A robot path planning method and system

Abstract

The application provides a robot path planning method, belonging to the field of data processing, to realize accurate dynamic path planning at a lower cost. The method comprises: a control device acquires a first image and a second image, the first image is an environment image in the moving direction collected by a first robot at its current position, and the second image is an environment image in the moving direction collected by a second robot at its current position; the first robot moves along a preset moving route, and the second robot moves along the preset moving route following the first robot; the control device determines a dynamic route for obstacle avoidance in a prior regional map according to the current position of the first robot, the current position of the second robot, the first image and the second image; the first robot and the second robot are currently located in a region contained in the prior regional map, and the dynamic route is connected with the preset moving route; and the control device instructs the first robot and the second robot to move along the dynamic route.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a robot path planning method and system. BACKGROUND

[0002] Robot path planning is an important research field in robotics, which involves how to move a robot from a starting point to a destination while avoiding obstacles and optimizing the path to achieve efficient and safe navigation. This technology has a wide range of applications in automation, manufacturing, logistics, medical services, and personal assistants. Path planning usually involves the following key steps: Environment perception: the robot needs to obtain information about the surrounding environment through sensors such as laser radar, camera, ultrasonic sensor, etc. Map construction: convert the perceived information into a map representation, such as a two-dimensional grid map, a topological map, or a three-dimensional map. Path search: search for a path from the starting point to the destination in the map. This usually involves the selection of search algorithms, such as Dijkstra's algorithm, A* algorithm, or heuristic search algorithm. Path optimization: optimize the searched path to reduce travel time or energy consumption.

[0003] With the development of artificial intelligence and machine learning technology, path planning technology is developing towards more intelligent and adaptive direction. How to achieve lower cost and accurate dynamic path planning is the current research problem. SUMMARY

[0004] The embodiments of the present application provide a robot path planning method and system to achieve accurate dynamic path planning at a lower cost.

[0005] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0006] In a first aspect, a robot path planning method is provided, applied to a control device, the method comprising: the control device acquiring a first image and a second image, the first image being an environmental image in the direction of travel collected by a first robot at its current location, the second image being an environmental image in the direction of travel collected by a second robot at its current location, the first robot moving along a preset travel route, and the second robot following the first robot along the preset travel route; the control device determining a dynamic route for obstacle avoidance within a priori area map according to the current position of the first robot, the current position of the second robot, the first image, and the second image, the first robot and the second robot currently being located within the area contained in the priori area map, and the dynamic route being connected to the preset travel route; and the control device instructing the first robot and the second robot to move along the dynamic route.

[0007] Optionally, the control device determines a dynamic route for obstacle avoidance within a priori region map based on the current position of the first robot, the current position of the second robot, the first image, and the second image. This includes: the control device determining the information correlation between the first image and the second image; if the information correlation is greater than a correlation threshold, the control device determines a priori region map based on the current position of the first robot, the current position of the second robot, the first image, and the second image; and the control device determines a dynamic route within the priori region map using a fast expanding tree algorithm based on a preset path and the priori region map.

[0008] Optionally, the control device determines the information correlation between the first image and the second image, including: the control device extracts features of the first image through a neural network model, and extracts features of the second image through a neural network model; the control device determines the information correlation between the features of the first image and the features of the second image.

[0009] Optionally, the control device extracts features of the first image and features of the second image through a neural network model, including: the control device performs a target-scale convolution on the first image using a neural network model to obtain features of the first image, and performs a target-scale convolution on the second image using a neural network model to obtain features of the second image. The first image and the second image have the same resolution. The feature structure of the first image is M1*N1, and the feature structure of the second image is M2*N2, where M1 = M2, N1 = N2, M1 and M2 are the number of rows, N1 and N2 are the number of columns, and M1, M2, N1, and N2 are all integers greater than 2. The control device determines the information correlation between the features of the first image and the features of the second image, including: the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image.

[0010] Optionally, the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image, including: the control device calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image. Listed to number Vector augmentation of the column yields the first... Listed to number The augmentation vector of the column, and the features of the first image from column 1 to column 2. Column and the The vector to the N1th column is embedded into the N1th column. Listed to number The enhancement vector of the column is used to obtain the enhanced features of the first image; the control device then selects the enhanced features of the second image from the enhanced vector of the column. column to the (M1+N1)th column of the first image, and embedding the vectors in the 1st column to the N1th column of the feature of the first image and the vectors in the 1st column to the N2th column of the feature of the second image into the enhanced vectors in the (M1+N2)th column to the (M1+2N2)th column, to obtain the enhanced feature of the second image; wherein, represents rounding up; the control device determines the information correlation between the feature of the first image and the feature of the second image by calculating the correlation between each column of enhanced vectors in the enhanced feature of the first image and the corresponding column of enhanced vectors in the enhanced feature of the second image.

[0011] Optionally, the control device enhances the vectors in the 1st column to the N1th column of the feature of the first image, to obtain the enhanced vectors in the (M1+N1)th column to the (M1+2N1)th column of the first image, and embeds the vectors in the 1st column to the N1th column of the feature of the first image and the vectors in the 1st column to the N2th column of the feature of the second image into the enhanced vectors in the (M1+N2)th column to the (M1+2N2)th column, to obtain the enhanced feature of the second image; wherein, correspondingly, the control device enhances the vectors in the 1st column to the N1th column of the feature of the second image, to obtain the enhanced vectors in the (M1+N1)th column to the (M1+2N1)th column of the second image, and embeds the vectors in the 1st column to the N1th column of the feature of the first image and the vectors in the 1st column to the N2th column of the feature of the second image into the enhanced vectors in the (M1+N2)th column to the (M1+2N2)th column, to obtain the enhanced feature of the second image; wherein, ​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​The vector to the N2th column is embedded into the N2th column. Listed to number The enhancement vector of the column is used to obtain the enhanced features of the second image, including: the control device extracts the enhanced features of the second image from the column's enhanced vector. Listed to number Each vector in the column is copied to the column containing that vector, resulting in the first vector. Listed to number The augmentation vector of the column, the first Listed to number The number of vectors in each column of the augmentation vector is 2*M²; the control device uses the features of the second image from column 1 to column 2. Column and the The N2th column contains a total of K2 vectors. These K2 vectors are then uniformly and randomly embedded into the N2th column. Listed to number The enhanced vectors of the columns are used to obtain the enhanced features of the second image; K2 is an integer greater than 2; where the number of column vectors included in the enhanced features of the second image is... The enhanced features of the second image include a number of row vectors.

