Robot path planning method and system

By collecting and analyzing image and perceived data in the hospital environment, generating optimized paths and making real-time adjustments, the real-time and accuracy problems in the path planning of the distribution robot are solved to ensure the timely delivery of medical supplies.

CN118963345APending Publication Date: 2024-11-15JIANGXI SAITE INTELLIGENT TECH CO LTD
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202411023375.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing distribution robot path planning methods have poor real-time and low accuracy in hospital environments, resulting in delays in the delivery of medical supplies and affecting the normal development of medical work.

Method used

By collecting image data and perceived data, transmitting it to the cloud platform, analyzing and integrating the generation of target obstacle avoidance data, building a priority system and path impact factor, generating target paths, and continuously optimizing and adjusting the paths during the robot movement process.

Benefits of technology

It improves the real-time and accuracy of the delivery robot in the face of complex road conditions, avoids delivery delays, and ensures the stable development of medical work.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963345A_ABST
    Figure CN118963345A_ABST
Patent Text Reader

Abstract

The invention provides a robot path planning method and system. The method comprises the steps that image data and sensing data are collected and transmitted to a cloud platform; analyzing the image data and the sensing data, and fusing analysis results to generate target obstacle avoidance data; constructing a priority system and a path influence factor of the target delivery task, and generating a target path based on the priority system, the path influence factor and the target obstacle avoidance data; and controlling the distribution robot to move according to the target path, and continuously optimizing and adjusting the target path. According to the invention, the problem that the route planning of the robot in a hospital scene cannot meet the real-time performance and accuracy of distribution can be solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a robot path planning method and system. Background Art

[0002] Delivery robots are deployed in hospital environments to deliver medical supplies. However, the public passages of hospitals are often congested. There are dynamically changing obstacles such as moving patients, medical staff, medical equipment, and beds in the public passages. Delivery robots need to cope with complex road conditions with multiple obstacles. In addition, delivery tasks are time-sensitive, and delivery robots will also face the challenge of completing delivery tasks in a timely manner. Therefore, the efficient operation of delivery robots is crucial to the timeliness and quality of medical services.

[0003] The existing delivery robot path planning methods have problems such as poor real-time performance and low accuracy when dealing with traffic congestion in public passages in hospitals, which can easily lead to delays in the delivery of medical supplies and affect the normal development of medical work. Therefore, a robot path planning method is urgently needed to solve the robot path planning problem in hospital scenarios. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention proposes a robot path planning method and system.

[0005] The first aspect of the present invention discloses a robot path planning method, comprising:

[0006] Collect image data and perception data and transmit them to the cloud platform;

[0007] The cloud platform analyzes the image data and the perception data, and generates target obstacle avoidance data by fusing the analysis results;

[0008] Constructing a priority system and a path influencing factor, and generating a target path based on the priority system, the path influencing factor and the target obstacle avoidance data;

[0009] The delivery robot is controlled to move according to the target path, and the target path is continuously optimized and adjusted.

[0010] As an optional implementation, in the first aspect of the present invention, the collecting image data and the sensing data are transmitted to the cloud platform, including:

[0011] The target area is determined based on the historical traffic data of the hospital's public area, and visual sensors are deployed in the target area to collect image data in real time. The network camera used as a visual sensor collects high-frame-rate image data of the environment in the target area.

[0012] A radar sensor is deployed on the delivery robot to collect perception data in real time. The radar sensor collects the real-time speed and real-time angular velocity of surrounding obstacles and the distance between the obstacles and the delivery robot as the perception data;

[0013] The image data and perception data are transmitted to the cloud platform via wireless network real-time communication.

[0014] As an optional implementation, in the first aspect of the present invention, the cloud platform analyzes the image data and the perception data, including:

[0015] The image recognition model was trained using historical traffic data from the hospital’s public areas;

[0016] Preprocessing and feature extraction are performed on the image data to generate a feature map, wherein the preprocessing includes noise elimination, outlier removal, and normalization processing;

[0017] The feature map is used as an input of the model, and a recognition result of the target obstacle in the target area is output as the first environmental data, wherein the first environmental data includes the first type, the first relative coordinate, the first motion state and the first environmental information of the target obstacle in the target area environment during the real-time acquisition time of the visual sensor;

[0018] Signal processing technology is used to perform preprocessing operations such as signal filtering, signal amplification, signal compression and signal enhancement on the perception data, and feature extraction and feature classification are performed on the preprocessed perception data through pattern recognition. The recognition result of the surrounding environment of the delivery robot is output as the second environmental data. The second environmental data includes the second type, second relative coordinates, second motion state and obstacle distribution map of the target obstacle within the real-time time collected by the radar sensor.

[0019] As an optional implementation, in the first aspect of the present invention, the fusion analysis result generates target obstacle avoidance data, including:

[0020] Performing a time synchronization operation and a space registration operation on the first environment data and the second environment data, wherein the time synchronization operation includes interpolation processing, resampling processing, and timestamp alignment, and the space registration operation includes performing coordinate transformation on the first relative coordinate and the second relative coordinate through the same coordinate system to obtain the absolute coordinates of the target obstacle in the target obstacle avoidance data;

[0021] The target obstacle type is determined by overlapping the first type and the second type, the position of the obstacle is confirmed based on the absolute coordinates, the movement trend of the target obstacle is predicted according to the first motion state and the second motion state, and the environmental information and the obstacle distribution map are integrated to generate an environmental perception map around the delivery robot in a mobile state as target obstacle avoidance data, wherein the movement trend includes the moving speed, moving angle, angular velocity and motion posture of the target obstacle.

