Robot autonomous obstacle avoidance method based on dynamic environment and related device

Abstract

The invention discloses a robot autonomous obstacle avoidance method based on a dynamic environment and a related device, and the method comprises the steps: a mobile robot receives an operation task in an operation region map, and determines an operation end point position of the operation task; path planning processing is carried out based on the current position and the operation end point position, and a planned operation path is generated; when the mobile robot executes the operation task, the corresponding positions and movement trends of the multiple moving obstacles are obtained; inputting the corresponding positions and motion trends of the plurality of moving obstacles into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining corresponding predicted trajectory data of the plurality of obstacles at the next moment; and performing autonomous obstacle avoidance adjustment control at the next moment when the mobile robot moves on the planned operation path based on the predicted trajectory data corresponding to the plurality of obstacles at the next moment. According to the embodiment of the invention, autonomous obstacle avoidance control in a complex working environment is realized, and the working safety is ensured.

Description

Technical Field

[0001] The present invention relates to the field of robot control technology, and in particular to a robot autonomous obstacle avoidance method based on a dynamic environment and a related device. Background Art

[0002] With the maturity of mobile robot technology, more and more mobile robots are used to perform some work tasks in relatively simple environments; however, in more complex working environments, there may be multiple mobile robots working together at the same time, and there will also be corresponding staff to synchronize other operations in the environment, and there may also be some irregular estimated obstacles in some areas; at this time, the existing mobile robots may not be able to achieve accurate active obstacle avoidance control when operating in such a complex environment, resulting in the inability to successfully complete the work tasks. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides a robot autonomous obstacle avoidance method and related devices based on a dynamic environment, which can realize autonomous and precise obstacle avoidance control in a complex working environment, successfully complete the working task, and ensure the safety of the operation.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a robot autonomous obstacle avoidance method in a dynamic environment, the method comprising:

[0005] The mobile robot receives a work task in a work area map and determines a work end position of the work task, wherein the work area map is marked with the position, shape and size of fixed obstacles;

[0006] Performing path planning based on the current position of the mobile robot and the operation end position to generate a planned operation path for the mobile robot within the operation area map;

[0007] When the mobile robot performs the operation task according to the planned operation path, positions and movement trends corresponding to a plurality of moving obstacles are obtained based on a sensor device provided on the mobile robot, wherein the movement trend includes a movement speed and a movement direction;

[0008] Inputting the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining the predicted trajectory data corresponding to the multiple obstacles at the next moment;

[0009] Based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned working path.

[0010] Optionally, the mobile robot receives a work task in a work area map and determines a work end position of the work task, including:

[0011] The mobile robot uses the robot ID to poll the job task list on the job task server at a preset interval. When a job task matching the robot ID is polled, the job task matching the robot ID is loaded into the local storage of the mobile robot. The job task list is formed by the job task server assigning and updating the job tasks according to preset rules after the control user uploads the job tasks to the job task server.

[0012] The mobile robot reads the locally stored operation task, analyzes the read operation task, and determines the operation end position of the operation task.

[0013] Optionally, performing path planning based on the current position of the mobile robot and the operation end position to generate a planned operation path of the mobile robot within the operation area map includes:

[0014] The mobile robot starts the positioning device to perform positioning processing on the current position to obtain the current position of the mobile robot;

[0015] Based on the current position and the operation end position, the improved A-Star algorithm is used to perform path planning processing according to the shortest operation path to generate a planned operation path for the mobile robot within the operation area map.

[0016] Optionally, the obtaining of positions and movement trends corresponding to a plurality of moving obstacles based on a sensor device arranged on the mobile robot includes:

[0017] Starting the laser sensor and the visual sensor provided on the mobile robot to collect and process data of the surrounding environment of the mobile robot, and obtaining laser point cloud data and visual image data;

[0018] The laser point cloud data and the visual image data are processed based on a multi-sensor fusion algorithm to obtain positions and movement trends corresponding to multiple moving obstacles.

[0019] Optionally, the laser point cloud data and the visual image data are processed based on a multi-sensor fusion algorithm to obtain positions and movement trends corresponding to a plurality of moving obstacles, including:

[0020] The laser point cloud data is subjected to filtering and noise reduction processing by using a preset filter, and after filtering and noise reduction, smoothing and hole filling processing are performed in sequence to form initially processed laser point cloud data;

[0021] Clustering is performed on the laser point cloud data after the initial processing, and a three-dimensional reconstruction algorithm is used to perform three-dimensional reconstruction of obstacles in the laser point cloud data according to the clustering results to obtain a number of three-dimensional reconstructed obstacles;

[0022] Performing edge detection processing on the visual image data, and performing extraction processing on a plurality of obstacles based on the edge detection result to obtain a plurality of extracted obstacles;

[0023] Performing obstacle fusion processing on the plurality of reconstructed obstacles and the plurality of extracted obstacles at the same angle to determine a plurality of obstacles around the mobile robot;

[0024] Based on the laser sensor, several obstacles around the mobile robot are positioned and tracked to obtain the positions and movement trends of the multiple moving obstacles around the mobile robot.

