated as the heuristic function value h, and The two nodes are A(x1,y1) and B(x2,y2). Path search: expand through the adjacent coordinate nodes in the four directions of the target position in the coordinate system, and skip if the adjacent node is an obstacle. Through backtracking, select the node with the smallest path f to the target node as the next node target to be expanded. f is the sum of the heuristic function value d and the g value.

[0049] Specifically, the A* algorithm combines the heuristic function and the actual cost g value for path search, which can improve the search efficiency while ensuring the optimal path is found, and by giving priority to more promising nodes for expansion, it reduces unnecessary search steps. At the same time, considering the existence of obstacles, it can plan a reasonable path in an environment with obstacles. The algorithm can automatically avoid obstacles, find a feasible path, and improve the robot's mobility, so that it can navigate to the target location more accurately. It reduces the possibility of path deviation and erroneous navigation, and improves the success rate of task execution.

[0050] In S5, the optimal path is output in the form of a coordinate sequence, which indicates the positions that the robot passes through in sequence, while the moving obstacles are monitored in real time through the camera.

[0051] Specifically, the output of the coordinate sequence enables the robot to accurately know its position and target position, achieve precise navigation, reduce the possibility of path deviation and getting lost, and improve the robot's mobility efficiency. At the same time, the real-time monitoring of the camera enables the robot to quickly respond to environmental changes, avoid stagnation or incorrect navigation due to obstacles, and ensure the smooth progress of the task.

[0052] In S6, there are new obstacles in the robot's moving path, causing the environment to change. The A* path planning algorithm is rerun to update the optimal path.

[0053] Specifically, when new obstacles appear on the robot's moving path, the environment changes and the original optimal path may no longer be applicable. Re-running the A* path planning algorithm can recalculate the optimal path based on the new environmental conditions, ensuring that the robot can effectively avoid new obstacles and continue to move towards its goal, thereby improving the robot's adaptability and flexibility, enabling it to continue to perform tasks in a dynamic environment, and enhancing the system's adaptability and reliability.

[0054] 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. Adaptive indoor positioning and path optimization method based on artificial intelligence, It is characterized in that The following steps are involved: S1: Visual positioning data acquisition, deploying the position and angle of the camera to cover the indoor area, and then collecting image data of the indoor environment in real time, calculating and determining the position of the target object and other obstacles in the image through visual artificial intelligence, and converting it into actual coordinate values; S2: Wireless Bluetooth positioning data collection. Multiple Bluetooth beacons are set up indoors. The robot receives the signal from the Bluetooth beacon, measures the signal strength and arrival time parameters, calculates the distance between the target object and the Bluetooth beacon, and uses the triangulation positioning algorithm to determine the position of the target object and obtain its coordinate value. S3: Positioning data fusion and evaluation, visual positioning data and wireless Bluetooth positioning data are matched to determine whether they overlap. If they match, they are bound to locate the precise target position. If not, the visual positioning and wireless Bluetooth positioning data are re-collected until the positions match each other; S4: Target setting and path planning, determine the target location, use the path planning algorithm to plan a path from the starting point to the target point, and calculate the optimal path through stationary obstacles; S5: Path execution and monitoring, executing the robot's running path to the target, and monitoring the robot's position and obstacles moving in front of it in real time; S6: Dynamic path optimization. When the robot is running, if someone is walking and blocking the robot's path, the robot needs to avoid the obstacle and replan the path.

2. The method for adaptive indoor positioning and path optimization based on artificial intelligence according to claim 1, It is characterized in that There are four cameras in S1, which cover multiple directions in the room, and the acquisition frequency is 30-60 frames per second. The coordinates of the target object and other obstacles in the image are calculated by a computer with the indoor corner as the center, and the coordinates are in a rectangular coordinate system.

3. The method for adaptive indoor positioning and path optimization based on artificial intelligence according to claim 1, It is characterized in that The S2 indoor Bluetooth beacon is set to three, denoted as A, B, and C. The robot is equipped with a Bluetooth receiver, which can receive the signal strength from the surrounding Bluetooth beacons. An indoor coordinate system is established with the indoor corner as the center. The triangulation positioning algorithm includes trilateral measurement. The position coordinates of the robot are (x, y) and solved by the following equations: (x-x1) 2 +(y-y1) 2 =d1 2 (x-x2) 2 +(y-y2) 2 =d2 2 <h2 style=";text-align:left;direction:ltr">(x-x3)<h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> +(y-y3)<h2 style=";text-align:left;direction:ltr"> 2 <h2 style=";text-align:left;direction:ltr"> =d3<h2 style=";text-align:left;direction:ltr"> 2 The distances of the three Bluetooth beacons A, B, and C are d1, d2, and d3 respectively; The coordinates of beacons A, B, and C are (x1, y1), (x2, y2), and (x3, y3), respectively.

4. The method for adaptive indoor positioning and path optimization based on artificial intelligence according to claim 1, It is characterized in that The visual positioning data in S3 and the target position coordinates of wireless Bluetooth positioning are in the same coordinate system, and the position threshold is 0.1-0.2 meters. If it exceeds, there is no match, and the data is re-collected for matching. If the distance deviation is within the range, matching and binding are performed, and a unique identifier is assigned to the target object, and the coordinate values ​​of the visual positioning and wireless Bluetooth positioning are associated with the identifier.

5. The method for adaptive indoor positioning and path optimization based on artificial intelligence according to claim 1, It is characterized in that In S4, information is obtained through wireless Bluetooth and visual signals to assist in confirming the target location, and the optimal path is calculated through the A* algorithm. First, two lists are created: an open list and a closed list, and the starting point coordinates are added to the open list, the g value of the starting point is set to 0, and then the Euclidean distance from the starting point to the target point is calculated as the heuristic function value h, and Two of the nodes are A(x 1 ,y 1 ) and B(x 2 ,y 2 ), Path search: Expand through the adjacent coordinate nodes in the four directions of the target position in the coordinate system, and skip if the adjacent node is an obstacle. By backtracking, select the node with the smallest path f to the target node as the next node target to be expanded, where f is the sum of the heuristic function value d and the g value.

6. The method for adaptive indoor positioning and path optimization based on artificial intelligence according to claim 1, It is characterized in that In S5, the optimal path is output in the form of a coordinate sequence, indicating the positions that the robot passes through in sequence, while the moving obstacles are monitored in real time through the camera.

7. The method for adaptive indoor positioning and path optimization based on artificial intelligence according to claim 1, It is characterized in that In S6, there are new obstacles in the robot's moving path, which causes the environment to change. The A* path planning algorithm is re-run to update the optimal path.

Images