[0022] As an optional implementation, in the first aspect of the present invention, constructing a priority system and a path influence factor, and generating a target path based on the priority system, the path influence factor and the target obstacle avoidance data, includes:

[0023] Determine the starting point and end point of the initial path according to the target delivery task;

[0024] The priority label of the target delivery task is constructed by the first urgency of the material priority classification, the second urgency of the patient-related materials and medical care-related materials, and the third urgency of the delivery time. The priority system is established through the priority label to mark the priority of the target delivery task;

[0025] Use the delivery robot’s historical movement data to train the path planning model;

[0026] The path planning model prioritizes the target paths generated by the delivery robot through the priority marking, and the target paths generated first will serve as obstacles for the target paths generated later;

[0027] The obstacle map is generated by environmental modeling based on the target obstacle avoidance data, and a path search algorithm is used to generate several path points according to the starting point and end point of the initial path. Several optional paths are generated by sequentially connecting the starting point, path points and end point.

[0028] As an optional implementation, in the first aspect of the present invention, generating a target path based on the priority system, the path impact factor and the target obstacle avoidance data includes:

[0029] The path length, path smoothness, safe distance between obstacles, number of turns and congestion level of the path area are configured as path influencing factors in each optional path, and a corresponding weight is assigned to each of the path influencing factors. The evaluation value of each path influencing factor is calculated by weighted calculation, and the evaluation score of the corresponding optional path is output. The optional path with the highest evaluation score is determined as the target path for the delivery robot to perform the target delivery task.

[0030] As an optional implementation, in the first aspect of the present invention, the continuously optimizing and adjusting the target path includes:

[0031] The motion state data of the delivery robot moving in the target path is collected in real time by radar sensors, and the environmental data of the delivery robot moving in the target path is collected in real time by visual sensors. The offset between the actual path of the delivery robot and the target path is calculated based on the environmental data and the motion state data, and the motion state of the delivery robot is continuously optimized based on the offset until the correction of the target path is completed.

[0032] A second aspect of the present invention discloses a robot path planning system, the system comprising:

[0033] Communication module, sensor module, cloud platform module, path planning module and central control module. The path planning module is deployed on the cloud platform module, and each module performs real-time data transmission and data interaction through the communication module.

[0034] Communication module, sensor module, cloud platform module, path planning module and central control module. The path planning module is deployed on the cloud platform module, and each module performs real-time data transmission and data interaction through the communication module.

[0035] The sensor module is used to collect image data and transmit perception data to the cloud platform;

[0036] The cloud platform module is used to analyze the image data and perception data, and fuse the analysis results to generate target obstacle avoidance data;

[0037] The path planning module is used to construct a priority system and path influencing factors, and generate a target path based on the priority system, path influencing factors and target obstacle avoidance data;

[0038] The central control module is used to control the movement of the delivery robot according to the target path and continuously optimize and adjust the target path.

[0039] As an optional implementation, in the second aspect of the present invention, the sensor module collects image data and perception data and transmits them to the cloud platform, including:

[0040] The target area is determined based on the historical traffic data of the hospital's public area, and visual sensors are deployed in the target area to collect image data in real time. The network camera used as a visual sensor collects high-frame-rate image data of the environment in the target area.

[0041] A radar sensor is deployed on the delivery robot to collect perception data in real time. The radar sensor collects the real-time speed and real-time angular velocity of surrounding obstacles and the distance between the obstacles and the delivery robot as the perception data;

[0042] The image data and perception data are transmitted to the cloud platform via wireless network real-time communication.

[0043] As an optional implementation, in the second aspect of the present invention, the analyzing the image data and the perception data includes:

[0044] The image recognition model was trained using historical traffic data from the hospital’s public areas;

[0045] Preprocessing and feature extraction are performed on the image data to generate a feature map, wherein the preprocessing includes noise elimination, outlier removal, and normalization processing;

[0046] The feature map is used as an input of the model, and a recognition result of the target obstacle in the target area is output as the first environmental data, wherein the first environmental data includes the first type, the first relative coordinate, the first motion state and the first environmental information of the target obstacle in the target area environment during the real-time acquisition time of the visual sensor;

[0047] Signal processing technology is used to perform preprocessing operations such as signal filtering, signal amplification, signal compression and signal enhancement on the perception data, and feature extraction and feature classification are performed on the preprocessed perception data through pattern recognition. The recognition result of the surrounding environment of the delivery robot is output as the second environmental data. The second environmental data includes the second type, second relative coordinates, second motion state and obstacle distribution map of the target obstacle within the real-time time collected by the radar sensor.

[0048] As an optional implementation, in the second aspect of the present invention, the cloud platform module generates target obstacle avoidance data by integrating the analysis results, including:

[0049] Performing a time synchronization operation and a space registration operation on the first environment data and the second environment data, wherein the time synchronization operation includes interpolation processing, resampling processing, and timestamp alignment, and the space registration operation includes performing coordinate transformation on the first relative coordinate and the second relative coordinate through the same coordinate system to obtain the absolute coordinates of the target obstacle in the target obstacle avoidance data;

[0050] The target obstacle type is determined by overlapping the first type and the second type, the position of the obstacle is confirmed based on the absolute coordinates, the movement trend of the target obstacle is predicted according to the first motion state and the second motion state, and the environmental information and the obstacle distribution map are integrated to generate an environmental perception map around the delivery robot in a mobile state as target obstacle avoidance data, wherein the movement trend includes the moving speed, moving angle, angular velocity and motion posture of the target obstacle.

