WO2020014990A1

WO2020014990A1 - Multi-map alignment method and device, terminal and storage medium

Info

Publication number: WO2020014990A1
Application number: PCT/CN2018/096539
Authority: WO
Inventors: 王超鹏; 林义闽; 廉士国
Original assignee: 深圳前海达闼云端智能科技有限公司
Priority date: 2018-07-20
Filing date: 2018-07-20
Publication date: 2020-01-23
Also published as: CN109073387B; CN109073387A

Abstract

A multi-map alignment method applied to a terminal or a cloud includes: acquiring N maps including description information of map points of a first path (S101); the description information in the N maps being different, N being an integer greater than one; selecting a map from the N maps as a reference map, the remaining maps being used as correction maps (S102); adjusting the description information of the map points of the first path in the correction maps according to the description information of the map point of the first path in the reference map (S103). The present invention also relates to a multi-map alignment device, a terminal and a storage medium.

Description

Multi-map alignment method, device, terminal and storage medium

Technical field

The present application relates to the field of computer vision, and in particular, to a method, device, terminal, and storage medium for multiple map alignment.

Background technique

Current visual localization and mapping (vSLAM) technology uses visual sensors to collect image data and build maps based on the collected image data.

technical problem

In the process of studying the prior art, the inventor found that the maps established by the vSLAM technology have relativity. For the same path, there is a difference between the image data collected in the forward motion and the image data collected in the reverse motion, resulting in a difference in the output results for the same location during the positioning process, that is, when the map directly established by the collected image data is used for positioning , You will get two different results. In the same way, for two maps with common paths, the above situation also exists in the process of mapping and positioning.

It can be seen that how to reduce the deviation of the localization results of the maps at the same location is a problem that needs to be solved.

Technical solutions

A technical problem to be solved in some embodiments of the present application is how to reduce the deviation of localization results of maps at the same location.

An embodiment of the present application provides a method for multi-map alignment, including: obtaining N maps including description information of map points of a first path; wherein the description information in N maps is different, and N is greater than 1. Integer; select one map from N maps as the reference map, and the remaining maps as the correction map; adjust the description information of the map points of the first path in the revised map according to the description information of the map points of the first path .

An embodiment of the present application further provides a multi-map alignment device, including: an obtaining module, configured to obtain N maps including description information of map points of a first path; wherein the description information in the N maps is different N is an integer greater than 1. The adjustment module is used to select one map from the N maps as the reference map and the remaining maps as the correction map; adjust the correction map according to the description information of the map points of the first path in the reference map. The description information of the map points of the first path.

An embodiment of the present application further provides a terminal, including at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor. The processor executes to enable at least one processor to execute the method for multi-map alignment mentioned in the above embodiment.

An embodiment of the present application further provides a computer-readable storage medium storing a computer program, wherein the computer program implements the method for multi-map alignment mentioned in the above embodiment when executed by a processor.

Beneficial effect

Compared with the prior art, the embodiments of the present application adjust the description information of the map points of the first path in the modified map according to the description information of the map points of the first path in the reference map, so that the reference map and the modified map pair The description information of the map points of the first path are consistent. Because the description information of the map points of the first path in the multiple maps are consistent, the terminal can obtain the same positioning result no matter which map is loaded during the positioning process, and reduce the deviation of the positioning results of the same location in the maps.

BRIEF DESCRIPTION OF THE DRAWINGS

One or more embodiments are exemplified by the pictures in the accompanying drawings. These exemplary descriptions do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the drawings in the drawings do not constitute a limitation on scale.

1 is a flowchart of a multi-map alignment method according to a first embodiment of the present application;

2 is a flowchart of a multi-map alignment method according to a second embodiment of the present application;

3a is a schematic diagram of a first path of a forward map according to a second embodiment of the present application;

3b is a schematic diagram of a first path of a reverse map according to a second embodiment of the present application;

3c is a schematic diagram of a path after A-B is divided into 6 segments in the second embodiment of the present application;

4 is a schematic structural diagram of a multi-map alignment device according to a third embodiment of the present application;

FIG. 5 is a schematic structural diagram of a terminal according to a fourth embodiment of the present application.

