Route correction method and autonomous mobile device

Abstract

A route correction method and an autonomous mobile device are disclosed. Reference position information corresponding to a reference point in a captured image is determined, wherein the captured image is taken from the reference point in an environment. Deviation information between estimated position information and the reference position information is determined, wherein the estimated position information is estimated based on motion information of the autonomous mobile device. A movement route information of the autonomous mobile device is corrected according to the deviation information, wherein the movement route information corresponds to a route between the estimated position information and the reference position information. Thus, the accuracy and reliability of positioning can be improved.

Description

Technical Field

[0001] This invention relates to a route planning technology, and more particularly to a route correction method and an autonomous mobility device. Background Technology

[0002] Current drones may encounter the following problems:

[0003] Human intervention: When the drone cannot accurately locate itself, its position needs to be corrected by human intervention. This will reduce mission efficiency and increase costs.

[0004] Installing sensors: The sensors on drones can be used for environmental perception. However, data processing from these sensors remains challenging in complex environments.

[0005] Path planning: Most current UAV systems use advanced path planning algorithms, but these are not suitable for closed environments lacking satellite positioning. Summary of the Invention

[0006] This invention provides a route correction method and an autonomous mobile device that can achieve accurate positioning in enclosed areas.

[0007] The route correction method of this invention is applicable to autonomous mobile devices. The route correction method includes (but is not limited to) the following steps: determining reference position information corresponding to a reference point in a captured image, wherein the captured image is obtained by photographing a reference point in the environment; determining deviation information between estimated position information and reference position information, wherein the estimated position information is estimated based on the motion information of the autonomous mobile device; and correcting the movement route information of the autonomous mobile device based on the deviation information, wherein the movement route information corresponds to the route between the estimated position information and the fixed-point position information.

[0008] The autonomous mobile device of this invention includes (but is not limited to) a memory and a processor. The memory stores program code. The processor is coupled to the memory. The processor loads the program code and executes: determining reference position information corresponding to a reference point in a captured image, wherein the captured image is obtained by photographing a reference point in the environment; determining deviation information between estimated position information and reference position information, wherein the estimated position information is estimated based on the motion information of the autonomous mobile device; and correcting the movement route information of the autonomous mobile device based on the deviation information, wherein the movement route information corresponds to the route between the estimated position information and the fixed-point position information.

[0009] Based on the above, the route correction method and autonomous mobile device of this invention correct the movement route information using the deviation information between reference position information determined based on imagery and estimated position information determined based on motion information. This improves positioning accuracy and, consequently, the efficiency of the patrol mission.

[0010] To make the above features and advantages of the present invention more apparent and understandable, specific embodiments are described below in conjunction with the accompanying drawings. Attached Figure Description

[0011] Figure 1 This is a block diagram of the components of an autonomous mobile device according to an embodiment of the present invention;

[0012] Figure 2 This is a flowchart of a route correction method according to an embodiment of the present invention;

[0013] Figure 3 This is a schematic diagram of deviation information according to an embodiment of the present invention;

[0014] Figure 4 This is a schematic diagram of movement route information according to an embodiment of the present invention;

[0015] Figure 5 This is a schematic diagram of movement route information according to an embodiment of the present invention;

[0016] Figures 6A to 6C This is a schematic diagram of a scaling factor according to an embodiment of the present invention;

[0017] Figure 7 This is a schematic diagram of movement route information according to an embodiment of the present invention;

[0018] Figure 8 This is a schematic diagram of movement route information according to an embodiment of the present invention.

[0019] Symbol Explanation

[0020] 100: Autonomous mobile device

[0021] 105: Image capturing device

[0022] 110: Sports Organization

[0023] 120: Memory

[0024] 130: Processor

[0025] S210~S230: Steps

[0026] 301: Estimated point

[0027] 302: Reference point

[0028] P1~P5: Location points

[0029] R0~R5: Actual points

[0030] P'1~P'5: New positioning points

[0031] P'3~P'5: Newly corrected positioning points Detailed Implementation

[0032] Figure 1 This is a block diagram of an autonomous mobile device 100 according to an embodiment of the present invention. Please refer to... Figure 1 The autonomous mobile device 100 includes (but is not limited to) an image capturing device 105, a motion mechanism 110, a memory 120, and a processor 130. The autonomous mobile device 100 can be an unmanned aerial vehicle (UAV), an unmanned aircraft, an autonomous aircraft, an autonomous mobile robot (AMR), an automated guided vehicle (AGV), an unmanned / computer-controlled vehicle, a robotic vacuum cleaner, or other mobile devices.