[0051] As an optional implementation, in the second aspect of the present invention, the path planning module constructs a priority system and a path influence factor, and generates a target path based on the priority system, the path influence factor and the target obstacle avoidance data, including:

[0052] Determine the starting point and end point of the initial path according to the target delivery task;

[0053] The priority label of the target delivery task is constructed by the first urgency of the material priority classification, the second urgency of the patient-related materials and medical care-related materials, and the third urgency of the delivery time. The priority system is established through the priority label to mark the priority of the target delivery task;

[0054] Use the delivery robot’s historical movement data to train the path planning model;

[0055] The path planning model prioritizes the target paths generated by the delivery robot through the priority marking, and the target paths generated first will serve as obstacles for the target paths generated later;

[0056] The obstacle map is generated by environmental modeling based on the target obstacle avoidance data, and a path search algorithm is used to generate several path points according to the starting point and end point of the initial path. Several optional paths are generated by sequentially connecting the starting point, path points and end point.

[0057] As an optional implementation, in the second aspect of the present invention, the path planning module generates a target path based on the priority system, the path influencing factor and the target obstacle avoidance data, including:

[0058] The path length, path smoothness, safe distance between obstacles, number of turns and congestion level of the path area are configured as path influencing factors in each optional path, and a corresponding weight is assigned to each of the path influencing factors. The evaluation value of each path influencing factor is calculated by weighted calculation, and the evaluation score of the corresponding optional path is output. The optional path with the highest evaluation score is determined as the target path for the delivery robot to perform the target delivery task.

[0059] As an optional implementation, in the second aspect of the present invention, the central control module continuously optimizes and adjusts the target path, including:

[0060] The motion state data of the delivery robot moving in the target path is collected in real time by radar sensors, and the environmental data of the delivery robot moving in the target path is collected in real time by visual sensors. The offset between the actual path of the delivery robot and the target path is calculated based on the environmental data and the motion state data, and the motion state of the delivery robot is continuously optimized based on the offset until the correction of the target path is completed.

[0061] A third aspect of the present invention discloses a robot path planning device, comprising:

[0062] at least one processor, and

[0063] a memory communicatively connected to the at least one processor; wherein,

[0064] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the robot path planning method as described in any one of the first aspects of the present invention.

[0065] The fourth aspect of the present invention discloses a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the robot path planning method as described in any one of the first aspect of the present invention.

[0066] Compared with the prior art, the present invention has the following advantages:

[0067] The present invention uses sensors to collect image data and perception data in real time and transmits them to a cloud platform for analysis. It relies on cloud computing and big data technology to quickly integrate and process environmental data fed back in real time by visual sensors and radar sensors, and identifies the traffic congestion status in the hospital in real time as an obstacle avoidance basis for generating a target path. It constructs a priority system and path influencing factors as screening conditions for the target path, outputs the optimal target path through a path planning model to control the robot to perform the delivery task, and continuously optimizes the path parameters during the movement of the robot to ensure that the delivery robot completes the delivery task efficiently and accurately, improves the real-time and accuracy of the delivery robot in performing delivery tasks when facing complex road conditions, avoids delivery delays to ensure the stable development of medical work, and solves the path planning problem of the delivery robot in the hospital scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] 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 labor.

[0069] Figure 1 is a flow chart of the robot path planning method of the present invention;

[0070] Figure 2 Schematic diagram of the robot path planning system of the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0072] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0073] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0074] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0075] Example 1

[0076] See also Figure 1 , an embodiment of the present invention discloses a robot path planning method, comprising:

[0077] 101. Collect image data and perception data and transmit them to the cloud platform.

[0078] Specifically, the target areas are areas where traffic congestion often occurs in the hospital environment, such as aisles, corridors, elevator entrances, stairwells, waiting rooms, cashier areas, and medicine collection areas. The demarcation of the target areas can be determined by evaluating the degree of congestion through historical image data and comprehensive confirmation through manual experience.

[0079] As an optional implementation, in the first aspect of the present invention, the collecting image data and the sensing data are transmitted to the cloud platform, including:

[0080] The target area is determined based on the historical traffic data of the hospital's public area, and visual sensors are deployed in the target area to collect image data in real time. The network camera used as a visual sensor collects high-frame-rate image data of the environment in the target area.

[0081] A radar sensor is deployed on the delivery robot to collect perception data in real time. The radar sensor collects the real-time speed and real-time angular velocity of surrounding obstacles and the distance between the obstacles and the delivery robot as the perception data;

[0082] The image data and perception data are transmitted to the cloud platform via wireless network real-time communication.

[0083] Furthermore, the visual sensor may be a network camera, wherein the clarity of the camera is adjusted according to the transmission speed required for acquiring the image. Since the size of images acquired with different clarity is related to the speed of data transmission, the camera can meet the clarity required by the cloud platform to extract features of obstacles in the environment while ensuring real-time transmission. For example, the camera can clearly capture dynamic obstacles such as moving beds, walking patients, and medical staff in the target area, as well as static obstacles such as displayed beds and medical equipment. The image data is an image containing environmental information in the target area acquired within a set unit time, and the image data will record the sampling rate and timestamp during the acquisition process for time synchronization of subsequent data fusion. The acquired image data can be transmitted to the cloud platform for processing via the 5G network connection communication established between the camera and the cloud platform. The 5G communication method in this application is used as a preferred implementation method, and other methods that can transmit data in real time may also be used.