[0012] Optionally, the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image. This includes: the control device calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image, for a total of... The correlation coefficient is determined by cosine similarity or Pearson correlation coefficient; the control equipment is determined. The number of relevance values ​​greater than the relevance threshold; if the number of relevance values ​​is greater than or equal to the number threshold, it means that the information relevance is greater than the relevance threshold; otherwise, the information relevance is less than or equal to the relevance threshold.

[0013] Optionally, the control device determines a priori region map based on the current position of the first robot, the current position of the second robot, the first image, and the second image. This includes: the control device processes the current position of the first robot, the current position of the second robot, the enhanced features of the first image, and the enhanced features of the second image through the feature processing layer of the neural network model to obtain a priori region map output by the feature processing layer of the neural network model. The priori region map is a mapping of the first image and the second image in three-dimensional space. The region of the priori region map is a triangular region with the current position of the second robot as the vertex and the boundary of the first image as the boundary. The priori region map contains obstacles.

[0014] Optionally, the control device determines a dynamic route in the prior area map according to the preset route and the prior area map by using a rapid expansion tree algorithm, including: determining, by the control device, whether the preset route overlaps with an obstacle in the prior area map; and if yes, determining, by the control device, the dynamic route in the prior area map by using the rapid expansion tree algorithm.

[0015] In a second aspect, a robot path planning system is provided, including a control device configured to: acquire, by the control device, a first image and a second image, the first image being an image of an environment in a moving direction of a first robot currently located at a position, and the second image being an image of an environment in the moving direction of a second robot currently located at a position, the first robot moving along a preset route, and the second robot moving along the preset route following the first robot; determine, by the control device, a dynamic route for obstacle avoidance in a prior area map according to a current position of the first robot, a current position of the second robot, the first image and the second image, the first robot and the second robot currently being located in the prior area map, and the dynamic route being connected with the preset route; and instruct, by the control device, the first robot and the second robot to move along the dynamic route.

[0016] The system of the second method can refer to the related description of the method of the first aspect, and details are not repeated here.

[0017] In a third aspect, a communication device is provided, including a processor and a memory, the memory being configured to store a computer program, when the processor executes the computer program, so that the communication device executes the method of the first aspect.

[0018] In a possible design, the communication device can further include a transceiver. The transceiver can be a transceiver circuit or an interface circuit. The transceiver can be used for communication between the communication device and other communication devices.

[0019] In the embodiments of the present application, the communication device can be the network device of the first aspect, or a chip (system) or other components or assemblies provided in the network device, or a device including the network device.

[0020] In a fourth aspect, a computer readable storage medium is provided, including a computer program or instructions, when the computer program or instructions are executed on a computer, so that the computer executes the method of the first aspect.

[0021] In summary, based on the above method and system, it can be known that:

[0022] In a case that the first robot moves along the preset moving route and the second robot moves along the preset moving route following the first robot, the control device can acquire a first image in a moving direction collected by the first robot at a current position of the first robot and a second image in the moving direction collected by the second robot at a current position of the second robot. Since the positions of the first robot and the second robot are different, the first image and the second image also have a parallax. Therefore, the control device can convert the first image and the second image into a prior regional map with spatial depth information according to the current position of the first robot, the current position of the second robot, and the parallax of the first image and the second image, and determine a dynamic route for obstacle avoidance in the prior regional map. That is, monocular images collected by the first robot and the second robot can realize dynamic obstacle avoidance path planning with spatial depth, that is, accurate dynamic path planning at low cost. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 An architecture schematic diagram of a system provided by an embodiment of the present application is shown in the following figure.

[0024] Figure 2 A flowchart of a robot path planning method provided by an embodiment of the present application is shown in the following figure.

[0025] Figure 3 An application scenario schematic diagram of a robot path planning method provided by an embodiment of the present application is shown in the following figure.

[0026] Figure 4 A structure schematic diagram of a communication device provided by an embodiment of the present application is shown in the following figure. DETAILED DESCRIPTION

[0027] The technical solutions in the present application will be described below with reference to the accompanying drawings.

[0028] In the embodiments of the present application, the indication can include direct indication and indirect indication, and can also include explicit indication and implicit indication. The information indicated by certain information is referred to as to-be-indicated information. In the implementation process, there are many ways to indicate the to-be-indicated information, for example, but not limited to, the to-be-indicated information itself or an index of the to-be-indicated information can be directly indicated. The to-be-indicated information can also be indirectly indicated by indicating other information, where the other information and the to-be-indicated information have an association relationship. The to-be-indicated information can also be indicated only by a part of the to-be-indicated information, and the other part of the to-be-indicated information is known or agreed in advance. For example, the arrangement order of each information agreed in advance (for example, a protocol) can also be used to realize the indication of a specific information, thereby reducing the indication overhead to a certain extent. Meanwhile, the common part of each information can be identified and uniformly indicated, so as to reduce the indication overhead caused by separately indicating the same information.