[0033] The image capturing device 105 may be a camera, video camera, or other device, module, or element equipped with an image capturing device. The image capturing device 105 may include an image sensor (e.g., a charge-coupled device (CCD), complementary metal-oxide-semiconductor (CMOS), etc.), an optical lens, image control circuitry, and other components. In this embodiment of the invention, the image capturing device 105 is used to capture images of the outside world. For example, the image capturing device 105 captures images of the environment in which the autonomous mobile vehicle 100 is located to obtain captured images. Capturing images means that the image capturing device 105 captures images of the environment. This environment may be, for example, a farm, a factory, or the ocean, and is not limited thereto.

[0034] The motion mechanism 110 may include a power unit (e.g., a motor or engine), a transmission system (e.g., a drive shaft or a gearbox), and a drive unit (e.g., wheels, tracks, propellers). In some embodiments, the motion mechanism 110 functions to change position, move, fly, or sail under the drive of a power unit (e.g., a motor or engine) that controls the movement of the drive unit.

[0035] The memory 120 can be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid-state drive (SSD), or similar component. In one embodiment, the memory 120 is used to store program code, software modules, configuration settings, data, or files (e.g., location information, deviation information, or movement route information), which will be detailed in subsequent embodiments.

[0036] Processor 130 is coupled to image capturing device 105, moving mechanism 110, and memory 120. Processor 130 may be a central processing unit (CPU), graphics processing unit (GPU), or other programmable general-purpose or special-purpose microprocessor, digital signal processor (DSP), programmable controller, field-programmable gate array (FPGA), application-specific integrated circuit (ASIC), neural network accelerator, or other similar components or combinations thereof. In one embodiment, processor 130 is used to execute all or part of the operations of autonomous moving device 100, and can load and execute program code, software modules, files, and data stored in memory 120.

[0037] The methods described in the embodiments of the present invention will be explained below in conjunction with the various mechanisms, devices, components and modules in the autonomous mobile device 100. The various processes of this method may be adjusted according to the implementation situation, and are not limited thereto.

[0038] Figure 2 This is a flowchart of a route correction method according to an embodiment of the present invention. Please refer to... Figure 2 The processor 130 determines the reference position information corresponding to the reference point in the captured image (step S210). Specifically, the captured image is obtained by the image capturing device 105 from a reference point in the environment. The reference position information corresponding to the reference point is pre-configured / determined. The reference position information can be coordinates of any coordinate system (e.g., latitude, longitude, and altitude or other custom two / three-dimensional coordinate systems) or the relative position (e.g., relative distance and / or direction) with respect to a reference object, and is used to indicate the position of the reference point.

[0039] In one embodiment, processor 130 can convert the captured image into vector code. This vector code is one-dimensional. That is, processor 130 converts a two-dimensional captured image into a one-dimensional vector code.

[0040] For example, processor 130 performs a convolution operation on the captured image: this captured image serves as the input image for the convolution operation and can be considered as a three-dimensional tensor of size H×W×C. H is the total number of pixels in the captured image corresponding to its height, W is the total number of pixels in the captured image corresponding to its width, and C is the number of channels (C is 3 if it is an RGB image). The mathematical expression for the convolution operation is: Y = X*W + b, where X is the input image (e.g., the captured image from image capturing device 105), W is the convolution kernel (or weights), * is the convolution operation, b is the bias term, and Y is the output feature map. The size of the output feature map is, for example, H'×W'×D, where H' is the total number of pixels in the output feature map corresponding to its height, W' is the total number of pixels in the output feature map corresponding to its width, and D is the number of channels in the output feature map.

[0041] Processor 130 flattens the 3D output feature map Y into a 1D vector, mathematically represented as: z = Flatten(Y), where z is a 1D vector of size 1 × (H' × W' × D). Next, processor 130 performs a linear projection on the flattened 1D vector to obtain the final embedding vector or feature representation (i.e., vector encoding): mathematically represented as: z' = W f ·z+b f , where is the projection matrix and its size is n×(H'×W'×D) (n is a positive integer), and b f This is the bias term. Ultimately, the output vector encoding z' is a one-dimensional vector of size n, which is the embedding vector in the feature space.

[0042] In some embodiments, the trained Vision Integrates Transformer (ViT) Mamba module can perform the aforementioned function of converting 3D images into 1D vectors. That is, the processor 130 inputs the captured image into the trained ViT Mamba module, which then generates the corresponding vector encoding. The ViT Mamba module is characterized by its lightweight design compared to other machine learning models.