[0084] Furthermore, the radar sensor is preferably connected to the cloud platform through a 5G network to establish a communication connection. The radar sensor collects the distance between the obstacles and the robot in the environment where the delivery robot is located, the speed of the obstacles, the movement trajectory of the obstacles, and the shape and size of the obstacles. The above collected data constitutes perception data, which is then transmitted to the cloud platform for processing through the 5G network. The perception data can be used in conjunction with image data to analyze the congestion status of the robot's current environment. For example, image data recognition can first detect that a group of people are gathered in the corridor, but the image data cannot be used to determine the precise distance between the crowd and the robot. At this time, the precise distance between the crowd and the robot can be obtained by analyzing and processing the perception data transmitted by the radar sensor, thereby controlling the robot's movement path and achieving the purpose of adjusting the robot's motion state in real time to adapt to different congestion scenarios.

[0085] Furthermore, in order to continuously optimize the robot walking on the planned route and monitor the robot's motion status on a daily basis, the delivery robot can deploy various sensors. For example, in addition to radar sensors, the delivery robot can also be equipped with positioning sensors (such as GPS, Beidou and other satellite positioning sensors or other positioning sensors) and motion sensors (such as gyroscopes, odometers, speedometers, accelerometers, etc.). The sensors can obtain the robot's position and motion status in real time. The motion status can be measured by measuring the delivery robot's moving speed, moving direction, moving acceleration, moving angular velocity and posture data. By monitoring the robot's motion status, the robot's path planning can be better and continuously optimized and controlled.

[0086] 102. The cloud platform analyzes the image data and the perception data, and fuses the analysis results to generate target obstacle avoidance data.

[0087] As an optional implementation, in the first aspect of the present invention, the cloud platform analyzes the image data and the perception data, including:

[0088] The image recognition model was trained using historical traffic data from the hospital’s public areas;

[0089] Preprocessing and feature extraction are performed on the image data to generate a feature map, wherein the preprocessing includes noise elimination, outlier removal, and normalization processing;

[0090] The feature map is used as an input of the model, and a recognition result of the target obstacle in the target area is output as the first environmental data, wherein the first environmental data includes the first type, the first relative coordinate, the first motion state and the first environmental information of the target obstacle in the target area environment during the real-time acquisition time of the visual sensor;

[0091] Signal processing technology is used to perform preprocessing operations such as signal filtering, signal amplification, signal compression and signal enhancement on the perception data, and feature extraction and feature classification are performed on the preprocessed perception data through pattern recognition. The recognition result of the surrounding environment of the delivery robot is output as the second environmental data. The second environmental data includes the second type, second relative coordinates, second motion state and obstacle distribution map of the target obstacle within the real-time time collected by the radar sensor.

[0092] Specifically, the image data is a high frame rate image, which is transmitted to the cloud platform and relies on the cloud platform and big data technology for rapid analysis and processing. Since the images collected by the visual sensor need to cover different scenes, lighting conditions and obstacle types, the image data needs to be preprocessed to improve the subsequent model training and prediction effects. The preprocessing steps may include image denoising, normalization, grayscale, resizing, etc. For example, grayscale can convert color images into grayscale images to reduce the amount of calculation while retaining key information, thereby converting the image data into a feature vector that can be used as a model input.

[0093] Preferably, the image recognition model is a neural network model in deep learning technology, preferably a convolutional neural network model CNN, which can automatically learn and recognize the features of the image. The historically collected image data is used as the training data of the convolutional neural network. The model automatically learns the local features of the robot and obstacles in the image, and finally obtains the global features of the robot and obstacles through forward propagation layer by layer in the network. Then, the error between the obstacle recognition result and the true result is verified through back propagation to update the model parameters for performance optimization. This can ensure that when the delivery robot actually performs the delivery task, the image data transmitted to the cloud platform can quickly respond and return the obstacle recognition result in the current environment of the delivery robot, that is, the first environment data. The recognition result of the first environment data includes the recognized robot, the type of obstacle, the shape of the obstacle, the size of the obstacle, the relative coordinates of the obstacle and the delivery robot, the movement state of the obstacle and other environmental information such as walls, stairs, corners, entrances and exits, door positions, etc. The first environment data obtained through recognition can be used as a data basis to more accurately plan the movement path of the appropriate delivery robot and control the delivery robot to complete the target task.

[0094] Furthermore, the initial data obtained by the radar sensor is reflected signal data, which needs to be preprocessed through signal processing technology to improve the quality and processability of the signal. The preprocessing methods include filtering (removing noise and interference in the radar signal, improving the signal-to-noise ratio, and helping subsequent signal processing to be more accurate and reliable), denoising (further reducing the noise component in the signal and improving the clarity of the signal) and data compression (reducing the amount of radar signal data and improving processing efficiency, which is especially important for processing large-scale radar data). Subsequently, the distance between the obstacle and the radar (i.e., the delivery robot body) is calculated by the time difference between the transmitted signal and the received reflected signal, and the speed of the obstacle is determined by calculating the change in signal frequency. Combined with the continuously collected distance and speed data, the motion trajectory of the obstacle can be constructed to facilitate subsequent tracking of moving targets and predict their future positions, thereby converting the signal data type into perception data of digital data type.

