Navigation method, device, electronic device and storage medium for delivery robot

Abstract

The embodiment of the present invention discloses a navigation method, device, electronic device and storage medium for a delivery robot. The method includes: receiving a delivery request from a target user, and determining a target map position matching the delivery request in a target map; the target map is OSM; determining a target robot matching the target map position, and obtaining the initial map position of the target robot; determining a global planning path according to the initial map position and the target map position, and sending the global planning path to the target robot so that the target robot delivers goods according to the global planning path; in response to the driving instruction of the target robot, generating a local path matching the target robot in real time, and adjusting the global planning path through the local path. The solution of the embodiment of the present invention solves the problem that robot navigation depends on the construction of a high-precision semantic map, realizes efficient and accurate navigation of the delivery robot, and improves the user experience.

Description

Technical Field

[0001] The present invention relates to the field of intelligent logistics technology, and in particular to a navigation method, device, electronic equipment and storage medium for a delivery robot. Background Art

[0002] With the rapid development of artificial intelligence and robotics, robots are gradually being used in all aspects of people’s daily lives. For example, delivery robots can significantly improve logistics efficiency, optimize customer experience, reduce operating costs, and significantly reduce dependence on manual labor.

[0003] At present, robot navigation mainly relies on building high-precision semantic maps. This method not only consumes a lot of resources, but also greatly limits the possibility of large-scale deployment.

[0004] How to achieve efficient and accurate navigation of delivery robots without relying on high-precision maps to meet users' delivery needs is a key research issue in the industry. Summary of the invention

[0005] The present invention provides a delivery robot navigation method, device, electronic device and storage medium to solve the problem that robot navigation depends on building a high-precision semantic map, and can achieve efficient and accurate navigation of the delivery robot without relying on a high-precision map, thereby improving the user experience.

[0006] According to one aspect of the present invention, a navigation method for a delivery robot is provided, the method comprising:

[0007] receiving a delivery request from a target user, and determining a target map location matching the delivery request in a target map; the target map is an Open Street Map (OSM);

[0008] Determining a target robot that matches the target map position, and obtaining an initial map position of the target robot;

[0009] Determining a global planning path according to the initial map position and the target map position, and sending the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path;

[0010] In response to the driving instruction of the target robot, a local path matching the target robot is generated in real time, and the global planning path is adjusted through the local path.

[0011] According to another aspect of the present invention, there is provided a navigation device for a delivery robot, the device comprising:

[0012] A target map location determination module, used for receiving a delivery request from a target user, and determining a target map location matching the delivery request in a target map; the target map is OSM;

[0013] An initial map position determination module, used to determine a target robot that matches the target map position and obtain the initial map position of the target robot;

[0014] A global planning path determination module, used to determine a global planning path according to the initial map position and the target map position, and send the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path;

[0015] The local path generation module is used to generate a local path matching the target robot in real time in response to the driving instruction of the target robot, and adjust the global planned path through the local path.

[0016] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the navigation method of the delivery robot described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the navigation method of the delivery robot described in any embodiment of the present invention when executed by a processor.

[0021] The technical solution of the embodiment of the present invention receives a delivery request from a target user, and determines a target map position matching the delivery request in a target map; determines a target robot matching the target map position, and obtains the initial map position of the target robot; determines a global planning path according to the initial map position and the target map position, and sends the global planning path to the target robot, so that the target robot delivers goods according to the global planning path; generates a local path matching the target robot in real time in response to the driving instruction of the target robot, and adjusts the global planning path through the local path, thereby solving the problem that robot navigation depends on building a high-precision semantic map, and can achieve efficient and accurate navigation of the delivery robot without relying on a high-precision map. The user experience is improved.