[0025] Optionally, the step of inputting the positions and movement trends corresponding to the multiple moving obstacles into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining predicted trajectory data corresponding to the multiple obstacles at the next moment, includes:

[0026] Indexing corresponding obstacle categories in a database according to the contour information corresponding to the plurality of moving obstacles, and marking the obstacle categories on the corresponding plurality of moving obstacles to form a plurality of marked moving obstacles;

[0027] Multiple marked moving obstacles, their corresponding positions and movement trends are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, so as to obtain the predicted trajectory data corresponding to the multiple obstacles at the next moment; wherein the obstacle trajectory prediction model is a model established and trained and converged by a radial basis function neural network.

[0028] Optionally, the autonomous obstacle avoidance adjustment control of the mobile robot at the next moment when it moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment includes:

[0029] Inputting the predicted trajectory data corresponding to the multiple obstacles at the next moment into the built-in behavior decision module of the mobile robot for decision processing, and generating the corresponding decision instructions at the next moment when moving on the planned working path according to the decision result;

[0030] Based on the decision instruction, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned working path.

[0031] In addition, an embodiment of the present invention further provides a robot autonomous obstacle avoidance device in a dynamic environment, the device comprising:

[0032] Determination module: used for the mobile robot to receive the operation task in the operation area map and determine the operation end position of the operation task, and the operation area map is marked with the position, shape and size of the fixed obstacles;

[0033] Planning module: used for performing path planning processing based on the current position of the mobile robot and the operation end position, and generating a planned operation path of the mobile robot within the operation area map;

[0034] An acquisition module: used for acquiring positions and movement trends corresponding to a plurality of moving obstacles based on a sensor device arranged on the mobile robot when the mobile robot performs the operation task according to the planned operation path, wherein the movement trend includes a movement speed and a movement direction;

[0035] Prediction module: used for inputting the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining the predicted trajectory data corresponding to the multiple obstacles at the next moment;

[0036] Adjustment control module: used for performing autonomous obstacle avoidance adjustment control on the mobile robot at the next moment when it moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment.

[0037] In addition, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the processor runs a computer program or code stored in the memory to implement the robot autonomous obstacle avoidance method as described in any one of the above.

[0038] In addition, an embodiment of the present invention further provides a computer-readable storage medium for storing a computer program or code. When the computer program or code is executed by a processor, the robot autonomous obstacle avoidance method as described in any one of the above is implemented.

[0039] In an embodiment of the present invention, a work task in a work area map is received by a mobile robot, and the work end position of the work task is determined; path planning processing is performed based on the current position and the work end position to generate a planned work path; when the mobile robot executes the work task, the positions and motion trends corresponding to multiple moving obstacles are obtained; the positions and motion trends corresponding to the multiple moving obstacles are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and the predicted trajectory data corresponding to the multiple obstacles at the next moment are obtained; based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned work path; autonomous and precise obstacle avoidance control is achieved in a complex working environment, and the work task is successfully completed to ensure work safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 is a flow chart of a robot autonomous obstacle avoidance method in a dynamic environment in an embodiment of the present invention;

[0042] Figure 2 is a flow chart of a robot autonomous obstacle avoidance method in a dynamic environment in another embodiment of the present invention;

[0043] Figure 3 Schematic diagram of the structure of a robot autonomous obstacle avoidance device in a dynamic environment according to an embodiment of the present invention;

[0044] Figure 4 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] 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.

[0046] For example, see Figure 1 , Figure 1 It is a flow chart of a robot autonomous obstacle avoidance method in a dynamic environment in an embodiment of the present invention.

[0047] like Figure 1 As shown, a robot autonomous obstacle avoidance method in a dynamic environment, the method comprising:

[0048] S101: The mobile robot receives a work task in a work area map and determines a work end position of the work task, wherein the work area map is marked with the position, shape and size of fixed obstacles;

[0049] In the specific implementation process of the present invention, the mobile robot receives the work tasks in the work area map and determines the work end position of the work tasks, including: the mobile robot uses the robot ID to poll the work task list on the work task server at a preset interval, and when the work task matching the robot ID is polled, the work task matching the robot ID is loaded into the local storage of the mobile robot, and the work task list is formed by the work task server according to preset rules after the control user uploads the work tasks to the work task server; the mobile robot reads the locally stored work tasks, and parses the read work tasks to determine the work end position of the work tasks.

[0050] Specifically, first, the mobile robot in the working area will poll the corresponding working task on the working task server according to the preset interval, that is, poll the working task list on the working task server through the robot ID corresponding to the mobile robot, and after polling the working task matching the robot ID, through the corresponding instructions, load the working task matching the robot ID to the local storage of the mobile robot, and then the mobile robot will read the corresponding working task in the local storage, and then perform corresponding analysis on the working task to obtain the working end position of the working task.

[0051] After receiving the job task uploaded by the control user, the job task server will assign a corresponding mobile robot to the job task to execute the job task; when allocating, it will first be based on the idle time and the distance to the location of the job task, that is, the mobile robot closest to the location of the job task is selected from the currently idle mobile robots as the mobile robot executing the job task, and the robot ID of the mobile robot is bound to the job task, and updated to the job task list in the job task server; the location, shape and size of fixed obstacles are marked in the work area map, so that it is easier to identify those mobile obstacles that do not move for a short time during tracking later.