[0095] Furthermore, the perception data as radar data usually includes a large number of raw data points. The data points may represent the position, speed and direction of each obstacle target in the current environment of the delivery robot. Therefore, it is necessary to extract features and classify the raw data points to obtain accurate identification and classification of each obstacle target. In pattern recognition, feature extraction can use waveform feature extraction, Doppler spectrum feature extraction, polarization feature extraction, point cloud data processing, etc., and then combine mathematical statistical tools such as principal cost analysis PCA and linear discriminant analysis for feature selection and optimization. The classification algorithm is applied to the extracted features to identify the target obstacles. The classification algorithm can use decision trees, support vector machines and neural networks to classify and identify the feature data of the perception data, and output the obstacle sensing map of the environment around the delivery robot as the second environment data. The second environment data combined with the first environment data can further increase the accuracy of the robot's path planning, providing strong data support for path planning.

[0096] It should be noted that the feature extraction and feature classification of sensing data can also be performed using only signal processing technology, specifically, signal imaging and parameter estimation are used to obtain detailed information of the target obstacle, where signal imaging performs synthetic aperture processing on the slow time dimension of the radar signal so that the target obstacle position is focused in the direction of radar movement, and energy focusing on the distance is achieved through pulse compression. This can form an image of the scattering characteristics of the target obstacle, which is helpful for the identification and classification of the target obstacle, and parameter estimation can realize the optimal estimation theory by performing Kalman filtering on the estimated positions within multiple dwell times, and realize the estimation of target position, speed and other information to obtain the trajectory estimation information of the target obstacle movement, and can also generate the second environmental data for evaluating the current environmental congestion of the delivery robot.

[0097] As an optional implementation, in the first aspect of the present invention, the fusion analysis result generates target obstacle avoidance data, including:

[0098] Performing a time synchronization operation and a space registration operation on the first environment data and the second environment data, wherein the time synchronization operation includes interpolation processing, resampling processing, and timestamp alignment, and the space registration operation includes performing coordinate transformation on the first relative coordinate and the second relative coordinate through the same coordinate system to obtain the absolute coordinates of the target obstacle in the target obstacle avoidance data;

[0099] The target obstacle type is determined by overlapping the first type and the second type, the position of the obstacle is confirmed based on the absolute coordinates, the movement trend of the target obstacle is predicted according to the first motion state and the second motion state, and the environmental information and the obstacle distribution map are integrated to generate an environmental perception map around the delivery robot in a mobile state as target obstacle avoidance data, wherein the movement trend includes the moving speed, moving angle, angular velocity and motion posture of the target obstacle.

[0100] Specifically, since the image data and radar data come from different sensors and have different sampling rates and timestamps, time synchronization is required, which can be achieved through interpolation, resampling or timestamp alignment. The image data and radar data also have different coordinate systems and resolutions, so they need to be converted to the same coordinate system and properly spatially aligned to ensure that the first environment data and the second environment data are in the same time and space, so that the fused target obstacle avoidance data can more accurately express the obstacle distribution and obstacle movement trend in the current environment of the delivery robot, thereby providing strong data support for path planning, and can adjust the motion state of the delivery robot in real time to cope with the congestion in the current environment.

[0101] 103. Construct a priority system and a path influencing factor, and generate a target path based on the priority system, the path influencing factor and the target obstacle avoidance data.

[0102] Specifically, the target delivery task is a delivery task that the delivery robot performs according to the work needs of the hospital. It can be the delivery of medicines for patients, the delivery of medical supplies required for medical work, etc. The delivery task will be set with a starting position and an end position. The path planning of how to reach the end position from the starting position requires path planning through an AI model. Delivery tasks have different urgency, so the priority of delivery tasks needs to be marked. Before generating the target path of the delivery robot, the target path of the delivery robot corresponding to the delivery task with a higher priority is generated first. Then, in order to avoid intersection and overlap of the target path of the delivery robot, when generating the target path of the delivery task with a lower priority, the robot in the target path with a higher priority is marked as an obstacle. The delivery robot determines the order of movement according to the task priority.

[0103] Furthermore, for special circumstances, such as when a patient's condition worsens and medical supplies are urgently needed, when the priority of the delivery task changes from low to high, the path planning of the corresponding delivery robot will be adjusted to the highest priority, and the target path of the highest priority delivery robot will be readjusted. The target paths of other delivery robots will be readjusted after the target path of the highest priority delivery robot is generated. The sensor-5G network transmission-cloud platform data transmission and data processing method adopted in this application can quickly respond to the adjustment of the target path due to changes in priority, and meet the requirements of timeliness and flexibility of delivery tasks in hospital scenarios.

[0104] As an optional implementation, in the first aspect of the present invention, constructing a priority system and a path influence factor, and generating a target path based on the priority system, the path influence factor and the target obstacle avoidance data, includes:

[0105] Determine the starting point and end point of the initial path according to the target delivery task;

[0106] The priority label of the target delivery task is constructed by the first urgency of the material priority classification, the second urgency of the patient-related materials and medical care-related materials, and the third urgency of the delivery time. The priority system is established through the priority label to mark the priority of the target delivery task;

[0107] Use the delivery robot’s historical movement data to train the path planning model;

[0108] The path planning model prioritizes the target paths generated by the delivery robot through the priority marking, and the target paths generated first will serve as obstacles for the target paths generated later;

[0109] The obstacle map is generated by environmental modeling based on the target obstacle avoidance data, and a path search algorithm is used to generate several path points according to the starting point and end point of the initial path. Several optional paths are generated by sequentially connecting the starting point, path points and end point.

[0110] As an optional implementation, in the first aspect of the present invention, generating a target path based on the priority system, the path impact factor and the target obstacle avoidance data includes:

[0111] The path length, path smoothness, safe distance between obstacles, number of turns and congestion level of the path area are configured as path influencing factors in each optional path, and a corresponding weight is assigned to each of the path influencing factors. The evaluation value of each path influencing factor is calculated by weighted calculation, and the evaluation score of the corresponding optional path is output. The optional path with the highest evaluation score is determined as the target path for the delivery robot to perform the target delivery task.