[0029] In addition, the specific indication manner can also be various existing indication manners, for example but not limited to the indication manners described above and various combinations thereof. The specific details of the various indication manners can refer to the prior art, and will not be described herein. As known from the above, for example, when multiple information of the same type needs to be indicated, the indication manners of different information can be different. In the specific implementation process, the required indication manner can be selected according to the specific needs, and the present embodiment does not limit the selected indication manner. In this way, the indication manner involved in the present embodiment should be understood as covering various methods that can enable the to-be-indicated party to know the to-be-indicated information.

[0030] It should be understood that the to-be-indicated information can be sent as a whole or divided into multiple sub-information and sent separately, and the sending period and / or sending time of the sub-information can be the same or different. The specific sending method is not limited in the present embodiment. The sending period and / or sending time of the sub-information can be predefined, for example, predefined according to a protocol, or configured by the sending end device by sending configuration information to the receiving end device.

[0031] The "predefined" or "preconfigured" can be realized by pre-storing corresponding codes, tables or other methods that can be used to indicate related information in the device, and the specific implementation manner is not limited in the present embodiment. The "storage" can mean storage in one or more memories. The one or more memories can be separately set or integrated in the encoder or decoder, processor or electronic device. The one or more memories can be partially separately set and partially integrated in the decoder, processor or electronic device. The type of the memory can be any form of storage medium, and the present embodiment does not limit this.

[0032] The "protocol" involved in the present embodiment can refer to a protocol family in the communication field, a standard protocol similar to the protocol family frame structure, or a related protocol in a future sound control method system combined with an environment effect, and the present embodiment does not specifically limit this.

[0033] In the present embodiment, "when", "in the case of", "if" and the like all refer to the device making corresponding processing under certain objective conditions, and are not limited by time, and do not require the device to have a judgment action when implemented, nor does it mean that there are other limitations.

[0034] In the description of the embodiments of the present application, unless otherwise specified, " / " represents that the objects before and after the " / " are in an "or" relationship, for example, A / B can represent A or B; "and / or" in the embodiments of the present application is only a description of the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. And, in the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two. "At least one of the following" or the like means any combination of the items, including any combination of single item or multiple items. For example, at least one of a, b or c can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple. In addition, in order to clearly describe the technical solutions of the embodiments of the present application, in the embodiments of the present application, "first", "second", and the like are used to distinguish the same items or similar items with basically the same function and effect. Those skilled in the art can understand that "first", "second", and the like do not limit the quantity and execution order, and "first", "second", and the like do not necessarily mean different. At the same time, in the embodiments of the present application, "exemplary" or "for example" means to serve as an example, illustration or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, "exemplary" or "for example" is used to present the relevant concept in a specific manner, for understanding.

[0035] The network architecture and service scenarios described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of network architecture and the appearance of new service scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0036] In order to understand the embodiments of the present application, first, the system shown in Figure 1 The system suitable for the embodiments of the present application is described in detail. The exemplary, Figure 1 The architecture of the system provided by the embodiments of the present application is shown in the figure.

[0037] As shown in Figure 1 , the system can include: a control device and a robot.

[0038] The network device can be a radio access network (RAN) device. The RAN device is also referred to as a target RAN device, and the RAN device can be a device that provides access for a terminal. For example, the RAN device can include: a gNB in a 5G, such as a new radio (NR) system, or one or a group (including multiple antenna panels) of antenna panels of a base station in a 5G, or can also be a network node that constitutes a gNB, a transmission and reception point (TRP or transmission point, TP), or a transmission measurement function (TMF), such as a building baseband unit (BBU), or a centralized unit (CU) or a distributed unit (DU), an RSU with base station functions, or a wired access gateway, or a core network element of a 5G. Alternatively, the RAN device can also include an access point (AP) in a wireless fidelity (WiFi) system, a wireless relay node, a wireless backhaul node, various forms of macro base stations, micro base stations (also referred to as small stations), relay stations, access points, wearable devices, vehicle-mounted devices, and the like. Alternatively, the RAN device can also include an access network device of a next-generation mobile communication system, such as a 6G base station, or in a next-generation mobile communication system, the network device can also have other naming ways, which are all within the protection scope of the embodiments of the present application, and the present application does not make any limitation thereto.

[0039] The robots can be multiple, such as a first robot and a second robot, which can be understood as a robot terminal, that is, a terminal, which can also be referred to as user equipment (UE), access terminal, subscriber unit, subscriber station, mobile station, mobile, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment. The terminal device in the embodiments of the present application can be a mobile phone, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, a vehicle-mounted terminal, a RSU with terminal function, etc. The terminal device of the present application can also be a vehicle-mounted module, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit built into a vehicle as one or more components or units. The vehicle can implement the method provided in the present application by built-in vehicle-mounted module, vehicle-mounted module, vehicle-mounted component, vehicle-mounted chip or vehicle-mounted unit. The communication between terminals can be communication between terminals, which can also be referred to as side communication.

[0040] The following will be combined with Figure 2 , the interaction process between each network element / device in the above communication system will be specifically introduced by method embodiments. The robot path planning method provided in the embodiments of the present application can be applied to the above system, which will be specifically introduced below.

[0041] Figure 2 The flowchart of the method provided in the embodiments of the present application is shown. The robot path planning method is applied to the above system, which can be executed by the control device in the system, such as Figure 2 , the specific process is as follows:

[0042] S201, the control device acquires the first image and the second image.

[0043] The first robot moves along a preset travel route, and the second robot moves along the preset travel route following the first robot, that is, the second robot can follow the first robot at a relatively close distance, such as 1-3 meters. The first robot and the second robot can both be transport robots or any other possible type of robot, such as a patrol robot. For ease of understanding, embodiments of the present application are described by way of example of transport robots, that is, in a transport scenario, multiple robots can be used for joint transport, such as the first robot and the second robot each carrying a device / article to be transported, and the second robot maintaining a relatively close distance from the first robot.