[0052] S102: performing path planning processing based on the current position of the mobile robot and the operation end position to generate a planned operation path of the mobile robot within the operation area map;

[0053] In the specific implementation process of the present invention, the path planning processing is performed based on the current position of the mobile robot and the operation end position to generate the planned operation path of the mobile robot in the operation area map, including: the mobile robot starts the positioning device to perform positioning processing on the current position to obtain the current position of the mobile robot; based on the current position and the operation end position, the improved A-Star algorithm is used to perform path planning processing according to the shortest operation path to generate the planned operation path of the mobile robot in the operation area map.

[0054] Specifically, in the planning of the working path, an improved A-Star algorithm is adopted, through which the shortest working path (or optimal working path) can be planned quickly and accurately; that is, it is first necessary to start the positioning device carried by the mobile robot itself to position the current position, so as to obtain the current position of the mobile robot; then, according to the current position and the end position of the work, the improved A-Star algorithm is used to perform path planning processing according to the shortest working path, and finally a planned working path of the mobile robot in the map of the working area can be generated; by performing path planning with the improved A-Star algorithm, paths that do not meet the requirements can be accurately and quickly removed, thereby achieving rapid path planning convergence and accurately planning the required working path.

[0055] S103: when the mobile robot performs the operation task according to the planned operation path, obtaining positions and movement trends corresponding to a plurality of moving obstacles based on a sensor device provided on the mobile robot, wherein the movement trend includes a movement speed and a movement direction;

[0056] In the specific implementation process of the present invention, the positions and movement trends corresponding to multiple moving obstacles are obtained based on the sensor equipment set on the mobile robot, including: starting the laser sensor and the visual sensor set on the mobile robot to collect and process the data of the surrounding environment of the mobile robot, and obtain laser point cloud data and visual image data; based on the multi-sensor fusion algorithm, the laser point cloud data and the visual image data are processed to obtain the positions and movement trends corresponding to multiple moving obstacles.

[0057] Furthermore, the laser point cloud data and the visual image data are processed based on a multi-sensor fusion algorithm to obtain the positions and movement trends corresponding to multiple moving obstacles, including: filtering and denoising the laser point cloud data using a preset filter, and after filtering and denoising, smoothing and hole filling processing are performed in sequence to form initially processed laser point cloud data; clustering the initially processed laser point cloud data, and performing three-dimensional reconstruction of obstacles in the laser point cloud data using a three-dimensional reconstruction algorithm according to the clustering results to obtain a number of three-dimensional reconstructed obstacles; edge detection processing is performed on the visual image data, and a number of obstacles are extracted based on the edge detection results to obtain a number of extracted obstacles; obstacle fusion processing is performed on the several reconstructed obstacles and the several extracted obstacles at the same angle to determine a number of obstacles around the mobile robot; positioning and tracking the several obstacles around the mobile robot based on the laser sensor to obtain the positions and movement trends corresponding to the multiple moving obstacles around the mobile robot.

[0058] Specifically, in this solution, obstacles are identified and tracked through the sensor equipment carried by the mobile robot, so as to obtain the positions and movement trends corresponding to multiple moving obstacles; and the sensor equipment is mainly laser sensors and visual sensors, that is, the surrounding environment of the mobile robot is collected and processed by laser sensors and visual sensors, so as to obtain laser point cloud data and visual image data; then the laser point cloud data and visual image data are processed by a multi-sensor fusion algorithm, so as to extract the positions and movement trends corresponding to multiple moving obstacles, wherein the movement trend includes movement speed and movement direction.

[0059] First, the laser point cloud data is filtered and denoised using a preset filter, and after filtering and denoising, it is smoothed and hole-filled in turn to form the initially processed laser point cloud data; then the initially processed laser point cloud data is clustered, and according to the clustering results, the 3D reconstruction algorithm is used to perform 3D reconstruction of obstacles in the laser point cloud data to obtain several 3D reconstructed obstacles; the several 3D reconstructed obstacles here include fixed obstacles marked in the work area map; then the color space of the collected visual image data is converted to realize the conversion of the visual image data into an HSV image, and The HSV image is processed for extracting the information of the H channel, so as to obtain the H channel image corresponding to the HSV image; the H channel image is processed for edge detection by using an edge detection algorithm, so as to extract the edge detection result; and several obstacles are extracted from the edge detection result; several obstacles around the mobile robot are determined by performing obstacle fusion processing on several reconstructed obstacles and several extracted obstacles at the same angle; then these several obstacles are used to match the positions, shapes and sizes of the fixed obstacles marked in the work area map one by one, and the suspected fixed obstacles among the several obstacles are marked by matching.

[0060] Then, the laser sensor is controlled to locate and track several obstacles around the mobile robot, and the suspected fixed obstacles among the obstacles are marked according to the positioning and tracking results, so as to determine the corresponding positions and movement trends of multiple moving obstacles around the mobile robot.