[0022] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0024] Figure 1 is a flowchart of a navigation method for a delivery robot provided according to Embodiment 1 of the present invention;

[0025] Figure 2 is a flow chart of a navigation method for a delivery robot provided according to Embodiment 2 of the present invention;

[0026] Figure 3 is a schematic diagram of determining an initial map position according to Embodiment 2 of the present invention;

[0027] Figure 4 is a flow chart of a navigation method for a delivery robot provided according to Embodiment 3 of the present invention;

[0028] Figure 5 is a flow chart of a navigation method for a delivery robot provided according to Embodiment 3 of the present invention;

[0029] Figure 6 is a structural schematic diagram of a navigation device of a delivery robot provided according to a fourth embodiment of the present invention;

[0030] Figure 7 It is a schematic diagram of the structure of an electronic device for implementing the navigation method of the delivery robot according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] Embodiment 1

[0034] Figure 1 This is a flow chart of a delivery robot navigation method provided according to the first embodiment of the present invention. This embodiment is applicable to the situation where a delivery robot is navigated without relying on a high-precision semantic map. The method can be executed by a delivery robot navigation device, which can be implemented in the form of hardware and / or software. The delivery robot navigation device can be configured in a computer, server, cloud server or other equipment, or in a robot, which is not limited in this embodiment. Figure 1 As shown, the method includes:

[0035] Step 110: Receive a delivery request from a target user, and determine a target map location matching the delivery request in a target map.

[0036] The target user may be any one or more users, which are not limited in this embodiment. In this embodiment, the target user may send a delivery request for the goods to be delivered to the cloud server through an application installed in the mobile terminal; the target user may also send a delivery request for the goods to be delivered to the cloud server through a small program in the target application in the mobile terminal, which are not limited in this embodiment.

[0037] It should be noted that, in this embodiment, the delivery request sent by the target user can be a voice request or a text request, which is not limited in this embodiment. It should also be noted that, in this embodiment, the target map location can be the XX building of the XX community displayed in the target map, or it can be the XX office building, etc., which is not limited in this embodiment.

[0038] Optionally, in this embodiment, after receiving the delivery request issued by the target user, the target map location matching the delivery request issued by the target user can be further determined in the target map, that is, the destination location in the delivery request can be queried in the target map.

[0039] The target map may be OSM, which is a collaborative online map project for constructing free content. Its goal is to create a world map with free content that can be edited by everyone, and to provide convenient navigation solutions for generally inexpensive mobile devices. OSM includes spatial data and attribute data, among which spatial data mainly includes three types: Nodes, Ways, and Relations. These three types of original spatial data constitute the entire map screen. Nodes define the position of points in space; Ways define lines or areas; Relations (optional) define the relationship between elements. Attribute data is used to describe vector data primitives.

[0040] In an optional implementation of the present embodiment, receiving a delivery request issued by a target user and determining a target map location matching the delivery request in a target map may include: parsing the delivery request through a large language model to obtain a parsing result; the parsing result includes a destination address corresponding to the delivery request; and querying the destination address in the target map to obtain the target map location.

[0041] In an optional implementation of this embodiment, after receiving the delivery request sent by the target user, the delivery request can be parsed by a large language model to obtain a parsing result including the destination address information; further, the location of the destination address can be queried in the target map and the location can be marked, and the marked location is the target map location.

[0042] It should be noted that in this embodiment, if the delivery request sent by the target user is a voice request, the voice request can be converted into text information and then processed by the large language model; if the delivery request sent by the target user is a text request, it can be directly processed by the large language model.

[0043] Step 120: Determine a target robot that matches the target map position, and obtain an initial map position of the target robot.

[0044] In an optional implementation of this embodiment, after determining the target map location that matches the delivery request issued by the target user in the target map, the target robot that matches the target map location can be further determined, and the initial map location of the target robot can be obtained.

[0045] Optionally, in the present embodiment, determining the target robot that matches the target map location may include: determining the target delivery station that is the shortest distance from the target map location, and determining an idle robot in the target delivery station as the target robot; and / or determining a relevant area of ​​the target map location, and determining the location information of each robot within the relevant area; wherein the relevant area is a circular area with the target map location as the center and a preset length as the radius; determining the target location information closest to the target map location based on the location information of each robot; and determining the robot corresponding to the target location information as the target robot.

[0046] In an optional implementation of this embodiment, after determining the target map location that matches the delivery request issued by the target user in the target map, the target delivery station with the shortest distance from the target map location can be further determined in the target map, and an idle robot in the target delivery station can be determined as the target robot.

[0047] The advantage of this setting is that the target robot can be quickly determined. Since the selected target robot is an idle robot, the delivery efficiency of the delivery request issued by the target user can be improved, providing a basis for improving the user experience.