[0044] The first image is an environmental image in the travel direction of the first robot at the current location of the first robot, and specifically can be an image captured by a monocular camera of the first robot (the shooting direction of the monocular camera is toward the front of the first robot, that is, the travel direction). Similarly, the second image is an environmental image in the travel direction of the second robot at the current location of the second robot, and specifically can be an image captured by a monocular camera of the second robot.

[0045] It can be understood that the monocular camera has a relatively low cost.

[0046] S202, the control device determines a dynamic route for obstacle avoidance in the prior regional map according to the current position of the first robot, the current position of the second robot, the first image, and the second image.

[0047] The first robot and the second robot are currently located in a region included in the prior regional map. The dynamic route is connected to the preset travel route, that is, it can be understood as a route modified from the preset travel route, for dynamic obstacle avoidance.

[0048] S202 can be implemented by the following steps:

[0049] Step 1: The control device determines the information correlation degree between the first image and the second image.

[0050] The information correlation degree can generally represent the similarity of the contents of the first image and the second image, that is, it can reflect whether the shooting directions of the first robot and the second robot are consistent. If the information correlation degree is very low, that is, the shooting directions of the first robot and the second robot are inconsistent, and the images captured are completely different, in this case, the first image and the second image cannot be used to generate a region of spatial depth. On the contrary, if the information correlation degree is relatively high, that is, the shooting directions of the first robot and the second robot are consistent, and the images captured are substantially the same, in this case, the first image and the second image can be used to generate a region of spatial depth.

[0051] Specifically, the control device extracts the feature of the first image through the neural network model, and extracts the feature of the second image through the neural network model.

[0052] For example, the control device performs target scale convolution on the first image through the neural network model to obtain the feature of the first image, and performs target scale convolution on the second image through the neural network model to obtain the feature of the second image, the first image and the second image have the same resolution, the structure of the feature of the first image is M1*N1, the structure of the feature of the second image is M2*N2, M1=M2, N1=N2, M1 and M2 are the number of rows, and N1 and N2 are the number of columns, M1, M2, N1 and N2 are all integers greater than 2.

[0053] The control device can determine the information correlation degree between the feature of the first image and the feature of the second image.

[0054] For example, the control device determines the information correlation degree between the feature of the first image and the feature of the second image by calculating the correlation degree between each column of vectors in the feature of the first image and the corresponding column of vectors in the feature of the second image.

[0055] In an implementation manner, since the convolution is performed according to the structure of the image, the structure of the feature of the first image also conforms to the structure of the image, that is, the column containing the edge of the matrix of M1*N1 contains the feature of the edge of the first image, and the structure of the feature of the second image also conforms to the structure of the image, that is, the column containing the edge of the matrix of M2*N2 contains the feature of the edge of the second image. Since the second robot is behind the first robot, the contents contained by the edges of the first image and the edges of the second image are usually different in the case of the same shooting direction, so the column containing the edge of the matrix of M1*N1 can be generalized to reduce its influence on feature matching, specifically as follows:

[0056] The control device can enhance the vectors from the (N1-1)th column to the Nth column in the feature of the first image to obtain enhanced vectors from the (N1-1)th column to the Nth column. The control device can enhance the vectors from the (N1-1)th column to the Nth column in the feature of the first image to obtain enhanced vectors from the (N1-1)th column to the Nth column. In an example, N1=20, and the vectors from the 3rd column to the 18th column in the feature of the first image are enhanced. Wherein, represents the upward rounding. In another example, N1=23, and the vectors from the 4th column to the 20th column in the feature of the first image are enhanced. For example, the control device can copy each vector in the (N1-1)th column to the Nth column in the feature of the first image in the column where the vector is located (and then added to the adjacent position of the vector) to obtain enhanced vectors from the (N1-1)th column to the Nth column. ​​​​​​Listed to number The enhancement vectors for each column contain 2*M1 vectors; this is to reduce the influence of edge features (vectors) on the result. The control device can select the features from the first column to the... Column and the The vector to the N1th column is embedded into the N1th column. Listed to number The enhancement vectors of the columns are used to obtain the enhancement features of the first image. For example, the control device can obtain the enhancement features of the first image based on the features of the first image from the first column to the second column. Column and the The N1th column contains a total of K1 vectors. These K1 vectors are then uniformly and randomly embedded into the N1th column. Listed to number The enhanced vectors of the columns are used to obtain the enhanced features of the first image; K1 is an integer greater than 2. In one example, N1 = 20, the columns of the features of the first image are the edge columns, and the vector enhancements of columns 1 to 2 and columns 19 to 20 of the features of the first image, for a total of 4 columns. If M1 = 40, then the 4 columns contain a total of 160 vectors, i.e., K1 = 160. That is, grouping 10 vectors together, the control device can randomly select 10 vectors from the first 4 columns of vector enhancement and embed them into the enhancement vector in the 3rd column. Then, the control device can randomly select 10 vectors from the first 4 columns and embed them into the 4th column, and so on, until the last 10 vectors are randomly embedded into the 18th column. In this case, the enhancement features of the first image include the following number of column vectors: The enhanced features of the first image include a number of row vectors. For example, if N1 = 20, the number of column vectors included in the enhancement features of the first image is 16; if M1 = 40, the number of row vectors included in the enhancement features of the first image is 50.