[0112] Specifically, the path planning model can be constructed using optimization algorithms such as reinforcement learning, genetic algorithms or ant colony algorithms, and the historical mobile data is the historical data of the delivery robot performing delivery tasks collected as a training set to train the path planning model. In this embodiment, the genetic algorithm is taken as an example. First, the delivery tasks are sorted by priority labels to determine the order in which different delivery robots generate paths. Then, the starting point, end point and obstacle information are determined by the initial path and target obstacle avoidance data to generate environmental information (such as a grid map). A set of initial solutions is randomly generated, each solution represents an optional path (i.e., an individual in the genetic algorithm), and each gene in each solution represents a point in the optional path (i.e., the row and column coordinates of the grid map). The fitness function is defined based on factors such as the length of the path and whether it intersects with obstacles. The higher the fitness function value, the better the path. Each solution (path) is evaluated and its fitness value is calculated. The optional paths can be further screened through selection operations, mutation operations, and crossover operations. Taking the selection operation as an example, according to the value of the fitness function, a part of excellent individuals are selected as parents to generate the next generation of individuals, thereby screening out the target path. In the selection of the target path, the path influencing factors are added as the evaluation values ​​of the optional paths, such as the length of the path, the possibility of passing through crowded areas, the distance to obstacles, the number of turns, etc., to increase the feasibility and accuracy of the generated target path.

[0113] 104. Control the movement of the delivery robot according to the target path, and continuously optimize and adjust the target path.

[0114] As an optional implementation, in the first aspect of the present invention, the continuously optimizing and adjusting the target path includes:

[0115] The motion state data of the delivery robot moving in the target path is collected in real time by radar sensors, and the environmental data of the delivery robot moving in the target path is collected in real time by visual sensors. The offset between the actual path of the delivery robot and the target path is calculated based on the environmental data and the motion state data, and the motion state of the delivery robot is continuously optimized based on the offset until the correction of the target path is completed.

[0116] Specifically, on one aspect, when the delivery robot is performing a delivery task specified by the target path, there may be a deviation between the actual motion state of the target path and the specified motion state. After the deviation occurs, the target path is no longer applicable to the current environment of the delivery robot. The actual effect of path planning will reduce the accuracy and real-time performance of completing the delivery task. On the other hand, when the delivery robot is performing a delivery task specified by the target path, due to changes in the real-time environment, there may be a better target path to increase the accuracy and real-time performance of the robot in completing the delivery task. Therefore, it is necessary to continuously optimize the motion state of the delivery robot. After generating the target path, the robot can again transmit the real-time data collected by the sensor to the cloud platform through the low-latency 5G network. The cloud platform responds to and processes the real-time data fed back by the sensor in real time to regenerate a new target path or correct the old target path. The delivery robot continues to be controlled to complete the delivery task through the corrected motion parameters, thereby improving the real-time and accuracy of the delivery robot in performing delivery tasks when facing complex road conditions, avoiding delivery delays to ensure the stable development of medical work, and solving the path planning problem of the delivery robot in the hospital scenario.

[0117] The present invention uses sensors to collect image data and perception data in real time and transmits them to a cloud platform for analysis. It relies on cloud computing and big data technology to quickly integrate and process environmental data fed back in real time by visual sensors and radar sensors, and identifies the traffic congestion status in the hospital in real time as an obstacle avoidance basis for generating a target path. It constructs a priority system and path influencing factors as screening conditions for the target path, outputs the optimal target path through a path planning model to control the robot to perform the delivery task, and continuously optimizes the path parameters during the movement of the robot to ensure that the delivery robot completes the delivery task efficiently and accurately, improves the real-time and accuracy of the delivery robot in performing delivery tasks when facing complex road conditions, avoids delivery delays to ensure the stable development of medical work, and solves the path planning problem of the delivery robot in the hospital scenario.

[0118] like Figure 2 As shown, the second aspect of the present invention discloses a robot path planning system, the system comprising:

[0119] Communication module, sensor module, cloud platform module, path planning module and central control module. The path planning module is deployed on the cloud platform module, and each module performs real-time data transmission and data interaction through the communication module.

[0120] Communication module, sensor module, cloud platform module, path planning module and central control module. The path planning module is deployed on the cloud platform module, and each module performs real-time data transmission and data interaction through the communication module.

[0121] The sensor module is used to collect image data and transmit perception data to the cloud platform;

[0122] The cloud platform module is used to analyze the image data and perception data, and fuse the analysis results to generate target obstacle avoidance data;

[0123] The path planning module is used to construct a priority system and path influencing factors, and generate a target path based on the priority system, path influencing factors and target obstacle avoidance data;

[0124] The central control module is used to control the movement of the delivery robot according to the target path and continuously optimize and adjust the target path.

[0125] Specifically, the communication module in the present application adopts 5G communication as a preferred wireless network data transmission method. Other wireless network communication methods can also solve the problem that the path planning of the robot in the hospital scenario in Example 1 of the present application cannot meet the real-time and accuracy of delivery. However, in order to achieve the technical effect of real-time adjustment of the target path of the delivery robot, the wireless transmission method of the 5G network is the best choice.