Embodiments of the invention

In order to make the purpose, technical solution, and advantages of the present application clearer, some embodiments of the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

The first embodiment of the present application relates to a multi-map alignment method, which is applied to a terminal or a cloud. The terminal may be a blind navigation device, an intelligent robot, and the like. The cloud communicates with the terminal, providing the terminal with a map for positioning or providing positioning results directly to the terminal. In this embodiment, a terminal is used as an example to describe the execution process of the multi-map alignment method. For the process of executing the multi-map alignment method in the cloud, reference may be made to the content of the embodiment of the present application. As shown in FIG. 1, the method for multi-map alignment includes the following steps:

Step 101: Obtain N maps including description information of map points of a first path.

Specifically, the description information in N maps is different, and N is an integer greater than 1. For example, the terminal obtains a forward map of the first path and a reverse map of the first path. The terminal moves from one point to the first direction, collects first image data of the first path, and establishes a forward map of the first path according to the first image data of the first path. Starting from the same point, the terminal moves in the second direction, collects second image data of the first path, and establishes a reverse map of the first path according to the second image data of the first path. Wherein, the directions of the first direction and the second direction are opposite.

Step 102: Select one map from the N maps as the reference map and the remaining maps as the correction map.

For example, the forward map of the first path is selected as the reference map, and the reverse map of the first path is used as the correction map.

Step 103: Adjust the description information of the map points of the first path in the map according to the description information of the map points of the first path in the reference map.

Specifically, the description information of the map points of the first path includes the direction information of the map points of the first path and the position information of the map points. The terminal adjusts the direction information of the map points of the first path in the correction map according to the direction information of the map points of the first path in the reference map, and adjusts the position of the correction map according to the position information of the map points of the first path in the reference map Location information for map points.

Compared with the prior art, the multi-map alignment method provided in this embodiment adjusts and corrects the description information of the map points of the first path in the map according to the description information of the map points of the first path in the reference map, so that The reference map and the modified map have the same description information of the map points of the first path. Because the description information of the map points of the first path in the multiple maps are consistent, the terminal can obtain the same positioning result no matter which map is loaded during the positioning process, and reduce the deviation of the positioning results of the same location in the maps.

The second embodiment of the present application relates to a method for multi-map alignment. This embodiment is a further refinement of the first embodiment, and specifically describes step 103.

As shown in FIG. 2, this embodiment includes steps 201 to 204. Step 201 and step 202 are substantially the same as step 101 and step 102 in the first embodiment, respectively, and will not be described in detail here. The differences are mainly introduced below:

Perform

steps

201 and 202.

Step 203: Divide the first path in each of the N maps into M segments of the second path. M is a positive integer.

In specific implementation, the terminal determines the key point information of the first path of each map in the N maps, and according to the key point information of the first path of each map in the N maps, each of the N maps The first path in the map is divided into M segments of the second path. The key points refer to special position points in the first path, such as inflection points. The terminal can divide the first path into different road segments according to these special position points.

The method for the terminal to determine the key point information of the first path of each of the N maps is as follows: each of the N maps performs the following operations: obtaining the direction information of the map points collected during the map establishment process ; Determining the key point information of the first path in the map according to the direction information of the map points.

The following describes the method for the terminal to obtain the direction information of the map points collected during the map establishment process.

Method 1: The terminal uses vSLAM technology to create a map based on the image data of the first path collected by the visual sensor set on the terminal, and outputs the position information and direction information corresponding to each frame of image, and uses the position information corresponding to each frame of the image as a map For the position information of the points, the direction information corresponding to each frame of the image is used as the direction information of the map points.

Method 2: The terminal determines the direction information of the map point according to data collected by other sensors such as an odometer or an Inertial Measurement Unit (IMU). Taking the odometer as an example, the terminal carries a visual sensor and an odometer, and moves along the first path. During the movement of the terminal, the vision sensor collects the image data of the first path, determines the time stamp information corresponding to each frame of image, and the odometer collects the direction information and position information of the terminal. The terminal aligns the time stamp of the visual sensor with the time stamp of the odometer, and determines the direction information and location information of the map point according to the time stamp information corresponding to each frame of the image. The terminal determines the description information of the map points according to the image information of each frame, the direction information and the position information corresponding to each frame of the image.