[0057] Similarly, the control device will select the first feature from the second image. Listed to number Vector augmentation of the column yields the first... Listed to number The augmentation vector of the column, and the features of the first image from column 1 to column 2. Column and the The vector to the N2th column is embedded into the N2th column. Listed to number The enhancement vector of the column is used to obtain the enhanced features of the second image. For example, the control device can extract the enhanced features of the second image from the column enhancement vector. Listed to number Each vector in the column is copied to the column containing that vector, resulting in the first vector. Listed to number the number of vectors in each column of the enhanced vector of the column the number of vectors in each column of the enhanced vector of the column The number of vectors in each column of the enhanced vector of the column The number of vectors in each column of the enhanced vector of the column The number of vectors in each column of the enhanced vector of the column The number of vectors in each column of the enhanced vector of the column The number of vectors in each column of the enhanced vector of the column The number of vectors in each column of the enhanced vector of the column

[0058] Finally, the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column of the enhanced vector of the enhanced features of the first image and the corresponding column of the enhanced vector of the enhanced features of the second image. For example, the control device calculates the correlation between each column of the enhanced vector of the enhanced features of the first image and the corresponding column of the enhanced vector of the enhanced features of the second image, a total of correlations, which can be cosine similarity or Pearson correlation coefficient. The control device can determine the number of correlations greater than the correlation threshold value in the correlations; wherein if the number of correlations is greater than or equal to the number threshold value, it indicates that the information correlation is greater than the correlation threshold value, otherwise, the information correlation is less than or equal to the correlation threshold value. For example, the number of column vectors included in the enhanced features of the first image and the second image is 16, and the number threshold value can be 12.

[0059] Step 2: In the case where the information correlation is greater than the correlation threshold value, the control device determines the prior regional map according to the current position of the first robot, the current position of the second robot, the first image and the second image.

[0060] The control device can process the current position of the first robot, the current position of the second robot, the enhanced features of the first image and the enhanced features of the second image through the feature processing layer of the neural network model to obtain the prior regional map output by the feature processing layer of the neural network model. The prior regional map is a mapping of the first image and the second image in a three-dimensional space, for example, as shown in Figure 3 , the region of the prior regional map is a triangular region with the current position of the second robot as the vertex to the boundary of the first image, and the prior regional map contains obstacles.

[0061] Step 3: The control device determines the dynamic route in the prior regional map according to the preset route and the prior regional map through the rapid expansion tree algorithm.

[0062] The control device can determine whether the preset moving route overlaps with the obstacle in the prior area map;

[0063] If yes, the control device determines the dynamic route in the prior area map by a rapid expansion tree algorithm. The rapid expansion tree algorithm can also be referred to as a rapid expansion random tree (RRT), or the rapid expansion tree algorithm can be replaced by a Dijkstra algorithm or an A* algorithm of heuristic search, without limitation of the prior art.

[0064] In S203, the control device instructs the first robot and the second robot to move along the dynamic route.

[0065] Specifically, the control device can send the dynamic route to the first robot and the second robot respectively, and the first robot and the second robot can set priorities, the priority of the dynamic route being higher than that of the preset moving route. In the case where the dynamic route is different from the preset moving route, the first robot and the second robot move according to the priority of the dynamic route, so as to achieve obstacle avoidance.

[0066] In addition, the above S201-S203 can be periodically executed in the process that the first robot and the second robot move along the preset moving route.

[0067] In summary, in the case where the first robot moves along the preset moving route and the second robot moves along the preset moving route following the first robot, the control device can acquire the first image in the moving direction collected by the first robot at the current position of the first robot, and the second image in the moving direction collected by the second robot at the current position of the second robot. Since the positions of the first robot and the second robot are different, the first image and the second image also have a parallax. Therefore, the control device can convert the first image and the second image into the prior area map with spatial depth information according to the current position of the first robot, the current position of the second robot, and the parallax of the first image and the second image, and determine the dynamic route for obstacle avoidance in the prior area map. That is, the monocular images collected by the first robot and the second robot can realize dynamic obstacle avoidance path planning with spatial depth, that is, accurate dynamic path planning at low cost.

[0068] The above Figure 2 The robot path planning method provided by the embodiments of the present application is described in detail. The following describes a robot path planning system for executing the robot path planning method provided by the embodiments of the present application.

[0069] The system includes a control device configured to:

[0070] The control device acquires a first image and a second image, the first image being an image of an environment in a direction of travel collected by a first robot at a current location of the first robot, and the second image being an image of the environment in the direction of travel collected by a second robot at a current location of the second robot, the first robot moving along a preset travel route, and the second robot moving along the preset travel route following the first robot; the control device determines a dynamic route for obstacle avoidance within a prior regional map according to the current location of the first robot, the current location of the second robot, the first image, and the second image, the first robot and the second robot being currently located within the prior regional map, and the dynamic route being connected to the preset travel route; and the control device instructs the first robot and the second robot to move along the dynamic route.

[0071] Optionally, the control device determines the dynamic route for obstacle avoidance within the prior regional map according to the current location of the first robot, the current location of the second robot, the first image, and the second image, including: the control device determining an information correlation degree between the first image and the second image; in a case where the information correlation degree is greater than a correlation degree threshold, the control device determining the prior regional map according to the current location of the first robot, the current location of the second robot, the first image, and the second image; and the control device determining the dynamic route within the prior regional map by a rapid expansion tree algorithm according to the preset travel route and the prior regional map.

[0072] Optionally, the control device determines the information correlation degree between the first image and the second image, including: the control device extracting a feature of the first image by a neural network model, and extracting a feature of the second image by the neural network model; and the control device determining the information correlation degree between the feature of the first image and the feature of the second image.

[0073] Optionally, the control device extracts the feature of the first image by the neural network model, and extracts the feature of the second image by the neural network model, including: the control device performing target-scale convolution on the first image by the neural network model to obtain the feature of the first image, and performing target-scale convolution on the second image by the neural network model to obtain the feature of the second image, the first image and the second image having the same resolution, the structure of the feature of the first image being M1*N1, the structure of the feature of the second image being M2*N2, M1=M2, N1=N2, M1 and M2 being the number of rows, and N1 and N2 being the number of columns, M1, M2, N1, and N2 each being an integer greater than 2; and the control device determining the information correlation degree between the feature of the first image and the feature of the second image, including: the control device determining the information correlation degree between the feature of the first image and the feature of the second image by calculating a correlation degree between each column of vectors in the feature of the first image and a corresponding column of vectors in the feature of the second image.