[0126] It should be explained that in order to continuously optimize the robot walking on the planned route and monitor the robot's motion status on a daily basis, the delivery robot can deploy various sensors. For example, in addition to radar sensors, the delivery robot can also be equipped with positioning sensors (such as GPS, Beidou and other satellite positioning sensors or other positioning sensors) and motion sensors (such as gyroscopes, odometers, speedometers, accelerometers, etc.). The sensors can obtain the robot's position and motion status in real time. The motion status can be measured by measuring the delivery robot's moving speed, moving direction, moving acceleration, moving angular velocity and posture data. By monitoring the robot's motion status, the robot's path planning can be better and continuously optimized and controlled.

[0127] As an optional implementation, in the second aspect of the present invention, the analyzing the image data and the perception data includes:

[0128] The image recognition model was trained using historical traffic data from the hospital’s public areas;

[0129] Preprocessing and feature extraction are performed on the image data to generate a feature map, wherein the preprocessing includes noise elimination, outlier removal, and normalization processing;

[0130] The feature map is used as an input of the model, and a recognition result of the target obstacle in the target area is output as the first environmental data, wherein the first environmental data includes the first type, the first relative coordinate, the first motion state and the first environmental information of the target obstacle in the target area environment during the real-time acquisition time of the visual sensor;

[0131] Signal processing technology is used to perform preprocessing operations such as signal filtering, signal amplification, signal compression and signal enhancement on the perception data, and feature extraction and feature classification are performed on the preprocessed perception data through pattern recognition. The recognition result of the surrounding environment of the delivery robot is output as the second environmental data. The second environmental data includes the second type, second relative coordinates, second motion state and obstacle distribution map of the target obstacle within the real-time time collected by the radar sensor.

[0132] As an optional implementation, in the second aspect of the present invention, the cloud platform module generates target obstacle avoidance data by integrating the analysis results, including:

[0133] Performing a time synchronization operation and a space registration operation on the first environment data and the second environment data, wherein the time synchronization operation includes interpolation processing, resampling processing, and timestamp alignment, and the space registration operation includes performing coordinate transformation on the first relative coordinate and the second relative coordinate through the same coordinate system to obtain the absolute coordinates of the target obstacle in the target obstacle avoidance data;

[0134] The target obstacle type is determined by overlapping the first type and the second type, the position of the obstacle is confirmed based on the absolute coordinates, the movement trend of the target obstacle is predicted according to the first motion state and the second motion state, and the environmental information and the obstacle distribution map are integrated to generate an environmental perception map around the delivery robot in a mobile state as target obstacle avoidance data, wherein the movement trend includes the moving speed, moving angle, angular velocity and motion posture of the target obstacle.

[0135] As an optional implementation, in the second aspect of the present invention, the path planning module constructs a priority system and a path influence factor, and generates a target path based on the priority system, the path influence factor and the target obstacle avoidance data, including:

[0136] Determine the starting point and end point of the initial path according to the target delivery task;

[0137] The priority label of the target delivery task is constructed by the first urgency of the material priority classification, the second urgency of the patient-related materials and medical care-related materials, and the third urgency of the delivery time. The priority system is established through the priority label to mark the priority of the target delivery task;

[0138] Use the delivery robot’s historical movement data to train the path planning model;

[0139] The path planning model prioritizes the target paths generated by the delivery robot through the priority marking, and the target paths generated first will serve as obstacles for the target paths generated later;

[0140] The obstacle map is generated by environmental modeling based on the target obstacle avoidance data, and a path search algorithm is used to generate several path points according to the starting point and end point of the initial path. Several optional paths are generated by sequentially connecting the starting point, path points and end point.

[0141] As an optional implementation, in the second aspect of the present invention, the path planning module generates a target path based on the priority system, the path influencing factor and the target obstacle avoidance data, including:

[0142] The path length, path smoothness, safe distance between obstacles, number of turns and congestion level of the path area are configured as path influencing factors in each optional path, and a corresponding weight is assigned to each of the path influencing factors. The evaluation value of each path influencing factor is calculated by weighted calculation, and the evaluation score of the corresponding optional path is output. The optional path with the highest evaluation score is determined as the target path for the delivery robot to perform the target delivery task.

[0143] As an optional implementation, in the second aspect of the present invention, the central control module continuously optimizes and adjusts the target path, including:

[0144] The motion state data of the delivery robot moving in the target path is collected in real time by radar sensors, and the environmental data of the delivery robot moving in the target path is collected in real time by visual sensors. The offset between the actual path of the delivery robot and the target path is calculated based on the environmental data and the motion state data, and the motion state of the delivery robot is continuously optimized based on the offset until the correction of the target path is completed.

[0145] A third aspect of the present invention discloses a robot path planning device, comprising:

[0146] at least one processor, and

[0147] a memory communicatively connected to the at least one processor; wherein,

[0148] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the robot path planning method as described in any one of the first aspects of the present invention.

[0149] The computer device may be a terminal, and the computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a robot path planning method is implemented. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0150] The fourth aspect of the present invention discloses a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the robot path planning method as described in any one of the first aspect of the present invention.

[0151] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of the robot path planning method. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0152] Alternatively, if the above modules of the present invention are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, RAM, ROM, magnetic disks or optical disks.

[0153] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0154] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0155] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0156] Finally, it should be noted that the robot path planning method and system disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A robot path planning method, characterized in that: The method comprises: Collect image data and perception data and transmit them to the cloud platform; The cloud platform analyzes the image data and the perception data, and generates target obstacle avoidance data by fusing the analysis results; Constructing a priority system and a path influencing factor, and generating a target path based on the priority system, the path influencing factor and the target obstacle avoidance data; The delivery robot is controlled to move according to the target path, and the target path is continuously optimized and adjusted.