It is worth mentioning that adding other sensors to assist in determining the location information and direction information of map points can avoid using vSLAM technology to map unlooped paths. Because the terminal does not detect a loop, the terminal's pose will not be optimized. The resulting cumulative error.

The following describes a method for determining key point information of a first path in a map according to direction information of a map point.

For example, the key point is an inflection point in the first path. In a straight path, the direction information of two adjacent map points is similar. At the turn, the direction information of two adjacent map points will change greatly. The terminal arranges the direction information of the map points of the first path in the order of obtaining the direction information of the map points of the first path. For the direction information of the map points of each first path, the following operations are respectively performed: calculating the direction information of the map points and The first difference value of the direction information of the next map point, to determine whether the first difference value is greater than the first threshold, and if determined to be, use the position information of the map point corresponding to the direction information of the map point as a key point candidate for the first path information. The terminal determines key point information of the first path according to all candidate key point information of the first path. The first threshold may be set according to actual needs.

In specific implementation, the terminal may set a certain moving distance dt, and separately record the direction information of the map point corresponding to each moving dt of the terminal. Set a certain direction threshold, that is, the first threshold. If the difference between the direction information of two adjacent map points is greater than the direction threshold, it is considered that there is a key point in the path between the two map points, and the starting point of the path The corresponding map point and / or the map point corresponding to the end point of the path are used as key points. In the actual running process, at a turning position, a plurality of points with large continuous movement directions can be detected. Therefore, the position information of the key points can be determined according to the position information of the points whose continuous movement direction changes greatly, as shown in formula 1:

Formula 1:

Among them, t _j represents the position information of key points, k _i represents the position information of map points with large changes in the direction of movement, and n represents the number of map points with large changes in the direction of movement.

Step 204: For each segment of the second path in each revised map, the following operations are respectively performed: determining the second path of the reference map corresponding to the second path of the revised map; and according to the second path of the corresponding reference map The description information of the map points is adjusted to modify the description information of the map points of the second path of the map.

The following describes a method for the terminal to determine the second path of the reference map corresponding to the second path of the modified map.

For example, the terminal obtains a forward map and a reverse map of the first path, and uses the forward map of the first path as a reference map and the reverse map of the first path as a correction map. The key points in the forward map of the first path are {t _P | (t _P0 , t _P1 , t _P2 ...... t _Pn-2 , t _Pn-1 )}, and the key points in the reverse map of the first path Is {t _N | (t _N0 , t _N1 , t _N2 ...... t _Nn-2 , t _Nn-1 )}. Since the starting point of the forward map and the reverse map of the first path are the same and only the directions are opposite, the order of the key points of the forward map of the first path and the order of the key points of the reverse map of the first path are reversed, that is, {t _P -t _N | t _P0 -t _N0 , t _P1 -t _N1 , t _P2 -t _N2 , ... t _Pn-2 -t _Nn-2 , t _Pn-1 -t _Nn-1 }. Since the second path is divided according to the key points in the forward map of the first path and the key points in the reverse map of the first path, the key points in the forward map of the first path and the first path are divided. The correspondence between the key points in the reverse map can determine the correspondence between the second path in the forward map and the second path in the reverse map.

The following is a method for adjusting the description information of the map points of the second path of the revised map based on the second path of the reference map corresponding to the second path of the revised map (hereinafter referred to as the corresponding path of the second path of the revised map). Instructions.

The terminal determines the direction information of the corresponding path of the second path of the modified map and the direction information of the second path of the modified map, according to the direction information of the corresponding path of the modified second path of the map, and the direction information of the second path of the modified map. , Adjust the direction information of the map points of the second path of the correction map. The terminal adjusts the position information of the map points of the second path of the modified map according to the position information of the map points of the second path of the reference map and the direction information of the adjusted map points of the second path of the modified map.

The terminal determines the direction information of the second path of each map in the N maps according to the direction information of the map points collected during the establishment of each map in the N maps. Specifically, the terminal performs the following operations respectively for the second path of each map in the N maps: obtaining all direction information of the map points of the second path, and for the direction information of the map points of each second path, respectively The following operations are performed: determining the second difference between the direction information of the map points of the second path and the direction information of the map points of the other second paths; and determining all the second differences corresponding to the direction information of the map points of the second path, Whether there is a second difference greater than the second threshold, and if it is determined not to exist, the direction information of the map points of the second path is used as the direction information of the second path.