[0074] Optionally, the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image, including: the control device calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image. Listed to number Vector augmentation of the column yields the first... Listed to number The augmentation vector of the column, and the features of the first image from column 1 to column 2. Column and the The vector to the N1th column is embedded into the N1th column. Listed to number The enhancement vector of the column is used to obtain the enhanced features of the first image; the control device then selects the enhanced features of the second image from the enhanced vector of the column. Listed to number Vector augmentation of the column yields the first... Listed to number The augmentation vector of the column, and the features of the first image from column 1 to column 2. Column and the The vector to the N2th column is embedded into the N2th column. Listed to number The enhancement vectors of the columns are used to obtain the enhancement features of the second image; where, This indicates rounding up; the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image.

[0075] Optionally, the control device selects the features of the first image as the first feature. Listed to number Vector augmentation of the column yields the first... Listed to number The augmentation vector of the column, and the features of the first image from column 1 to column 2. Column and the The vector to the N1th column is embedded into the N1th column. Listed to number The enhancement vector of the column is used to obtain the enhanced features of the first image, including: the control device extracts the enhanced features of the first image from the column enhancement vector. Listed to number Each vector in the column is copied to the column containing that vector, resulting in the first vector. Listed to number The augmentation vector of the column, the first Listed to number The number of vectors in each column of the augmentation vector is 2*M1; the control device uses the features of the first image from column 1 to column 2 to column 3 to obtain the vectors from the first image. Column and the The N1th column contains a total of K1 vectors. These K1 vectors are then uniformly and randomly embedded into the N1th column. Listed to number The enhanced vectors of the columns are used to obtain the enhanced features of the first image; K1 is an integer greater than 2; where the enhanced features of the first image include the number of column vectors. The enhanced features of the first image include a number of row vectors. Accordingly, the control device will select the first feature from the second image Listed to number Vector augmentation of the column yields the first... Listed to number The augmentation vector of the column, and the features of the first image from column 1 to column 2. Column and the The vector to the N2th column is embedded into the N2th column. Listed to number The enhancement vector of the column is used to obtain the enhanced features of the second image, including: the control device extracts the enhanced features of the second image from the column's enhanced vector. Listed to number Each vector in the column is copied to the column containing that vector, resulting in the first vector. Listed to number The augmentation vector of the column, the first Listed to number The number of vectors in each column of the augmentation vector is 2*M²; the control device uses the features of the second image from column 1 to column 2. Column and the The N2th column contains a total of K2 vectors. These K2 vectors are then uniformly and randomly embedded into the N2th column. Listed to number The enhanced vectors of the columns are used to obtain the enhanced features of the second image; K2 is an integer greater than 2; where the number of column vectors included in the enhanced features of the second image is... The enhanced features of the second image include a number of row vectors.

[0076] Optionally, the control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image. This includes: the control device calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image, for a total of... The correlation coefficient is determined by cosine similarity or Pearson correlation coefficient; the control equipment is determined. The number of the relevancies greater than the relevancy threshold in the relevancies; wherein, if the number of the relevancies is greater than or equal to the number threshold, the information relevancy is greater than the relevancy threshold, otherwise, the information relevancy is less than or equal to the relevancy threshold.

[0077] Optionally, the control device determines the prior regional map according to the current position of the first robot, the current position of the second robot, the first image and the second image, comprising: the control device processes the current position of the first robot, the current position of the second robot, the enhanced features of the first image and the enhanced features of the second image through the feature processing layer of the neural network model to obtain the prior regional map output by the feature processing layer of the neural network model, the prior regional map being a mapping of the first image and the second image in a three-dimensional space, the region of the prior regional map being a triangular region with the current position of the second robot as a vertex to the boundary of the first image, and the prior regional map containing obstacles.

[0078] Optionally, the control device determines the dynamic route in the prior regional map according to the preset route and the prior regional map through the rapid expansion tree algorithm, comprising: the control device determines whether the preset route overlaps with the obstacles in the prior regional map; if yes, the control device determines the dynamic route in the prior regional map through the rapid expansion tree algorithm.

[0079] Figure 4 A structural diagram of a communication device is provided for the embodiments of the present application. Exemplarily, the communication device can be a network device, or a chip (system) or other components or assemblies that can be arranged in the network device. As shown in Figure 4 The communication device 600 can include a processor 601. Optionally, the communication device 600 can also include a memory 602 and / or a transceiver 603. The processor 601 is coupled with the memory 602 and the transceiver 603, for example, through a communication bus.

[0080] The specific introduction of each component of the communication device 600 is as follows: Figure 4 The specific introduction of each component of the communication device 600 is as follows:

[0081] The processor 601 is a control center of the communication apparatus 600, which can be one processor or collectively refer to a plurality of processing elements. For example, the processor 601 is one or more central processing units (CPUs), application specific integrated circuits (ASICs), or one or more integrated circuits configured to perform the embodiments of the present application, such as one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0082] Optionally, the processor 601 can perform various functions of the communication apparatus 600 by running or executing software programs stored in the memory 602 and calling data stored in the memory 602, such as performing the above-mentioned Figure 2 robot path planning method.

[0083] In a specific implementation, as an embodiment, the processor 601 can include one or more CPUs, such as the CPU0 and the CPU1 shown in FIG. 1. Figure 4

[0084] In a specific implementation, as an embodiment, the communication apparatus 600 can also include a plurality of processors. Each of the processors can be a single-CPU or a multi-CPU. The processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (for example, computer program instructions).

[0085] The memory 602 is configured to store software programs for performing the schemes of the present application, and the processor 601 is configured to control the execution. The specific implementation can refer to the above-mentioned method embodiments, and details are not described herein.