[0048] In another optional implementation of the present embodiment, after determining the target map location that matches the delivery request issued by the target user in the target map, the relevant area of ​​the target map location can be further determined, and the location information of each robot in the relevant area can be determined; illustratively, the relevant area of ​​the target map location can be a circular area formed by taking the target map location as the center and a preset length (for example, 200 meters, 500 meters, or 1 kilometer, etc., which is not limited in the present embodiment) as the radius; further, the robot at the target location information closest to the target map location can be determined as the target robot.

[0049] In an example of this embodiment, if the target map location is Building 1 in Community A, and the relevant area of ​​the target map location is Community A, then the robots in Community A and the location information of each robot can be further determined, and the robot closest to Building 1 can be determined as the target robot.

[0050] The advantage of this setting is that it can determine the robot closest to the destination, which can improve the efficiency of cargo delivery and provide a basis for improving user experience.

[0051] In another optional implementation of this embodiment, after determining the target robot that matches the target map location, the delivery request can also be sent directly to the target robot, so that the target robot generates a navigation path according to the target map location and its own initial map position in the target map, thereby completing the cargo delivery.

[0052] Optionally, in this embodiment, after determining that a target robot matching the target map position is obtained, the position of the target robot in the target map may be further determined, which is referred to as an initial map position in this embodiment.

[0053] Step 130: determine a global planning path according to the initial map position and the target map position, and send the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path.

[0054] In an optional implementation of the present embodiment, after determining the target map position of the destination address and the initial map position of the target robot, the global planning path for the target robot to deliver the goods can be further determined based on the target map position and the initial map position, and the global planning path can be sent to the target robot; further, the target robot can deliver the goods according to the received global planning path.

[0055] Optionally, in this embodiment, the global planning path is determined according to the initial map position and the target map position, including: determining the road network structure of the initial map position and the target map position in the target map, and preprocessing each of the road network structures to obtain each target road network structure; generating the global planning path according to each of the target road network structures.

[0056] It is understandable that the OSM map may be composed of multiple road network structures, and multiple road network structures may be included from the initial map position to the target map position. The final global planning path can be formed by selecting a suitable road network structure.

[0057] In an optional implementation of the present embodiment, each road network structure from the initial map position to the target map position may be determined first, and each road network structure may be pre-processed, for example, by filtering out obstacles, maintenance sections or congested areas in each road network structure, which are not limited in the present embodiment, to obtain at least one target road network structure; finally, a global planning path may be generated based on each target road network structure; in an example of the present embodiment, after obtaining each target road network structure, adjacent target road network structures may be spliced ​​in pairs to generate the final global planning path.

[0058] In the specific implementation, the Contraction Hierarchies algorithm can be used to efficiently calculate the optimal path from the target map location to the initial map location. This algorithm significantly speeds up the path query by preprocessing the road network structure of the OSM map, and can quickly generate a global path even in a complex urban environment, providing efficient and reliable navigation guidance for the robot.

[0059] Step 140: In response to the driving instruction of the target robot, a local path matching the target robot is generated in real time, and the global planned path is adjusted by the local path.

[0060] In an optional implementation of this embodiment, after receiving the global planned path, the target robot can start delivering goods according to the global planned path; after receiving the instruction for the target robot to start driving, that is, the instruction for the target robot to start delivering goods, a local path matching the target robot can be generated in real time, and the global planned path can be adjusted through the local path.

[0061] It can be understood that in this embodiment, the global planning path refers to the driving path from the initial map position to the target map position, which is a longer driving path; and the local path refers to the driving path from the current position of the target robot to the next position (for example, the position of a reference object), which is a shorter driving path.

[0062] The solution of this embodiment can continuously generate shorter local paths during the driving process of the target robot, and can timely revise the global planned path to prevent emergencies such as traffic accidents from occurring on the driving route of the global planned path and affecting delivery.

[0063] The technical solution of this embodiment receives a delivery request from a target user, and determines a target map position matching the delivery request in a target map; determines a target robot matching the target map position, and obtains the initial map position of the target robot; determines a global planning path according to the initial map position and the target map position, and sends the global planning path to the target robot, so that the target robot delivers goods according to the global planning path; generates a local path matching the target robot in real time in response to the driving instruction of the target robot, and adjusts the global planning path through the local path, thereby solving the problem that robot navigation depends on building a high-precision semantic map, and can achieve efficient and accurate navigation of the delivery robot without relying on a high-precision map. The user experience is improved.