The following describes the process in which the terminal updates the description information of the corrected map according to the description information of the reference map in different situations.

Case 1: The number of direction information of the corresponding path of the second path of the modified map is one, and the number of direction information of the second path of the corrected map is one. In this case, the terminal adjusts the direction information of the map points of the second path of the modified map according to the direction information of the corresponding path of the second path of the modified map and the direction information of the second path of the modified map; In the path, the position information of the map point corresponding to the map point at the start of the second path of the revised map is substituted for the position information of the map point at the start of the second path of the revised map; from the second path of the revised map after the replacement The position information of the starting map point starts, and is calculated according to the scale of the modified map, the position information of the current map point, the adjusted direction information of the current map point, and the distance information between the current map point and the next map point. Replacement information used to replace the position information of the next map point, and replace the position information of the next map point with the replacement information until the position information of all map points on the second path of the modified map is replaced.

Case 2: The number of direction information of the corresponding path of the second path of the modified map is greater than one, or the number of direction information of the second path of the corrected map is greater than one. In this case, the terminal performs the following operations on the direction information of the corresponding path of the second path of each modified map: sequentially selects one direction information from the direction information of the second path of the modified map, and performs the following operations: The direction information of the corresponding path of the second path of the map and the direction information of the second path of the selected modified map are used to adjust the direction information of the map points of the second path of the modified map; The position information of the map point corresponding to the map point starting from the second path of the map replaces the position information of the map point starting from the second path of the revised map; the map point starting from the second path of the revised map after the replacement Starting from the position information of the map, the calculation is used to replace the next one according to the scale of the modified map, the position information of the current map point, the adjusted direction information of the current map point, and the distance information of the current map point and the next map point. Replacement information of the position information of the map point, and replace the position information of the next map point with the replacement information until the replacement Map location information of all points on the path to complete the second amendment of the map. The terminal determines the position information of the map points corresponding to the map points at which the second path of the modified map ends in the second path of the reference map. The terminal calculates a third difference between the position information of the map point corresponding to the map point where the second path of the modified map ends and the position information of the map point where the second path of the modified map ends after the replacement. After the terminal completes the foregoing operations for all the direction information of the corresponding path of the second path of the modified map and the direction information of the second path of the corrected map, the terminal determines the smallest third difference value, and performs the operation corresponding to the smallest third difference value. As a result, the adjustment result of the description information of the map points of the first path in the final revised map.

For example, the first path of the forward map is shown in FIG. 3a, and the first path of the reverse map is shown in FIG. 3b. The forward map is used as the reference map and the reverse map is used as the modified map. The terminal determines the key points of the first path of the forward map as A, B, C, D, E, and F. According to the above key points, the first path of the forward map is divided into six second paths, which are: AB, BC, CD, DE, EF, and FA. The terminal determines the key points of the first path of the reverse map as a, b, c, d, e, and f. According to the above key points, the first path of the reverse map is divided into 6 second paths, which are ab and bc, respectively. , Cd, de, ef, and fa. Among them, AB corresponds to ab, BC corresponds to bc, CD corresponds to cd, DE corresponds to de, EF corresponds to ef, and FA corresponds to fa. Because the direction of a single map point is not sufficient to represent the overall direction of a straight line, after the map is created using vSLAM technology, many points can be obtained on a path, and the direction of directly using one point will cause a large difference. Therefore, the terminal divides the second path into multiple third paths. For example, a schematic diagram of the path after dividing AB into 6 segments is shown in FIG. 3c. Determine the difference between the direction information of map point h and the direction information of other map points (i, j, k, l, m, and n). If the difference between the direction information of map point h and other map points is not greater than the second The threshold value uses the direction information of the map point h as the direction information of the second path. By analogy, the direction information of other map points of AB is judged. Finally, if the direction information of each map point meets the requirements, the direction information of AB may be the direction information of map point h, the direction information of map point i, the direction information of map point j, the direction information of map point k, and the map. Any one of the direction information of the point l, the direction information of the map point m, and the direction information of the map point n. With reference to the above method, the direction information of the other second paths of the forward map and the direction information of the second path of the reverse map are determined. Because there may be multiple possible direction information for each second path of the reference map and each second path of the reverse map, the adjustment results corresponding to each type of direction information are different. Therefore, corrections need to be made according to each possible direction information. The map is adjusted. The following uses AB in the forward map and ab in the reverse map as examples to illustrate the adjustment process. Assume that the direction information of only one map point u in the ab in the reverse map meets the requirements. Let the direction information of map point u be the direction information of ab. The direction information of map point h and map point j in AB meet the requirements. Using the direction information of the map point h as the direction information of AB, the difference θ _d between the direction information of the map point h and the direction information of the map point u is calculated. According to the difference θ _d , the direction information of each map point in ab is adjusted according to Formula 2.