[0061] S104: inputting the positions and movement trends corresponding to the multiple moving obstacles into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining predicted trajectory data corresponding to the multiple obstacles at the next moment;

[0062] In a specific implementation of the present invention, the positions and movement trends corresponding to the multiple moving obstacles are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtain predicted trajectory data corresponding to the multiple obstacles at the next moment, including: indexing the corresponding obstacle category in a database according to the contour information corresponding to the multiple moving obstacles, and marking the obstacle category on the corresponding multiple moving obstacles to form a plurality of marked mobile obstacles; inputting the multiple marked mobile obstacles and the positions and movement trends corresponding to the multiple marked mobile obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtain predicted trajectory data corresponding to the multiple obstacles at the next moment; wherein the obstacle trajectory prediction model is a model established and trained and converged by a radial basis function neural network.

[0063] Specifically, firstly, the corresponding obstacle categories are indexed in the database according to the contour information corresponding to the multiple moving obstacles, and the obstacle categories are marked on the corresponding multiple moving obstacles to form multiple marked moving obstacles; then, the multiple marked moving obstacles, the positions corresponding to the multiple marked moving obstacles, and the motion trends are input into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and the predicted trajectory data corresponding to the multiple obstacles at the next moment are obtained; wherein, the obstacle trajectory prediction model is a model established and trained and converged by a radial basis function neural network.

[0064] The structure of the radial basis function neural network is similar to that of a multi-layer forward network, which is a three-layer forward network; the input layer is composed of signal source nodes; the second layer is the hidden layer, and the number of hidden units depends on the needs of the problem being described. The transformation function of the hidden unit is the RBF radial basis function, which is a non-negative nonlinear function that is radially symmetric and decays about the center point; the third layer is the output layer, which responds to the action of the input pattern; the transformation from the input space to the hidden layer space is nonlinear, while the transformation from the hidden layer space to the output layer space is linear; and it has high accuracy when applied to trajectory prediction of moving objects.

[0065] S105: Based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when the mobile robot moves on the planned working path.

[0066] In the specific implementation process of the present invention, the autonomous obstacle avoidance adjustment control of the mobile robot at the next moment when it moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment includes: inputting the predicted trajectory data corresponding to the multiple obstacles at the next moment into the built-in behavior decision module of the mobile robot for decision processing, and generating a decision instruction corresponding to the next moment when moving on the planned working path according to the decision result; and autonomous obstacle avoidance adjustment control of the mobile robot at the next moment when it moves on the planned working path based on the decision instruction.

[0067] In the specific implementation process of the present invention, the predicted trajectory data corresponding to multiple obstacles at the next moment are input into the built-in behavior decision module of the mobile robot for decision processing, and then the corresponding decision instructions for the next moment when moving on the planned working path are generated according to the decision results; in the present technical solution, the behavior decision model is a decision model formed by adjusting the parameters of the PID controller according to the actual situation; it can give corresponding feedback decisions according to the specific situation.

[0068] When receiving the decision instruction, the mobile robot will realize the autonomous obstacle avoidance adjustment control at the next moment when it moves on the planned working path according to the decision instruction.

[0069] In an embodiment of the present invention, a work task in a work area map is received by a mobile robot, and the work end position of the work task is determined; path planning processing is performed based on the current position and the work end position to generate a planned work path; when the mobile robot executes the work task, the positions and motion trends corresponding to multiple moving obstacles are obtained; the positions and motion trends corresponding to the multiple moving obstacles are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and the predicted trajectory data corresponding to the multiple obstacles at the next moment are obtained; based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned work path; autonomous and precise obstacle avoidance control is achieved in a complex working environment, and the work task is successfully completed to ensure work safety.

[0070] For example 2, please refer to Figure 2 , Figure 2 It is a flowchart of a robot autonomous obstacle avoidance method in a dynamic environment in another embodiment of the present invention.

[0071] like Figure 2 As shown, a robot autonomous obstacle avoidance method in a dynamic environment, the method comprising:

[0072] S201: The mobile robot receives a work task in a work area map and determines a work end position of the work task, wherein the work area map is marked with the position, shape and size of fixed obstacles;

[0073] S202: performing path planning processing based on the current position of the mobile robot and the operation end position to generate a planned operation path of the mobile robot within the operation area map;

[0074] S203: When the mobile robot performs the operation task according to the planned operation path, a laser sensor and a visual sensor provided on the mobile robot are started to collect and process data of the surrounding environment of the mobile robot, and obtain laser point cloud data and visual image data;

[0075] S204: performing filtering and noise reduction processing on the laser point cloud data using a preset filter, and performing smoothing and hole filling processing in sequence after filtering and noise reduction to form initially processed laser point cloud data;

[0076] S205: clustering the initially processed laser point cloud data, and performing three-dimensional reconstruction of obstacles in the laser point cloud data using a three-dimensional reconstruction algorithm according to the clustering result to obtain a plurality of three-dimensionally reconstructed obstacles;

[0077] S206: performing edge detection processing on the visual image data, and performing extraction processing on a plurality of obstacles based on the edge detection result to obtain a plurality of extracted obstacles;

[0078] S207: performing obstacle fusion processing on the plurality of reconstructed obstacles and the plurality of extracted obstacles at the same angle to determine a plurality of obstacles around the mobile robot;

[0079] S208: positioning and tracking a plurality of obstacles around the mobile robot based on the laser sensor to obtain positions and movement trends corresponding to the plurality of moving obstacles around the mobile robot;

[0080] S209: inputting the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining predicted trajectory data corresponding to the multiple obstacles at the next moment;

[0081] S210: Based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when the mobile robot moves on the planned working path.