[0064] Embodiment 2

[0065] Figure 2 1 is a flow chart of a navigation method for a delivery robot provided according to Embodiment 2 of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0066] Step 210: Receive a delivery request from a target user, and determine a target map location matching the delivery request in a target map.

[0067] Step 220: Determine a target robot that matches the target map position.

[0068] Step 230: acquiring a target image based on the image sensor of the target robot; performing instance segmentation on the target image according to the text information of the target robot in the target map, and performing semantic filtering processing on the instance segmentation result and the point cloud data matching the target image to obtain target point cloud data; projecting the target point cloud data onto a preset plane to obtain a projection result; and aligning the projection result with the target map to obtain an initial map position of the target robot.

[0069] The preset plane may be a bird's-eye view plane or a plane where an OSM map is located, which is not limited in this embodiment.

[0070] In an optional implementation of the present embodiment, after determining that a target robot matching the position of the target map is obtained, an image of the surrounding environment of the target robot captured by an image sensor deployed in the target robot, i.e., the target image in the present embodiment, can be obtained; further, the target image can be instance segmented according to text information of the target robot in the target map, wherein the text information of the target robot in the target map is the text information of the approximate position of the target robot in the target map, for example, Street A, Community B, or Intersection C, etc., which is not limited in the present embodiment.

[0071] Furthermore, semantic filtering processing can be performed based on the instance segmentation results and the point cloud data matching the target image to obtain target point cloud data; further, the target point cloud data can be projected, for example, the target point cloud data is projected onto a bird's-eye view plane; further, the projection result is aligned with the target map to obtain the initial map position of the target robot on the target map.

[0072] Figure 3 is a schematic diagram of determining an initial map position according to Embodiment 2 of the present invention. Figure 3 As shown, in this embodiment, the target image collected by the target robot can be input into the trained MobileSAM model or other image segmentation model for segmentation, thereby obtaining the masks of various objects in the target image; at the same time, the text information of each object is extracted from OSM and encoded by the CLIP model; further, in the instance segmentation module, the CLIP model is used to encode various objects in the target image, and the cosine similarity between each object encoding and each text encoding is calculated, and finally the text with the highest similarity is used as the semantic label of the object, thereby realizing instance segmentation. Further, in the semantic filtering module, the system projects the point cloud data onto the target image to obtain semantic information, and retains the point cloud data with semantic labels. Subsequently, the retained point cloud data is projected onto the bird's-eye view plane. Finally, the geometric shape information in OSM is used for two-dimensional ICP (iterative closest point) registration to obtain the precise global pose of the robot in the OSM map.

[0073] Step 240: determine a global planning path according to the initial map position and the target map position, and send the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path.

[0074] Step 250: In response to the driving instruction of the target robot, a local path matching the target robot is generated in real time, and the global planned path is adjusted through the local path.

[0075] According to the scheme of this embodiment, after determining the target robot for delivering goods, a target image can be acquired based on the image sensor of the target robot; instance segmentation is performed on the target image according to the text information of the target robot in the target map, and semantic filtering is performed according to the instance segmentation result and the point cloud data matching the target image to obtain target point cloud data; the target point cloud data is projected onto a preset plane to obtain a projection result; the projection result is aligned with the target map to obtain the initial map position of the target robot, which can solve the problem of deviation between the global positioning system and the OSM coordinate system, and the problem of unstable global positioning system signal in areas such as dense buildings, and can accurately determine the initial map position of the target robot in the target map, providing a basis for the subsequent generation of an accurate global planning path.

[0076] Embodiment 3

[0077] Figure 4 1 is a flow chart of a navigation method for a delivery robot provided according to Embodiment 3 of the present invention. This embodiment is a further refinement of the above technical solution. The technical solution in this embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 4 As shown, the method includes:

[0078] Step 410: Receive a delivery request from a target user, and determine a target map location matching the delivery request in a target map.

[0079] Step 420: Determine a target robot that matches the target map position, and obtain an initial map position of the target robot.

[0080] Step 430: determine a global planning path according to the initial map position and the target map position, and send the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path.

[0081] Step 440: In response to the driving instruction of the target robot, a local path matching the target robot is generated in real time, and the global planned path is adjusted through the local path.