Formula 2: θ _cor = θ _org + θ _d

Among them, θ _cor indicates the direction information after the map points are adjusted in the second path of the modified map, θ _org indicates the direction information before the map points are adjusted in the second path of the modified map, and θ _d indicates the second path of the modified map. The difference between the direction information and the direction information of the second path of the reference map.

Replace the position information of the start point a of ab with the position information of the start point A of AB, and calculate the replacement information for replacing the position information of the next map point according to formulas 3 and 4, until all the information on the second path of the modified map is replaced. Location information for map points.

Formula 3: x '= x + s * d * cos (θ _cor )

Formula 4: y '= y + s * d * sin (θ _cor )

Among them, (x, y) represents the position information of the current map point in the modified map, s represents the scale of the modified map; d represents the distance between the current map point and the next map point, and θ _cor represents the adjusted current map point. Direction information, (x ', y') represents replacement information for replacing the direction information of the next map point.

After replacing the position information of each map point in ab, the difference q ₁ between the position information of the end point (map point b) of ab and the position information of the end point (map point B) of ab is calculated according to Formula 5.

Equation 5:

Among them, q represents the difference between the adjusted position information of the map point at the end of the second path of the modified map and the position information of the map point at the end of the second path of the reference map, and (x _pe , y _pe ) represents each segment in the reference map. The position information of the end point of the path, (x _ne , y _ne ) represents the adjusted position information of the end point of each segment of the path in the correction map.

Then, use the direction information of the map point j as the direction information of the AB of the forward map, and refer to the direction information of the map point h as the direction information of the AB of the forward map, and adjust the position information of the map points of the ab of the reverse map. A difference q ₂ between the position information of the end point (map point b) of ab of the reverse map and the end point (map point B) of AB of the forward map corresponding to the adjustment result is determined. The terminal judges the sizes of q ₁ and q ₂ and determines that q _{1 is} less than q ₂ , then uses the direction information of the map point h as the direction information of AB of the forward map, and the direction information and position information of the map points of ab of the reverse map. The adjustment result is used as the final adjustment result.

The third embodiment of the present application relates to an apparatus for multi-map alignment. As shown in FIG. 4, the apparatus includes an obtaining module 301 and an adjusting module 302. The obtaining module 301 is configured to obtain N maps including description information of map points of the first path; the description information in the N maps is different, and N is an integer greater than 1. The adjustment module 302 is configured to retrieve N maps from N maps. One of the maps is selected as the reference map, and the rest of the maps are used as the correction map. According to the description information of the map points of the first path in the reference map, the description information of the map points of the first path in the correction map is adjusted.

It is not difficult to find that this embodiment is a system embodiment corresponding to the first embodiment, and this embodiment can be implemented in cooperation with the first embodiment. The related technical details mentioned in the first embodiment are still valid in this embodiment. To reduce repetition, details are not described here. Accordingly, the related technical details mentioned in this embodiment can also be applied in the first embodiment.

A fourth embodiment of the present application relates to a terminal. As shown in FIG. 5, the terminal includes at least one processor 401; and a memory 402 that is communicatively connected to the at least one processor 401. The memory 402 stores instructions that can be executed by the at least one processor 401, and the instructions are executed by the at least one processor 401, so that the at least one processor 401 can execute the above-mentioned multi-map alignment method.