[0043] Processor 130 determines the reference position information corresponding to the vector code. Processor 130 may pre-store the correspondence between one or more reference vector codes and their corresponding reference position information, and find the reference position information corresponding to the reference point in the captured image from this correspondence. For example, processor 130 finds a reference vector code that is the same as the vector code of the captured image, and obtains the reference position information of this reference vector code in the correspondence.

[0044] In another embodiment, the vector encoding is in the form of text, symbols, or numbers, and the processor 130 can directly derive the reference position information from the vector encoding in the form of text, symbols, or numbers. For example, the vector encoding is binary latitude and longitude coordinates or three-dimensional coordinates.

[0045] Please refer to Figure 2 The processor 130 determines the deviation information between the estimated position information and the reference position information (step S220). Specifically, the estimated position information is estimated based on the motion information of the autonomous mobile device 100. The estimated position information can be coordinates in any coordinate system (e.g., latitude, longitude, and altitude, or other custom two / three-dimensional coordinate systems) or its relative position to a reference object (e.g., relative distance and / or direction), and is used to indicate the estimated position. Motion information includes, for example, the distance / speed and direction of movement.

[0046] In one embodiment, the processor 130 can determine the estimated position information corresponding to the motion information of the autonomous mobile device 100 based on the initial position information, and the initial position information is known. For example, if the initial position information is coordinates (0,0,0) and the motion information is moving 300 meters per minute, then the estimated position information after one minute is coordinates (300,0,0). In one application scenario, the autonomous mobile device 100 is an unmanned aerial vehicle (UAV), and the processor 130 can determine the current estimated position information through an automatic dead reckoning algorithm.

[0047] In one embodiment, the estimated position information includes first estimated coordinates for the current time point. For example, coordinates determined by an automated dead reckoning algorithm, consisting of latitude, longitude, and altitude. The reference position information includes first reference coordinates for the current time point. For example, coordinates converted from captured imagery, consisting of latitude, longitude, and altitude. The deviation information includes the deviation value between the first estimated coordinates and the first reference coordinates. The deviation value is, for example, the difference between the longitude of the estimated position information and the longitude of the reference position information, the difference between the latitude of the estimated position information and the latitude of the reference position information, and the difference between the altitude of the estimated position information and the altitude of the reference position information. However, the coordinates are not measured in latitude, longitude, or altitude. For example, coordinates in a custom two-dimensional or three-dimensional coordinate system.

[0048] For example, Figure 3This is a schematic diagram of deviation information according to an embodiment of the present invention. Please refer to... Figure 3 The estimated position information is the estimated coordinates (X1, Y1, Z1) of estimated point 301, and the reference position information is the reference coordinates (X2, Y2, Z2) of reference point 302. The deviation value is the difference between the estimated coordinates of estimated point 301 and the reference coordinates of reference point 302 on the three axes (e.g., X-axis, Y-axis, and Z-axis). For example, [X2-X1, Y2-Y1, Z2-Z1]. The arrow / vector from estimated point 301 to reference point 302 shown in the figure represents this deviation value.

[0049] Please refer to Figure 2 The processor 130 corrects the movement route information of the autonomous mobile device 100 based on the deviation information (step S230). Specifically, the (original) movement route information corresponds to the route between the estimated position information and the fixed-point position information. The fixed-point position information corresponds to one or more positioning points in the overall route of the autonomous mobile device 10. The positioning position information includes the position information of one or more positioning points. The positioning position information can be coordinates of any coordinate system (e.g., latitude, longitude, and altitude or other custom two / three-dimensional coordinate systems) or relative positions (e.g., relative distance and / or direction) with reference objects, and is used to indicate the position of the positioning point. The processor 130 is pre-configured to move between multiple positioning points. The processor 130 pre-determines the order of these positioning points and determines the route between two positioning points and their corresponding reference route information based on this order. The reference route information is, for example, a combination of one or more distances and corresponding directions from the first positioning point to the second positioning point. The overall route of the autonomous mobile device 10 is the set of routes passing through these positioning points.

[0050] Before correction, processor 130 assumes that the estimated point corresponding to the estimated position information should be on the route corresponding to the reference route information between two positioning points. Therefore, the (original) movement route information corresponds to the route between the estimated point corresponding to the estimated position information and the next positioning point corresponding to the fixed position information. For example, it is a combination of one or more distances and corresponding directions for moving from the estimated point at the current time to the next positioning point.