[0082] In an optional implementation of the present embodiment, a local path matching the target robot is generated in real time, and the global planned path is adjusted through the local path, which may include: acquiring the current position information of the target robot in real time, and generating a local path based on the current position information and the target reference position information in the global planned path; determining the attribute information of the local path and the global planned path respectively; and replacing the target driving path of the section in the global planned path that matches the local path with the local path when the attribute information of the global planned path does not meet a preset condition, so that the target robot navigates according to the local path.

[0083] Among them, the target reference position information is the position information of the reference object that is closest to the current position information among the reference position information in the global planning path; the reference position information may include at least one of the following: the position information of the landmark, the position information of the traffic light, and the intersection position information, etc., which is not limited in this embodiment.

[0084] The attribute information may include at least one of the following: whether there is an obstacle, the type of obstacle, the number of traffic lights, and whether a traffic accident has occurred, etc., which is not limited in this embodiment.

[0085] Among them, the preset condition can be a comparison condition, for example, the attribute information of the global planning path is better than the local path; it can also be a limitation condition, for example, the global planning path does not contain obstacles, etc., which is not limited in this embodiment.

[0086] Optionally, in this embodiment, while the target robot is traveling to the destination address, i.e., the target map location, the current position information of the target robot can be acquired in real time, and a local path can be generated based on the current position information and the reference position information in the global planning path that is closest to the current position information; exemplarily, if the reference object closest to the current position information is the first traffic light, then a first local path can be generated based on the current position information and the position information of the first traffic light.

[0087] Furthermore, the attribute information of the first local path and the part of the global planned path that overlaps with the first local path (the starting points overlap, but the paths do not overlap) can be determined; for example, it can be determined whether there are obstacles, the types of obstacles, the number of traffic lights, or whether a traffic accident has occurred in the first local path and the part of the global planned path that overlaps with the first local path.

[0088] Furthermore, the attribute information of the first local path and the part of the global planned path that overlaps with the first local path can be compared to determine whether the attribute information of the part of the global planned path that overlaps with the first local path is better than the attribute information of the first local path, for example, whether the number of traffic lights is less than the number of traffic lights in the first local path, and whether the number of obstacles is less than the number of obstacles in the first local path; if it is determined that the attribute information of the part of the global planned path that overlaps with the first local path is not better than the attribute information of the first local path, that is, it is determined that the attribute information of the global planned path does not meet the preset conditions, then the part of the global planned path that overlaps with the first local path can be replaced with the first local path.

[0089] It is understandable that in this embodiment, global path planning provides a reference path for local path planning, guiding the robot to move towards the target direction. Local path planning can make real-time adjustments to the global planning path to cope with unexpected obstacles and other uncertain factors. The combination of the two enables the robot to effectively plan routes in a large area and flexibly respond to changes in the local environment.

[0090] The advantage of this setting is that the pre-planned global planning path can be adjusted in real time, which can prevent emergencies from causing delivery delays and affecting customer experience.

[0091] In a specific example of this embodiment, the FastLIO algorithm can be used to estimate the local odometer to ensure real-time and accurate tracking of the robot's position changes. The local path planning can optimize and correct the global trajectory through the Time Elastic Band (TEB) algorithm to generate a local motion trajectory that is more in line with the actual environment and the robot's motion characteristics. In the control module, the system selects a suitable controller based on the robot's chassis model to ensure that the robot can perform smooth and accurate motion control according to the optimized path.

[0092] Step 450: During the driving process of the target robot, a driving image fed back in real time by the target robot is obtained; whether the driving image contains a preset object of interest is detected; if the target driving image contains the target object of interest, first position information of the target object of interest in the target map is determined according to the target driving image and text information of the target object of interest; and the target map is updated according to the first position information.

[0093] In an optional implementation of the present embodiment, the navigation method of the delivery robot may further include: during the driving process of the target robot, obtaining a driving image that is fed back in real time by the target robot; detecting whether the driving image contains a preset object of interest; if the target driving image contains the target object of interest, determining the first position information of the target object of interest in the target map based on the target driving image and the text information of the target object of interest; and updating the target map based on the first position information.

[0094] The object of interest may be a telephone pole, a traffic light, an obstacle sign, or a camera, etc., which is not limited in this embodiment.