[0082] The specific implementation of the second embodiment can be found in the above embodiment and will not be described in detail here.

[0083] For example 3, please refer to Figure 3 , Figure 3 It is a schematic diagram of the structural composition of a robot autonomous obstacle avoidance device in a dynamic environment in an embodiment of the present invention.

[0084] like Figure 3 As shown, a robot autonomous obstacle avoidance device in a dynamic environment, the device comprising:

[0085] Determination module 301: used for the mobile robot to receive a work task in a work area map and determine the work end position of the work task, wherein the work area map is marked with the position, shape and size of fixed obstacles;

[0086] In the specific implementation process of the present invention, the mobile robot receives the work tasks in the work area map and determines the work end position of the work tasks, including: the mobile robot uses the robot ID to poll the work task list on the work task server at a preset interval, and when the work task matching the robot ID is polled, the work task matching the robot ID is loaded into the local storage of the mobile robot, and the work task list is formed by the work task server according to preset rules after the control user uploads the work tasks to the work task server; the mobile robot reads the locally stored work tasks, and parses the read work tasks to determine the work end position of the work tasks.

[0087] Specifically, first, the mobile robot in the working area will poll the corresponding working task on the working task server according to the preset interval, that is, poll the working task list on the working task server through the robot ID corresponding to the mobile robot, and after polling the working task matching the robot ID, through the corresponding instructions, load the working task matching the robot ID to the local storage of the mobile robot, and then the mobile robot will read the corresponding working task in the local storage, and then perform corresponding analysis on the working task to obtain the working end position of the working task.

[0088] After receiving the job task uploaded by the control user, the job task server will assign a corresponding mobile robot to the job task to execute the job task; when allocating, it will first be based on the idle time and the distance to the location of the job task, that is, the mobile robot closest to the location of the job task is selected from the currently idle mobile robots as the mobile robot executing the job task, and the robot ID of the mobile robot is bound to the job task, and updated to the job task list in the job task server; the location, shape and size of fixed obstacles are marked in the work area map, so that it is easier to identify those mobile obstacles that do not move for a short time during tracking later.

[0089] Planning module 302: used to perform path planning processing based on the current position of the mobile robot and the operation end position, and generate a planned operation path of the mobile robot in the operation area map;

[0090] In the specific implementation process of the present invention, the path planning processing is performed based on the current position of the mobile robot and the operation end position to generate the planned operation path of the mobile robot in the operation area map, including: the mobile robot starts the positioning device to perform positioning processing on the current position to obtain the current position of the mobile robot; based on the current position and the operation end position, the improved A-Star algorithm is used to perform path planning processing according to the shortest operation path to generate the planned operation path of the mobile robot in the operation area map.

[0091] Specifically, in the planning of the working path, an improved A-Star algorithm is adopted, through which the shortest working path (or optimal working path) can be planned quickly and accurately; that is, it is first necessary to start the positioning device carried by the mobile robot itself to position the current position, so as to obtain the current position of the mobile robot; then, according to the current position and the end position of the work, the improved A-Star algorithm is used to perform path planning processing according to the shortest working path, and finally a planned working path of the mobile robot in the map of the working area can be generated; by performing path planning with the improved A-Star algorithm, paths that do not meet the requirements can be accurately and quickly removed, thereby achieving rapid path planning convergence and accurately planning the required working path.

[0092] An acquisition module 303 is used for obtaining positions and movement trends corresponding to a plurality of moving obstacles based on a sensor device provided on the mobile robot when the mobile robot performs the operation task according to the planned operation path, wherein the movement trend includes a movement speed and a movement direction;

[0093] In the specific implementation process of the present invention, the positions and movement trends corresponding to multiple moving obstacles are obtained based on the sensor equipment set on the mobile robot, including: starting the laser sensor and the visual sensor set on the mobile robot to collect and process the data of the surrounding environment of the mobile robot, and obtain laser point cloud data and visual image data; based on the multi-sensor fusion algorithm, the laser point cloud data and the visual image data are processed to obtain the positions and movement trends corresponding to multiple moving obstacles.

[0094] Furthermore, the laser point cloud data and the visual image data are processed based on a multi-sensor fusion algorithm to obtain the positions and movement trends corresponding to multiple moving obstacles, including: filtering and denoising the laser point cloud data using a preset filter, and after filtering and denoising, smoothing and hole filling processing are performed in sequence to form initially processed laser point cloud data; clustering the initially processed laser point cloud data, and performing three-dimensional reconstruction of obstacles in the laser point cloud data using a three-dimensional reconstruction algorithm according to the clustering results to obtain a number of three-dimensional reconstructed obstacles; edge detection processing is performed on the visual image data, and a number of obstacles are extracted based on the edge detection results to obtain a number of extracted obstacles; obstacle fusion processing is performed on the several reconstructed obstacles and the several extracted obstacles at the same angle to determine a number of obstacles around the mobile robot; positioning and tracking the several obstacles around the mobile robot based on the laser sensor to obtain the positions and movement trends corresponding to the multiple moving obstacles around the mobile robot.