[0051] In one embodiment, the processor 130 can correct the positioning point corresponding to the fixed position information based on the deviation value to become a new positioning point. For example, the processor 130 adds the deviation value to the coordinates of the positioning point. This new positioning point has a deviation value from the original positioning point.

[0052] For example, Figure 4 This is a schematic diagram of movement route information according to an embodiment of the present invention. Please refer to... Figure 4 The actual points R0 and R1 have coordinates (x, y, y) respectively. R0 ,y R0 ,z R0 ), (xR1 ,y R1 ,z R1 The position (e.g., reference coordinates) of the autonomous mobile device 10 at two points in time is determined based on reference position information. Therefore, the matrix of the actual route R is R = [x R0 y R0 z R0 ],[x R1 y R1 z R1 For example, R = [70 37 47], [120 60 90]. Locating points P1, P2, P3, P4, and P5 (whose coordinates are (x...)... P1 ,y P1 ,z P1 ), (x P2 ,y P2 ,z P2 ), (x P3 ,y P3 ,z P3 ), (x P4 ,y P4 ,z P4 ), (x P5 ,y P5 ,z P5 This corresponds to the fixed-point location information of the autonomous mobile device 10, determined based on the estimated location information. The matrix of the total route P corresponding to the fixed-point location information is P = [x P1 y P1 z P1 ]、[x P2 y P2 z P2 ]、[x P3 y P3 z P3 ]、[x P4 y P4 z P4 ]、[x P5 y P5 z P5 For example, P = [50 4860], [95 70 100], [128 107 125], [144 150 123], [132 199 100].

[0053] Assuming processor 130 considers the current time point to be located at positioning point P1, processor 130 will set the coordinates (x, y) of positioning point P1. P1 ,y P1 ,z P1 The first estimated coordinate is (x, y). The actual coordinates of point R1 at the current time are (x, y). R1 ,y R1 ,zR1 The first reference coordinate is used. Therefore, the deviation value is Δ = R1 - P1 = [Δ]. x Δ y Δ z For example, Δ = [20 -11 -13]. The new positioning points P'1, P'2, P'3, P'4, and P'5 have coordinates (x...). P’1 ,y P’1 ,z P’1 ), (x P’2 ,y P’2 ,z P’2 ), (x P’3 ,y P’3 ,z P’3 ), (x P’4 ,y P’4 ,z P’4 ), (x P’5 ,y P’5 ,z P’5 The position corresponds to the correction based on the deviation information. The matrix of the total route P' corresponding to the corrected movement route information is P' = P + Δ = [x P1 +Δ x y P1 +Δ y z P1 +Δ z ]、[x P2 +Δ x y P2 +Δ y z P2 +Δ z ]、[x P3 +Δ x y P3 +Δ y z P3 +Δ z ]、[x P4 +Δ x y P4 +Δ y z P4 +Δ z ]、[x P5 +Δ x y P5 +Δ y z P5 +Δ z For example, P' = [7037 47], [115 5987], [148 96 112], [164 139 110], [152 188 87]. In this case, the new positioning point P'1 overlaps with the reference point R1. That is, the coordinates of the new positioning point P'1 are the same as the coordinates of the reference point R1.

[0054] In one embodiment, the processor 130 can determine whether the deviation information meets a correction condition. The correction condition is that the deviation value corresponding to the deviation information is greater than a deviation threshold value. For example, the deviation value of any of the three axes is greater than 10 meters (i.e., the deviation threshold value). In response to the deviation information meeting the correction condition, the processor 130 can correct the movement route information of the autonomous mobile device 10. For example, if the deviation value corresponding to the deviation information is greater than the deviation threshold value, the processor 130 corrects the positioning point to a new positioning point. In response to the deviation information not meeting the correction condition, the processor 130 can prohibit / not correct the movement route information of the autonomous mobile device 10. For example, if the deviation value corresponding to the deviation information is not greater than the deviation threshold value, the positioning point remains unchanged.

[0055] In one embodiment, the corrected movement route information includes multiple new location coordinates for multiple new location points. For example, the above... Figure 4 The embodiments provide the coordinates of new positioning points P'1, P'2, P'3, P'4, and P'5. The reference position information includes multiple reference coordinates corresponding to multiple reference points, respectively. For example, the above... Figure 4 The coordinates of actual points R0 and R1 in the embodiment.