[0095] Optionally, during the driving process of the target robot, driving images can be captured in real time and the captured driving images can be fed back to the server; after receiving the driving images fed back by the target robot, the server can detect whether the driving images contain objects of interest such as telephone poles, traffic lights, obstacle signs or cameras; if it is determined that the target driving image (any driving image) contains an object of interest, the first position information of the target object of interest in the target map can be determined based on the target driving image and the text information of the target object of interest; further, the target map can be updated according to the first position information.

[0096] Optionally, in an optional implementation of the present embodiment, determining the first position information of the target object of interest in the target map based on the target driving image and the text information of the target object of interest may include: performing instance segmentation on the target driving image according to the text information of the target object of interest; and performing semantic filtering processing based on the instance segmentation result and the point cloud data matching the target driving image to obtain target point cloud data; projecting the target point cloud data onto a bird's-eye view plane to obtain a projection result; and aligning the projection result with the target map to obtain the first position information.

[0097] It should be noted that the above process is similar to the process of determining the initial map position, the difference is that in the above steps, the focus is on the object of interest. First, the object of interest is segmented and the corresponding point cloud data is projected onto the bird's-eye view plane. Further, it is updated to OSM to ensure the accuracy and real-time nature of the map. This process helps to better reflect the actual situation of the environment and provide richer and more reliable data support for subsequent path planning and navigation.

[0098] It should also be noted that in this embodiment, step 440 is introduced first and then step 450. This is only for the convenience of introduction and does not represent the sequence of steps 440 and 450. In some embodiments, step 450 may occur at the same time as step 440. This is not a limitation of this embodiment.

[0099] The solution of this embodiment can also obtain a driving image of the target robot that is fed back in real time during the driving process of the target robot; detect whether the driving image contains a preset object of interest; if the target driving image contains the target object of interest, determine the first position information of the target object of interest in the target map according to the target driving image and the text information of the target object of interest; and update the target map according to the first position information, thereby ensuring the accuracy and real-time performance of the target map and providing richer and more reliable data support for subsequent path planning and navigation.

[0100] In order to better understand the navigation method of the delivery robot involved in this embodiment, Figure 5 is a flow chart of a navigation method for a delivery robot provided according to Embodiment 3 of the present invention. Figure 5 As shown, it mainly includes the following:

[0101] 1. Task planning based on LLM: First, use the large language model to parse the text-based delivery request into several specific delivery addresses, then query the approximate location of the target address through the OSM interface, and finally send multiple tasks to the robot; in this embodiment, a laser radar camera is deployed in the robot to collect images.

[0102] 2. Global positioning based on visual basic model: In order to solve the possible deviation problem between GPS and OSM map coordinate systems, as well as the problem of unstable GPS signal in areas with dense buildings, the present invention is based on the visual basic model and realizes global positioning directly on the OSM map, that is, determines the initial map position of the target robot in OSM.

[0103] 3. Global path planning based on OSM: After determining the delivery target, global path planning is performed based on the OSM map. Specifically, the Contraction Hierarchies algorithm is used to efficiently calculate the optimal path from the starting point to the target location. This algorithm significantly speeds up path queries by preprocessing the road network structure of the OSM map. It can quickly generate global paths even in complex urban environments, providing efficient and reliable navigation guidance for the robot.

[0104] 4. OSM online update: First, the objects of interest are segmented and the corresponding point cloud data is projected onto the bird's-eye view plane. Then, it is updated to the OSM map to ensure the accuracy and real-time performance of the map. This process helps to better reflect the actual situation of the environment and provide richer and more reliable data support for subsequent path planning and navigation.

[0105] 5. Local navigation: In terms of local positioning, the FastLIO algorithm is used to estimate the local odometer to ensure real-time and accurate tracking of the robot's position changes. The local path planning uses the TEB algorithm to optimize and correct the global trajectory, thereby generating a local motion trajectory that is more in line with the actual environment and the robot's motion characteristics. In the control module, the system selects a suitable controller based on the robot's chassis model to ensure that the robot can perform smooth and precise motion control according to the optimized path.

[0106] The solution of this embodiment adopts OSM as a lightweight and flexible map representation form, which does not require map construction in advance. It can dynamically interpret and adapt to various outdoor environments in real time, reduce deployment costs, and provide a more efficient solution for the field of autonomous delivery.