In this embodiment, the processor 401 uses a central processing unit (CPU) as an example, and the memory 402 uses a readable and writable memory (Random Access Memory, RAM) as an example. The processor 401 and the memory 402 may be connected through a bus or other methods. In FIG. 5, the connection through the bus is taken as an example. The memory 402 is a non-volatile computer-readable storage medium and can be used to store non-volatile software programs, non-volatile computer executable programs, and modules. For example, the reference map and the modified map are stored in the embodiment of the present application. In memory 402. The processor 401 executes various functional applications and data processing of the device by running non-volatile software programs, instructions, and modules stored in the memory 402, that is, a method for implementing the multi-map alignment described above.

The memory 402 may include a storage program area and a storage data area, where the storage program area may store an operating system and an application program required for at least one function; the storage data area may store a list of options and the like. In addition, the memory 402 may include a high-speed random access memory, and may further include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 402 may optionally include a memory remotely set relative to the processor, and these remote memories may be connected to an external device through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

One or more modules are stored in the memory, and when executed by one or more processors, the multi-map alignment method in any of the foregoing method embodiments is executed.

The above products can execute the method provided in the embodiment of the present application, and have the corresponding functional modules and beneficial effects of the execution method. For technical details not described in this embodiment, refer to the method provided in the embodiment of the present application.

A fifth embodiment of the present application relates to a computer-readable storage medium storing a computer program. When the computer program is executed by the processor, the method for multi-map alignment described in any of the above method embodiments is implemented.

That is, those skilled in the art can understand that all or part of the steps in the method of the above embodiments can be implemented by a program instructing related hardware. The program is stored in a storage medium and includes several instructions for making a device ( It may be a single-chip microcomputer, a chip, or the like) or a processor to perform all or part of the steps of the method described in each embodiment of the present application. The foregoing storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks and other media that can store program codes .

Those of ordinary skill in the art can understand that the foregoing embodiments are specific embodiments for implementing the present application, and in practical applications, various changes can be made in form and details without departing from the spirit and range.

Claims

A multi-map alignment method, including:

Obtaining N maps including description information of map points of the first path; wherein the description information in the N maps is different, and N is an integer greater than 1;

Selecting one map from the N maps as a reference map and the remaining maps as correction maps;

And adjusting the description information of the map points of the first path in the modified map according to the description information of the map points of the first path in the reference map.
The method for multi-map alignment according to claim 1, wherein the description information of the map points of the first path in the modified map is adjusted according to the description information of the map points of the first path in the reference map , Including:

Dividing the first path of each of the N maps into M segments of the second path, where M is a positive integer;

For each segment of the second path in the revised map, the following operations are respectively performed:

Determining a second path of the reference map corresponding to the second path of the modified map;

And adjusting the description information of the map points of the second path of the modified map according to the description information of the map points of the second path of the reference map.
The method for multi-map alignment according to claim 2, wherein the description information includes position information of a map point and direction information of the map point;

Adjusting the description information of the map points of the second path of the revised map according to the description information of the map points of the second path of the corresponding reference map specifically includes:

Determining direction information of a corresponding second path of the reference map, and direction information of a second path of the modified map;

Adjusting the direction information of the map points of the second path of the modified map according to the corresponding direction information of the second path of the reference map and the direction information of the second path of the modified map; according to the The position information of the map points of the second path of the reference map and the direction information of the adjusted map points of the second path of the modified map adjust the position information of the map points of the second path of the modified map.
The method for multi-map alignment according to claim 2 or 3, wherein the dividing the first path of each of the N maps into M segments of the second path specifically includes:

Determining key point information of the first path of each of the N maps;

According to the key point information of the first path of each of the N maps, the first path of each of the N maps is divided into M segments of the second path.
The method for multi-map alignment according to claim 4, wherein the determining key point information of the first path in each of the N maps specifically comprises:

The following operations are performed on each of the N maps: obtaining the direction information of the map points collected during the map establishment process; and determining the first of the maps according to the direction information of the map points. Key information for the path.
The method for multi-map alignment according to claim 5, wherein determining the key point information of the first path in the map according to the direction information of the map points comprises:

Arrange the direction information of the map points of the first path according to the order of obtaining the direction information of the map points of the first path;

For the direction information of the map point of each of the first paths, the following operations are respectively performed: calculating a first difference between the direction information of the map point and the direction information of the next map point, and determining whether the first difference is Is greater than the first threshold, and if it is determined to be, the position information of the map point corresponding to the direction information of the map point is used as the candidate key point information of the first path;