[0095] Specifically, in this solution, obstacles are identified and tracked through the sensor equipment carried by the mobile robot, so as to obtain the positions and movement trends corresponding to multiple moving obstacles; and the sensor equipment is mainly laser sensors and visual sensors, that is, the surrounding environment of the mobile robot is collected and processed by laser sensors and visual sensors, so as to obtain laser point cloud data and visual image data; then the laser point cloud data and visual image data are processed by a multi-sensor fusion algorithm, so as to extract the positions and movement trends corresponding to multiple moving obstacles, wherein the movement trend includes movement speed and movement direction.

[0096] First, the laser point cloud data is filtered and denoised using a preset filter, and after filtering and denoising, it is smoothed and hole-filled in turn to form the initially processed laser point cloud data; then the initially processed laser point cloud data is clustered, and according to the clustering results, the 3D reconstruction algorithm is used to perform 3D reconstruction of obstacles in the laser point cloud data to obtain several 3D reconstructed obstacles; the several 3D reconstructed obstacles here include fixed obstacles marked in the work area map; then the color space of the collected visual image data is converted to realize the conversion of the visual image data into an HSV image, and The HSV image is processed for extracting the information of the H channel, so as to obtain the H channel image corresponding to the HSV image; the H channel image is processed for edge detection by using an edge detection algorithm, so as to extract the edge detection result; and several obstacles are extracted from the edge detection result; several obstacles around the mobile robot are determined by performing obstacle fusion processing on several reconstructed obstacles and several extracted obstacles at the same angle; then these several obstacles are used to match the positions, shapes and sizes of the fixed obstacles marked in the work area map one by one, and the suspected fixed obstacles among the several obstacles are marked by matching.

[0097] Then, the laser sensor is controlled to locate and track several obstacles around the mobile robot, and the suspected fixed obstacles among the obstacles are marked according to the positioning and tracking results, so as to determine the corresponding positions and movement trends of multiple moving obstacles around the mobile robot.

[0098] Prediction module 304: used to input the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtain the predicted trajectory data corresponding to the multiple obstacles at the next moment;

[0099] In a specific implementation of the present invention, the positions and movement trends corresponding to the multiple moving obstacles are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtain predicted trajectory data corresponding to the multiple obstacles at the next moment, including: indexing the corresponding obstacle category in a database according to the contour information corresponding to the multiple moving obstacles, and marking the obstacle category on the corresponding multiple moving obstacles to form a plurality of marked mobile obstacles; inputting the multiple marked mobile obstacles and the positions and movement trends corresponding to the multiple marked mobile obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtain predicted trajectory data corresponding to the multiple obstacles at the next moment; wherein the obstacle trajectory prediction model is a model established and trained and converged by a radial basis function neural network.

[0100] Specifically, firstly, the corresponding obstacle categories are indexed in the database according to the contour information corresponding to the multiple moving obstacles, and the obstacle categories are marked on the corresponding multiple moving obstacles to form multiple marked moving obstacles; then, the multiple marked moving obstacles, the positions corresponding to the multiple marked moving obstacles, and the motion trends are input into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and the predicted trajectory data corresponding to the multiple obstacles at the next moment are obtained; wherein, the obstacle trajectory prediction model is a model established and trained and converged by a radial basis function neural network.

[0101] The structure of the radial basis function neural network is similar to that of a multi-layer forward network, which is a three-layer forward network; the input layer is composed of signal source nodes; the second layer is the hidden layer, and the number of hidden units depends on the needs of the problem being described. The transformation function of the hidden unit is the RBF radial basis function, which is a non-negative nonlinear function that is radially symmetric and decays about the center point; the third layer is the output layer, which responds to the action of the input pattern; the transformation from the input space to the hidden layer space is nonlinear, while the transformation from the hidden layer space to the output layer space is linear; and it has high accuracy when applied to trajectory prediction of moving objects.

[0102] The adjustment control module 305 is used to perform autonomous obstacle avoidance adjustment control on the mobile robot at the next moment when the mobile robot moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment.

[0103] In the specific implementation process of the present invention, the autonomous obstacle avoidance adjustment control of the mobile robot at the next moment when it moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment includes: inputting the predicted trajectory data corresponding to the multiple obstacles at the next moment into the built-in behavior decision module of the mobile robot for decision processing, and generating a decision instruction corresponding to the next moment when moving on the planned working path according to the decision result; and autonomous obstacle avoidance adjustment control of the mobile robot at the next moment when it moves on the planned working path based on the decision instruction.

[0104] In the specific implementation process of the present invention, the predicted trajectory data corresponding to multiple obstacles at the next moment are input into the built-in behavior decision module of the mobile robot for decision processing, and then the corresponding decision instructions for the next moment when moving on the planned working path are generated according to the decision results; in the present technical solution, the behavior decision model is a decision model formed by adjusting the parameters of the PID controller according to the actual situation; it can give corresponding feedback decisions according to the specific situation.