[0086] ​Optionally, the memory 602 can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), or other type of dynamic storage device that can store information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory 602 can be integrated with the processor 601 or exist independently and be coupled to the processor 601 through the interface circuit (not shown in the figure) of the communication device 600. The embodiments of the present application are not limited in this regard. Figure 4

[0087] The transceiver 603 is configured to communicate with other communication devices. For example, the communication device 600 is a terminal, and the transceiver 603 can be configured to communicate with a network device or another terminal. For another example, the communication device 600 is a network device, and the transceiver 603 can be configured to communicate with a terminal or another network device.

[0088] Optionally, the transceiver 603 can include a receiver and a transmitter (not shown separately in the figure). The receiver is configured to implement the receiving function, and the transmitter is configured to implement the transmitting function. Figure 4

[0089] Optionally, the transceiver 603 can be integrated with the processor 601 or exist independently and be coupled to the processor 601 through the interface circuit (not shown in the figure) of the communication device 600. The embodiments of the present application are not limited in this regard. Figure 4

[0090] It can be understood that the structure of the communication device 600 shown in the figure does not constitute a limitation on the communication device. The actual communication device can include more or fewer components than those shown in the figure, or combine certain components, or different arrangement of components. Figure 4

[0091] In addition, the technical effects of the communication device 600 can refer to the technical effects of the methods described in the above method embodiments, which will not be described here.

[0092] ​​​​It should be appreciated that a processor in the embodiments of the present application can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can be any conventional processor.

[0093] It should also be understood that the memory in the embodiments of the present application can be a volatile memory or a nonvolatile memory, or can include both volatile and nonvolatile memory. Among them, the nonvolatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically EPROM (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM) and direct rambus RAM (DR RAM).

[0094] The above-described embodiments can be implemented in part or in whole through software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When loaded and executed by a computer, the computer instructions or computer programs can produce the processes or functions described above in accordance with the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, such as from a website site, a computer, a server, or a data center to another website site, a computer, a server, or a data center through a wired (e.g., infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium or a collection of medium accessible by a computer or a data storage device such as a server, a data center, etc. containing one or more available medium. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0095] It should be understood that the term "and / or" in this document is merely used to describe an associated relationship between associated objects, and can represent three relationships, for example, A and / or B can represent three cases of A alone, A and B together, and B alone, where A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the front and rear associated objects, but can also represent an "and / or" relationship. The specific meaning can be understood according to the context before and after.

[0096] In this application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or similar expressions means any combination of the items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0097] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined according to their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0098] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0099] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0100] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0101] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0102] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0103] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0104] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A robot path planning method, characterized in that, Applied to control equipment, the method includes: The control device acquires a first image and a second image. The first image is an environmental image captured by the first robot at its current location along its path, and the second image is an environmental image captured by the second robot at its current location along its path. The first robot moves along a preset path, and the second robot follows the first robot along the preset path. The control device determines a dynamic route for obstacle avoidance within a priori area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image. The first robot and the second robot are currently located within the area contained in the priori area map, and the dynamic route is connected to the preset path route. The control device instructs the first robot and the second robot to move along the dynamic route; The control device determines a dynamic route for obstacle avoidance within a priori area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image, including: The control device determines the information correlation between the first image and the second image; When the information relevance is greater than the relevance threshold, the control device determines the prior area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image; The control device determines the dynamic route within the prior area map based on the preset route and the prior area map using a fast expanding tree algorithm. The control device determines the information correlation between the first image and the second image, including: The control device extracts features from the first image using a neural network model, and also extracts features from the second image using the same neural network model. The control device determines the information correlation between the features of the first image and the features of the second image; The control device determines the prior area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image, including: The control device processes the current position of the first robot, the current position of the second robot, the enhanced features of the first image, and the enhanced features of the second image through the feature processing layer of the neural network model to obtain the prior region map output by the feature processing layer of the neural network model. The prior region map is a mapping of the first image and the second image in three-dimensional space. The region of the prior region map is a triangular region with the current position of the second robot as the vertex and extending to the boundary of the first image. The prior region map contains obstacles.

2. The method according to claim 1, characterized in that, The control device extracts features from the first image using a neural network model, and extracts features from the second image using the same neural network model, including: The control device performs a target-scale convolution on the first image using the neural network model to obtain the features of the first image, and performs the target-scale convolution on the second image using the neural network model to obtain the features of the second image. The first image and the second image have the same resolution. The feature structure of the first image is M1*N1, and the feature structure of the second image is M2*N2, where M1=M2, N1=N2, M1 and M2 are the number of rows, N1 and N2 are the number of columns, and M1, M2, N1 and N2 are all integers greater than 2. The control device determines the information correlation between the features of the first image and the features of the second image, including: The control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image.

3. The method according to claim 2, characterized in that, The control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column vector in the features of the first image and the corresponding column vector in the features of the second image, including: The control device enhances the vectors from column ⌈N1 / 10⌉+1 to column N1-⌈N1 / 10⌉ in the features of the first image to obtain enhanced vectors from column ⌈N1 / 10⌉+1 to column N1-⌈N1 / 10⌉, and embeds the vectors from column 1 to column ⌈N1 / 10⌉ and column N1-⌈N1 / 10⌉+1 to column N1 in the features of the first image into the enhanced vectors from column ⌈N1 / 10⌉+1 to column N1-⌈N1 / 10⌉ to obtain the enhanced features of the first image; The control device enhances the vectors from column ⌈N2 / 10⌉+1 to column N2-⌈N2 / 10⌉ in the features of the second image to obtain enhanced vectors from column ⌈N2 / 10⌉+1 to column N2-⌈N2 / 10⌉, and embeds the vectors from column 1 to column ⌈N2 / 10⌉ and column N2-⌈N2 / 10⌉+1 to column N2 in the features of the second image into the enhanced vectors from column ⌈N2 / 10⌉+1 to column N2-⌈N2 / 10⌉ to obtain the enhanced features of the second image; Where ⌈⌉ represents rounding up; The control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image.