[0107] Embodiment 4

[0108] Figure 6 Schematic diagram of the structure of a navigation device for a delivery robot according to Embodiment 4 of the present invention. Figure 6 As shown, the device includes: a target map position determination module 610, an initial map position determination module 620, a global planning path determination module 630 and a local path generation module 640.

[0109] The target map location determination module 610 is used to receive a delivery request from a target user and determine a target map location matching the delivery request in a target map; the target map is OSM;

[0110] An initial map position determination module 620 is used to determine a target robot that matches the target map position and obtain the initial map position of the target robot;

[0111] A global planning path determination module 630 is used to determine a global planning path according to the initial map position and the target map position, and send the global planning path to the target robot so that the target robot performs cargo delivery according to the global planning path;

[0112] The local path generation module 640 is used to generate a local path matching the target robot in real time in response to the driving instruction of the target robot, and adjust the global planned path through the local path.

[0113] The solution of this embodiment receives a delivery request issued by a target user through a target map position determination module, and determines a target map position matching the delivery request in the target map; determines a target robot matching the target map position through an initial map position determination module, and obtains the initial map position of the target robot; determines a global planning path according to the initial map position and the target map position through a global planning path determination module, and sends the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path; generates a local path matching the target robot in real time through a local path generation module in response to the driving instruction of the target robot, and adjusts the global planning path through the local path, thereby solving the problem that robot navigation depends on building a high-precision semantic map, realizing efficient and accurate navigation of the delivery robot, and improving user experience.

[0114] In an optional implementation of this embodiment, the target map location determination module 610 is specifically configured to parse the delivery request through a large language model to obtain a parsing result; the parsing result includes a destination address corresponding to the delivery request;

[0115] The destination address is queried in the target map to obtain the target map location.

[0116] In an optional implementation of this embodiment, the initial map position determination module 620 is specifically used to determine the target delivery station with the shortest distance from the target map position, and determine an idle robot in the target delivery station as the target robot; and / or,

[0117] Determine the relevant area of ​​the target map position, and determine the position information of each robot in the relevant area; wherein the relevant area is a circular area with the target map position as the center and a preset length as the radius;

[0118] Determine the target position information closest to the target map position according to the position information of each of the robots;

[0119] A robot corresponding to the target position information is determined as a target robot.

[0120] In an optional implementation of this embodiment, the initial map position determination module 620 is further specifically configured to acquire a target image based on an image sensor of the target robot;

[0121] Performing instance segmentation on the target image according to the text information of the target robot in the target map, and performing semantic filtering processing on the instance segmentation result and the point cloud data matching the target image to obtain target point cloud data;

[0122] Projecting the target point cloud data onto a preset plane to obtain a projection result;

[0123] The projection result is aligned with the target map to obtain an initial map position of the target robot.

[0124] In an optional implementation of this embodiment, the global planning path determination module 630 is specifically used to determine the road network structure of the initial map position and the target map position in the target map, and pre-process each of the road network structures to obtain each target road network structure;

[0125] The global planning path is generated according to each of the target road network structures.

[0126] In an optional implementation of this embodiment, the local path generation module 640 is specifically used to obtain the current position information of the target robot in real time, and generate a local path according to the current position information and the target reference position information in the global planning path;

[0127] Determine attribute information of the local path and the global planned path respectively; the attribute information includes at least one of the following: whether there is an obstacle, the type of obstacle, the number of traffic lights, and whether a traffic accident occurs;

[0128] When the attribute information of the global planning path does not meet the preset conditions, the target driving path of the section in the global planning path that matches the local path is replaced by the local path, so that the target robot navigates according to the local path.

[0129] In an optional implementation of this embodiment, the navigation device of the delivery robot further includes: an updating module, which is used to obtain a driving image fed back in real time by the target robot during the driving process of the target robot;

[0130] Detecting whether the driving image contains a preset object of interest;

[0131] In a case where the target driving image includes a target object of interest, determining first position information of the target object of interest in the target map according to the target driving image and text information of the target object of interest;

[0132] The target map is updated according to the first location information.

[0133] The navigation device of the delivery robot provided in the embodiment of the present invention can execute the navigation method of the delivery robot provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0134] It should be noted that in the technical solutions of the embodiments of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of user personal information (such as delivery information, voice information, etc.) are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0135] Embodiment 5

[0136] Figure 7 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0137] like Figure 7 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0138] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0139] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the navigation method of the delivery robot.