[0105] When receiving the decision instruction, the mobile robot will realize the autonomous obstacle avoidance adjustment control at the next moment when it moves on the planned working path according to the decision instruction.

[0106] In an embodiment of the present invention, a work task in a work area map is received by a mobile robot, and the work end position of the work task is determined; path planning processing is performed based on the current position and the work end position to generate a planned work path; when the mobile robot executes the work task, the positions and motion trends corresponding to multiple moving obstacles are obtained; the positions and motion trends corresponding to the multiple moving obstacles are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and the predicted trajectory data corresponding to the multiple obstacles at the next moment are obtained; based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned work path; autonomous and precise obstacle avoidance control is achieved in a complex working environment, and the work task is successfully completed to ensure work safety.

[0107] A computer-readable storage medium provided in an embodiment of the present invention stores a computer program on the computer-readable storage medium, and when the program is executed by a processor, the robot autonomous obstacle avoidance method of any one of the above embodiments is implemented. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (EraSable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or optical card. That is, the storage device includes any medium that can store or transmit information in a readable form by a device (for example, a computer, a mobile phone), which can be a read-only memory, a disk or an optical disk, etc.

[0108] An embodiment of the present invention further provides a computer application program, which runs on a computer and is used to execute the robot autonomous obstacle avoidance method of any one of the above embodiments.

[0109] also, Figure 4 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention.

[0110] The embodiment of the present invention further provides an electronic device, such as Figure 4 The electronic device includes a processor 402, a memory 403, an input unit 404, a display unit 405 and other devices. Those skilled in the art will understand that Figure 4The structural components of the electronic device shown do not constitute a limitation on all devices, and may include more or fewer components than shown, or combine certain components. The memory 403 can be used to store the application 401 and various functional modules, and the processor 402 runs the application 401 stored in the memory 403, thereby executing various functional applications and data processing of the device. The memory can be an internal memory or an external memory, or include both internal and external memories. The internal memory may include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, or a random access memory. The external memory may include a hard disk, a floppy disk, a ZIP disk, a U disk, a magnetic tape, etc. The memory disclosed in the present invention includes but is not limited to these types of memories. The memory disclosed in the present invention is only used as an example and not as a limitation.

[0111] The input unit 404 is used to receive the input of the signal and the keyword input by the user. The input unit 404 may include a touch panel and other input devices. The touch panel can collect the user's touch operation on or near it (such as the user's operation on or near the touch panel using any suitable object or accessory such as a finger, stylus, etc.), and drive the corresponding connection device according to a pre-set program; other input devices may include but are not limited to one or more of a physical keyboard, a function key (such as a playback control key, a switch key, etc.), a trackball, a mouse, a joystick, etc. The display unit 405 can be used to display the information input by the user or the information provided to the user and various menus of the terminal device. The display unit 405 can be in the form of a liquid crystal display, an organic light emitting diode, etc. The processor 402 is the control center of the terminal device, which uses various interfaces and lines to connect the various parts of the entire device, and executes various functions and processes data by running or executing software programs and / or modules stored in the memory 403, and calling the data stored in the memory.

[0112] As an embodiment, the electronic device includes: one or more processors 402, a memory 403, and one or more applications 401, wherein the one or more applications 401 are stored in the memory 403 and are configured to be executed by the one or more processors 402, and the one or more applications 401 are configured to execute the corresponding robot autonomous obstacle avoidance method in any of the above-mentioned embodiments.

[0113] In an embodiment of the present invention, a work task in a work area map is received by a mobile robot, and the work end position of the work task is determined; path planning processing is performed based on the current position and the work end position to generate a planned work path; when the mobile robot executes the work task, the positions and motion trends corresponding to multiple moving obstacles are obtained; the positions and motion trends corresponding to the multiple moving obstacles are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and the predicted trajectory data corresponding to the multiple obstacles at the next moment are obtained; based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned work path; autonomous and precise obstacle avoidance control is achieved in a complex working environment, and the work task is successfully completed to ensure work safety.

[0114] In addition, the above is a detailed introduction to a robot autonomous obstacle avoidance method and related devices based on a dynamic environment provided by an embodiment of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A robot autonomous obstacle avoidance method in a dynamic environment, characterized in that: The method comprises: The mobile robot receives a work task in a work area map and determines a work end position of the work task, wherein the work area map is marked with the position, shape and size of fixed obstacles; Performing path planning based on the current position of the mobile robot and the operation end position to generate a planned operation path for the mobile robot within the operation area map; When the mobile robot performs the operation task according to the planned operation path, positions and movement trends corresponding to a plurality of moving obstacles are obtained based on a sensor device provided on the mobile robot, wherein the movement trend includes a movement speed and a movement direction; Inputting the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining the predicted trajectory data corresponding to the multiple obstacles at the next moment; Based on the predicted trajectory data corresponding to the multiple obstacles at the next moment, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned working path.