Determining key point information of the first path according to all candidate key point information of the first path.
The method for multi-map alignment according to claim 3, wherein the determining direction information of the second path of the corresponding reference map and the direction information of the second path of the modified map specifically include :

Determine the direction information of the second path of each map in the N maps according to the direction information of the map points collected during the establishment of each map in the N maps.
The method for multi-map alignment according to claim 7, wherein the determining the first position of each map in the N maps according to the direction information of the map points collected during the establishment of each map in the N maps The direction information of the two paths includes:

For the second path of each of the N maps, perform the following operations:

Acquiring all direction information of the map points of the second path;

For the direction information of the map points of each of the second paths, the following operations are respectively performed: determining a second difference between the direction information of the map points of the second path and the direction information of the map points of other second paths; Whether there is a second difference greater than a second threshold value among all the second differences corresponding to the direction information of the map points of the second path, and if it is determined not to exist, the map point direction information of the second path is used as the first Direction information of two paths.
The method for multi-map alignment according to claim 8, wherein, according to the direction information of the corresponding second path of the reference map and the direction information of the second path of the modified map, adjusting Adjusting the direction information of the map points of the second path of the modified map; adjusting the position information of the map points of the second path of the reference map and the direction information of the map points after the second path of the modified map is adjusted The position information of the map points on the second path of the modified map specifically includes:

If it is determined that the number of direction information of the second path of the corresponding reference map is 1, and the number of direction information of the second path of the modified map is 1, according to the corresponding Use the direction information of the second path of the reference map and the direction information of the second path of the modified map to adjust the direction information of the map points of the second path of the modified map; using the second path of the reference map, Position information of a map point corresponding to a map point starting from a second path of the revised map, replacing position information of a map point starting from the second path of the revised map; The position information of the map point at the start of the second path of the map starts, according to the scale of the modified map, the position information of the current map point, the adjusted direction information of the current map point, and the current map point and the next. Distance information of map points, calculating replacement information for replacing position information of the next map point, and replacing the position information of the next map point with the replacement information until the replacement is completed Map the location information of all points on the map a second correction path;

If it is determined that the number of direction information of the corresponding second path of the reference map is greater than 1, or that the number of direction information of the second path of the modified map is greater than 1, for each of the corresponding For the direction information of the second path of the reference map, the following operations are respectively performed: from the direction information of the second path of the modified map, one direction information is sequentially selected, and the following operations are performed: according to the corresponding one of the Direction information of the second path of the reference map and direction information of the selected second path of the modified map, adjusting direction information of map points of the second path of the modified map; using the second path of the reference map The position information of the map point corresponding to the map point starting from the second path of the modified map replaces the position information of the map point starting from the second path of the modified map; The position information of the map point at the start of the second path of the modified map is described, according to the scale of the modified map, the position information of the current map point, and the adjustment of the current map point. Subsequent direction information, and distance information between the current map point and the next map point, calculating replacement information for replacing the position information of the next map point, and replacing the position information of the next map point with the replacement information, Until the position information of all the map points on the second path of the modified map is replaced; determining the positions of the map points in the second path of the reference map corresponding to the map points where the second path of the modified map ends Information; calculating a third position information of the map point corresponding to the map point where the second path of the modified map ends and a third position information of the map point where the second path of the modified map ends Difference

After the above operations are completed for all the corresponding direction information of the second path of the reference map and the direction information of the second path of the modified map, the smallest third difference is determined, and the The operation result corresponding to the smallest third difference is used as the final adjustment result of the description information of the map points of the first path in the modified map.
The method for multi-map alignment according to any one of claims 5 to 9, wherein the direction information of the map points is determined according to data collected by an odometry or inertial measurement device IMU.
A multi-map alignment device, including:

An obtaining module, configured to obtain N maps including description information of map points of a first path; wherein the description information in the N maps is different, and N is an integer greater than 1;

An adjustment module, configured to select one map from the N maps as a reference map and the remaining maps as modified maps; and adjust the Description information of the map points of the first path.
A terminal comprising at least one processor; and

A memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processing The device can execute the method for multi-map alignment according to any one of claims 1 to 10.
A computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the method for multi-map alignment according to any one of claims 1 to 10 is implemented.