[0140] In some embodiments, the navigation method of the delivery robot may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the navigation method of the delivery robot described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the navigation method of the delivery robot in any other appropriate manner (e.g., by means of firmware).

[0141] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0143] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0144] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0145] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0146] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0147] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0148] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A navigation method for a delivery robot, characterized in that: include: receiving a delivery request from a target user, and determining a target map location in a target map that matches the delivery request; The target map is an open street map OSM; Determining a target robot that matches the target map position, and obtaining an initial map position of the target robot; Determining a global planning path according to the initial map position and the target map position, and sending the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path; In response to the driving instruction of the target robot, a local path matching the target robot is generated in real time, and the global planning path is adjusted by the local path; The determining of the target robot matching the target map position includes: determining a target delivery station with the shortest distance from the target map position, and determining an idle robot in the target delivery station as the target robot.

2. The navigation method of the delivery robot according to claim 1, characterized in that: The receiving a delivery request from a target user and determining a target map location matching the delivery request in a target map includes: Parsing the delivery request using a large language model to obtain a parsing result; the parsing result includes a destination address corresponding to the delivery request; The destination address is queried in the target map to obtain the target map location.

3. The navigation method of the delivery robot according to claim 1, characterized in that: The determining of the target robot matching the target map position further includes: Determine the relevant area of ​​the target map position, and determine the position information of each robot in the relevant area; wherein the relevant area is a circular area with the target map position as the center and a preset length as the radius; Determine the target position information closest to the target map position according to the position information of each of the robots; A robot corresponding to the target position information is determined as a target robot.

4. The navigation method of the delivery robot according to claim 1, characterized in that: The obtaining of the initial map position of the target robot comprises: Acquiring a target image based on an image sensor of the target robot; Performing instance segmentation on the target image according to the text information of the target robot in the target map, and performing semantic filtering processing on the instance segmentation result and the point cloud data matching the target image to obtain target point cloud data; Projecting the target point cloud data onto a preset plane to obtain a projection result; The projection result is aligned with the target map to obtain an initial map position of the target robot.

5. The navigation method of the delivery robot according to claim 1, characterized in that: The determining of the global planning path according to the initial map position and the target map position includes: Determining the road network structure of the initial map position and the target map position in the target map, and preprocessing each of the road network structures to obtain each target road network structure; The global planning path is generated according to each of the target road network structures.

6. The navigation method of the delivery robot according to claim 1, characterized in that: The real-time generation of a local path matching the target robot and adjusting the global planning path by the local path include: Acquire the current position information of the target robot in real time, and generate a local path according to the current position information and the target reference position information in the global planning path; Determine attribute information of the local path and the global planned path respectively; the attribute information includes at least one of the following: whether there is an obstacle, the type of obstacle, the number of traffic lights, and whether a traffic accident occurs; When the attribute information of the global planning path does not meet the preset conditions, the target driving path of the section in the global planning path that matches the local path is replaced by the local path, so that the target robot navigates according to the local path.

7. The navigation method of the delivery robot according to claim 1, characterized in that: Also includes: During the driving process of the target robot, obtaining a driving image fed back in real time by the target robot; Detecting whether the driving image contains a preset object of interest; In a case where the target driving image includes a target object of interest, determining first position information of the target object of interest in the target map according to the target driving image and text information of the target object of interest; The target map is updated according to the first location information.

8. A navigation device for a delivery robot, characterized in that: include: A target map location determination module, used to receive a delivery request from a target user and determine a target map location matching the delivery request in a target map; The target map is OSM; An initial map position determination module, used to determine a target robot that matches the target map position and obtain the initial map position of the target robot; A global planning path determination module, used to determine a global planning path according to the initial map position and the target map position, and send the global planning path to the target robot, so that the target robot performs cargo delivery according to the global planning path; A local path generation module, for generating a local path matching the target robot in real time in response to a driving instruction of the target robot, and adjusting the global planning path through the local path; The initial map position determination module is specifically used to determine a target delivery station that is the shortest distance from the target map position, and determine an idle robot in the target delivery station as the target robot.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the navigation method of the delivery robot according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the navigation method of the delivery robot according to any one of claims 1 to 7 when executed.

Images