4. The method according to claim 3, characterized in that, The control device enhances the vectors from column ⌈N1 / 10⌉+1 to column N1-⌈N1 / 10⌉ in the features of the first image to obtain enhanced vectors from column ⌈N1 / 10⌉+1 to column N1-⌈N1 / 10⌉, and embeds the vectors from column 1 to column ⌈N1 / 10⌉ and column N1-⌈N1 / 10⌉+1 to column N1 in the features of the first image into the enhanced vectors from column ⌈N1 / 10⌉+1 to column N1, to obtain enhanced features of the first image, including: The control device copies each vector in the column from ⌈N1 / 10⌉+1 to N1-⌈N1 / 10⌉ of the features of the first image to the column where the vector is located, thus obtaining the enhanced vectors in the column from ⌈N1 / 10⌉+1 to N1-⌈N1 / 10⌉. The number of vectors in each column of the enhanced vectors in the column from ⌈N1 / 10⌉+1 to N1-⌈N1 / 10⌉ is 2*M1. The control device, based on the K1 vectors contained in columns 1 to ⌈N1 / 10⌉ and columns N1-⌈N1 / 10⌉+1 to N1 of the features of the first image, uniformly and randomly embeds the K1 vectors into the enhancement vectors in columns ⌈N1 / 10⌉+1 to N1-⌈N1 / 10⌉ to obtain the enhancement features of the first image; K1 is an integer greater than 2. Wherein, the number of column vectors included in the enhancement features of the first image is N1-2*⌈N1 / 10⌉, and the number of row vectors included in the enhancement features of the first image is 2*M1+⌈K1 / (N1-2*⌈N1 / 10⌉)⌉; Accordingly, the control device enhances the vectors from column ⌈N2 / 10⌉+1 to column N2-⌈N2 / 10⌉ in the features of the second image to obtain enhanced vectors from column ⌈N2 / 10⌉+1 to column N2-⌈N2 / 10⌉, and embeds the vectors from column 1 to column ⌈N2 / 10⌉ and column N2-⌈N2 / 10⌉+1 to column N2 in the features of the first image into the enhanced vectors from column ⌈N2 / 10⌉+1 to column N2, to obtain the enhanced features of the second image, including: The control device copies each vector in the features of the second image from column ⌈N2 / 10⌉+1 to column N2-⌈N2 / 10⌉ to the column containing the vector, thus obtaining the enhanced vectors in columns ⌈N2 / 10⌉+1 to N2-⌈N2 / 10⌉. The number of vectors in each column of the enhanced vectors in columns ⌈N2 / 10⌉+1 to N2-⌈N2 / 10⌉ is 2*M2. The control device, based on the K2 vectors contained in columns 1 to ⌈N2 / 10⌉ and columns N2-⌈N2 / 10⌉+1 to N2 of the features of the second image, uniformly and randomly embeds the K2 vectors into the enhancement vectors in columns ⌈N2 / 10⌉+1 to N2-⌈N2 / 10⌉ to obtain the enhancement features of the second image; K2 is an integer greater than 2. The enhancement features of the second image include a column vector number of N² - 2*⌈N² / 10⌉ and a row vector number of 2*M² + ⌈K² / (N² - 2*⌈N² / 10⌉)⌉.

5. The method according to claim 4, characterized in that, The control device determines the information correlation between the features of the first image and the features of the second image by calculating the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image, including: The control device calculates the correlation between each column of enhancement vectors in the enhancement features of the first image and the corresponding column of enhancement vectors in the enhancement features of the second image, for a total of N1-2*⌈N1 / 10⌉ correlations, where the correlation is cosine similarity or Pearson correlation coefficient. The control device determines the number of correlations greater than the correlation threshold among the N1-2*⌈N1 / 10⌉ correlations; Wherein, if the number of relevance values ​​is greater than or equal to the number threshold, it means that the information relevance is greater than the relevance threshold; otherwise, the information relevance is less than or equal to the relevance threshold.

6. The method according to claim 1, characterized in that, The control device determines the dynamic route within the prior area map based on the preset route and the prior area map using a fast expanding tree algorithm, including: The control device determines whether the preset route overlaps with the obstacles in the prior area map; If so, the control device determines the dynamic route within the prior area map using the fast expanding tree algorithm.

7. A robot path planning system, characterized in that, The system includes a control device, which is configured to: The control device acquires a first image and a second image. The first image is an environmental image captured by the first robot at its current location along its path, and the second image is an environmental image captured by the second robot at its current location along its path. The first robot moves along a preset path, and the second robot follows the first robot along the preset path. The control device determines a dynamic route for obstacle avoidance within a priori area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image. The first robot and the second robot are currently located within the priori area map, and the dynamic route is connected to the preset path route. The control device instructs the first robot and the second robot to move along the dynamic route; The control device determines a dynamic route for obstacle avoidance within a priori area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image, including: The control device determines the information correlation between the first image and the second image; When the information relevance is greater than the relevance threshold, the control device determines the prior area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image; The control device determines the dynamic route within the prior area map based on the preset route and the prior area map using a fast expanding tree algorithm. The control device determines the information correlation between the first image and the second image, including: The control device extracts features from the first image using a neural network model, and also extracts features from the second image using the same neural network model. The control device determines the information correlation between the features of the first image and the features of the second image; The control device determines the prior area map based on the current position of the first robot, the current position of the second robot, the first image, and the second image, including: The control device processes the current position of the first robot, the current position of the second robot, the enhanced features of the first image, and the enhanced features of the second image through the feature processing layer of the neural network model to obtain the prior region map output by the feature processing layer of the neural network model. The prior region map is a mapping of the first image and the second image in three-dimensional space. The region of the prior region map is a triangular region with the current position of the second robot as the vertex and extending to the boundary of the first image. The prior region map contains obstacles.

Images