2. The robot autonomous obstacle avoidance method according to claim 1, characterized in that: The mobile robot receives a work task in a work area map and determines a work end position of the work task, including: The mobile robot uses the robot ID to poll the job task list on the job task server at a preset interval. When a job task matching the robot ID is polled, the job task matching the robot ID is loaded into the local storage of the mobile robot. The job task list is formed by the job task server assigning and updating the job tasks according to preset rules after the control user uploads the job tasks to the job task server. The mobile robot reads the locally stored operation task, analyzes the read operation task, and determines the operation end position of the operation task.

3. The robot autonomous obstacle avoidance method according to claim 1, characterized in that: The path planning process is performed based on the current position of the mobile robot and the operation end position to generate a planned operation path of the mobile robot within the operation area map, including: The mobile robot starts the positioning device to perform positioning processing on the current position to obtain the current position of the mobile robot; Based on the current position and the operation end position, the improved A-Star algorithm is used to perform path planning processing according to the shortest operation path to generate a planned operation path for the mobile robot within the operation area map.

4. The robot autonomous obstacle avoidance method according to claim 1, characterized in that: The obtaining of positions and movement trends corresponding to a plurality of moving obstacles based on a sensor device arranged on the mobile robot includes: Starting the laser sensor and the visual sensor provided on the mobile robot to collect and process data of the surrounding environment of the mobile robot, and obtaining laser point cloud data and visual image data; The laser point cloud data and the visual image data are processed based on a multi-sensor fusion algorithm to obtain positions and movement trends corresponding to multiple moving obstacles.

5. The robot autonomous obstacle avoidance method according to claim 4, characterized in that: The laser point cloud data and the visual image data are processed based on a multi-sensor fusion algorithm to obtain positions and movement trends corresponding to multiple moving obstacles, including: The laser point cloud data is subjected to filtering and noise reduction processing by using a preset filter, and after filtering and noise reduction, smoothing and hole filling processing are performed in sequence to form initially processed laser point cloud data; Clustering is performed on the laser point cloud data after the initial processing, and a three-dimensional reconstruction algorithm is used to perform three-dimensional reconstruction of obstacles in the laser point cloud data according to the clustering results to obtain a number of three-dimensional reconstructed obstacles; Performing edge detection processing on the visual image data, and performing extraction processing on a plurality of obstacles based on the edge detection result to obtain a plurality of extracted obstacles; Performing obstacle fusion processing on the plurality of reconstructed obstacles and the plurality of extracted obstacles at the same angle to determine a plurality of obstacles around the mobile robot; Based on the laser sensor, several obstacles around the mobile robot are positioned and tracked to obtain the positions and movement trends of the multiple moving obstacles around the mobile robot.

6. The robot autonomous obstacle avoidance method according to claim 1, characterized in that: The step of inputting the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining the predicted trajectory data corresponding to the multiple obstacles at the next moment, includes: Indexing corresponding obstacle categories in a database according to the contour information corresponding to the plurality of moving obstacles, and marking the obstacle categories on the corresponding plurality of moving obstacles to form a plurality of marked moving obstacles; Multiple marked moving obstacles, their corresponding positions and movement trends are input into an obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, so as to obtain the predicted trajectory data corresponding to the multiple obstacles at the next moment; wherein the obstacle trajectory prediction model is a model established and trained and converged by a radial basis function neural network.

7. The robot autonomous obstacle avoidance method according to claim 1, characterized in that: The autonomous obstacle avoidance adjustment control of the mobile robot at the next moment when it moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment includes: Inputting the predicted trajectory data corresponding to the multiple obstacles at the next moment into the built-in behavior decision module of the mobile robot for decision processing, and generating the corresponding decision instructions at the next moment when moving on the planned working path according to the decision result; Based on the decision instruction, autonomous obstacle avoidance adjustment control is performed on the mobile robot at the next moment when it moves on the planned working path.

8. A robot autonomous obstacle avoidance device in a dynamic environment, characterized in that: The device comprises: Determination module: used for the mobile robot to receive the operation task in the operation area map and determine the operation end position of the operation task, and the operation area map is marked with the position, shape and size of the fixed obstacles; Planning module: used for performing path planning processing based on the current position of the mobile robot and the operation end position, and generating a planned operation path of the mobile robot within the operation area map; An acquisition module: used for acquiring positions and movement trends corresponding to a plurality of moving obstacles based on a sensor device arranged on the mobile robot when the mobile robot performs the operation task according to the planned operation path, wherein the movement trend includes a movement speed and a movement direction; Prediction module: used for inputting the positions and movement trends corresponding to the multiple moving obstacles into the obstacle trajectory prediction model to perform trajectory prediction processing at the next moment, and obtaining the predicted trajectory data corresponding to the multiple obstacles at the next moment; Adjustment control module: used for performing autonomous obstacle avoidance adjustment control on the mobile robot at the next moment when it moves on the planned working path based on the predicted trajectory data corresponding to the multiple obstacles at the next moment.

9. An electronic device comprising a processor and a memory, characterized in that: The processor runs the computer program or code stored in the memory to implement the robot autonomous obstacle avoidance method according to any one of claims 1 to 7.

10. A computer-readable storage medium for storing a computer program or code, characterized in that: When the computer program or code is executed by a processor, the robot autonomous obstacle avoidance method according to any one of claims 1 to 7 is implemented.

Images