2. The robot path planning method according to claim 1, characterized in that: The collected image data and the sensed data are transmitted to the cloud platform, including: The target area is determined based on the historical traffic data of the hospital's public area, and visual sensors are deployed in the target area to collect image data in real time. The network camera used as a visual sensor collects high-frame-rate image data of the environment in the target area. A radar sensor is deployed on the delivery robot to collect perception data in real time. The radar sensor collects the real-time speed and real-time angular velocity of surrounding obstacles and the distance between the obstacles and the delivery robot as the perception data; The image data and perception data are transmitted to the cloud platform via wireless network real-time communication.

3. The robot path planning method according to claim 1, characterized in that: The cloud platform analyzes the image data and the perception data, including: The image recognition model was trained using historical traffic data from the hospital’s public areas; Preprocessing and feature extraction are performed on the image data to generate a feature map, wherein the preprocessing includes noise elimination, outlier removal, and normalization processing; The feature map is used as an input of the model, and a recognition result of the target obstacle in the target area is output as the first environmental data, wherein the first environmental data includes the first type, the first relative coordinate, the first motion state and the first environmental information of the target obstacle in the target area environment during the real-time acquisition time of the visual sensor; Signal processing technology is used to perform preprocessing operations such as signal filtering, signal amplification, signal compression and signal enhancement on the perception data, and feature extraction and feature classification are performed on the preprocessed perception data through pattern recognition. The recognition result of the surrounding environment of the delivery robot is output as the second environmental data. The second environmental data includes the second type, second relative coordinates, second motion state and obstacle distribution map of the target obstacle within the real-time time collected by the radar sensor.

4. The robot path planning method according to claim 3, characterized in that: The fusion analysis results generate target obstacle avoidance data, including: Performing a time synchronization operation and a space registration operation on the first environment data and the second environment data, wherein the time synchronization operation includes interpolation processing, resampling processing, and timestamp alignment, and the space registration operation includes performing coordinate transformation on the first relative coordinate and the second relative coordinate through the same coordinate system to obtain the absolute coordinates of the target obstacle in the target obstacle avoidance data; The target obstacle type is determined by overlapping the first type and the second type, the position of the obstacle is confirmed based on the absolute coordinates, the movement trend of the target obstacle is predicted according to the first motion state and the second motion state, and the environmental information and the obstacle distribution map are integrated to generate an environmental perception map around the delivery robot in a mobile state as target obstacle avoidance data, wherein the movement trend includes the moving speed, moving angle, angular velocity and motion posture of the target obstacle.

5. The robot path planning method according to claim 1, characterized in that: The constructing of a priority system and a path influencing factor, and generating a target path based on the priority system, the path influencing factor and the target obstacle avoidance data, comprises: Determine the starting point and end point of the initial path according to the target delivery task; The priority label of the target delivery task is constructed by the first urgency of the material priority classification, the second urgency of the patient-related materials and medical care-related materials, and the third urgency of the delivery time. The priority system is established through the priority label to mark the priority of the target delivery task; Use the delivery robot’s historical movement data to train the path planning model; The path planning model prioritizes the target paths generated by the delivery robot through the priority marking, and the target paths generated first will serve as obstacles for the target paths generated later; The obstacle map is generated by environmental modeling based on the target obstacle avoidance data, and a path search algorithm is used to generate several path points according to the starting point and end point of the initial path. Several optional paths are generated by sequentially connecting the starting point, path points and end point.

6. The robot path planning method according to claim 5, characterized in that: The generating of the target path based on the priority system, the path influencing factor and the target obstacle avoidance data comprises: The path length, path smoothness, safe distance between obstacles, number of turns and congestion level of the path area are configured as path influencing factors in each optional path, and a corresponding weight is assigned to each of the path influencing factors. The evaluation value of each path influencing factor is calculated by weighted calculation, and the evaluation score of the corresponding optional path is output. The optional path with the highest evaluation score is determined as the target path for the delivery robot to perform the target delivery task.

7. The robot path planning method according to claim 1, characterized in that: The continuously optimizing and adjusting the target path includes: The motion state data of the delivery robot moving in the target path is collected in real time by radar sensors, and the environmental data of the delivery robot moving in the target path is collected in real time by visual sensors. The offset between the actual path of the delivery robot and the target path is calculated based on the environmental data and the motion state data, and the motion state of the delivery robot is continuously optimized based on the offset until the correction of the target path is completed.

8. A robot path planning system, characterized in that: The system includes: a communication module, a sensor module, a cloud platform module, a path planning module and a central control module. The path planning module is deployed on the cloud platform module, and each module performs real-time data transmission and data interaction through the communication module; The sensor module is used to collect image data and transmit perception data to the cloud platform; The cloud platform module is used to analyze the image data and perception data, and fuse the analysis results to generate target obstacle avoidance data; The path planning module is used to construct a priority system and path influencing factors, and generate a target path based on the priority system, path influencing factors and target obstacle avoidance data; The central control module is used to control the movement of the delivery robot according to the target path and continuously optimize and adjust the target path.

9. A robot path planning device, characterized in that: include: at least one processor, and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the robot path planning method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the robot path planning method according to any one of claims 1 to 7.

Citation Information

Cited By

  • Path planning method and system based on ant colony algorithm and dynamic window

    CN119204379A

  • Path planning method and device for 5G nursing robot

    CN119336038A

  • Vaccine sorting operation path optimization method

    CN119359207A

  • Robot transportation path intelligent planning system

    CN120370958A