[0056] Processor 130 can determine a loss function based on the distance between two of these new positioning points and the distance between two of multiple reference points. The distance between two of these new positioning points is the difference between the corresponding two new positioning coordinates. For example, as described above... Figure 4 The new positioning coordinates (x, y) of the new positioning points P'1 and P'2 in the embodiment P’1 ,y P’1 ,z P’1 ), (x P’2 ,y P’2 ,z P’2 The difference (x) P’2 -x P’1 ,y P’2 -y P’1 ,z P’2 -z P’1 The distance between any two of these reference points is the difference between their corresponding coordinates. For example, the above... Figure 4 The actual points R0 and R1 in the embodiment have two reference coordinates (x) R0 ,y R0 ,z R0 ), (x R1 ,y R1 ,z R1 The difference (x) R1 -x R0 ,y R1 -y R0 ,z R1 -z R0 ).

[0057] In some application scenarios, the corrected route information may not correctly guide the overall route P'. In this case, it is also necessary to consider the scaling factor (hereinafter referred to as the scaling factor) corresponding to the distance between the two new positioning points and the distance between the two actual points. The difference between the two new positioning coordinates in the loss function corresponds to the scaling factor. That is to say, the loss function is the error between the distance between the two actual points and the scaled distance between the two new positioning points.

[0058] In one embodiment, the loss function is the difference between a first value and a second value, where the first value is the distance between two of a plurality of new localization points, and the second value is the product of the distance between two of a plurality of reference points and a scaling factor. For example, the matrix of the scaling factor S is S = [S x S y S z The coordinates of two points A and B are (A, B, C, D, E, F, F, G, F, F, x A y A z (B) x B y B z The vector d corresponding to the distance between )) is d A-B =[A x -B x A y -B y A z -B z The loss function is... Where vector Corresponding to the (i+1)th actual point R (i+1) With the i-th actual point R i The distance between them (i.e., the difference between the two reference coordinates), Corresponding to the (i+1)th new positioning point P′ (i+1) With the i-th new positioning point P′ i The distance between them (i.e., the difference between the two new positioning coordinates), and n is a positive integer.

[0059] For example, Figure 5 This is a schematic diagram of movement route information according to an embodiment of the present invention. Please refer to... Figure 5 The new coordinates of the new location point P'1 are (70, 37, 47), and the new coordinates of the new location point P'2 are (95, 70, 100). Furthermore, the new coordinates of the actual point R1 are (70, 37, 47), and the new coordinates of the actual point R2 are (120, 60, 90). These coordinates can be substituted into the aforementioned loss function, Loss(S).

[0060] Next, the processor 130 minimizes the loss function and determines the scaling factor. That is, it minimizes the error between the distance between the two actual points and the scaled distance between the two new localized points, and finds the scaling factor that minimizes the error in the loss function. At this point, the scaling factor is used to correct the two new localized coordinates.

[0061] For example, Figures 6A to 6C This is a schematic diagram of a scaling factor according to an embodiment of the present invention. Please refer to... Figures 6A to 6C These figures represent the correspondence between the loss function's loss (e.g., the range of the loss function Loss(S)) on the X, Y, and Z axes and the values ​​of the scaling factors. In these figures, the matrix of the scaling factor S is S = [1.111 1.045 1.075]. That is, when the values ​​of the scaling factor S on the X, Y, and Z axes are 1.111, 1.045, and 1.075, respectively, the loss function has the minimum loss / error on the X, Y, and Z axes, respectively.

[0062] In one embodiment, processor 130 can determine the minimum value of the loss function using gradient descent. The minimum value (i.e., the minimum loss / error mentioned above) corresponds to a scaling factor. Gradient descent is a method that continuously updates parameters (e.g., scaling factors) to find a solution (i.e., the value of the loss function). Therefore, processor 130 can first randomly generate a set of initial parameter solutions, and then use this randomly generated set of solutions to calculate the gradient direction and magnitude of the solution. Then, by continuously updating the parameters to change the gradient direction and magnitude, the loss function gradually approaches the minimum value. (See calculation...) Figures 6A to 6C The minimum loss in the loss function is the valley / lowest point, and the gradient descent method is an algorithm that gradually finds the valley / lowest point from a certain point on the loss function.

[0063] In other embodiments, the processor 130 may also use momentum gradient descent, adaptive momentum estimation (Adam), Newton's method, or other optimization algorithms to determine a scaling factor that yields the minimum of the loss function.

[0064] Next, the processor 130 can use the product of the scaling factor and a new positioning coordinate as the corrected new positioning coordinate. For example, the corrected new positioning coordinate P” = S.P’ = [S x .x P’1 S y .y P’1 S z .z P’1 ],[S x .x P’2 S y .yP’2 S z .z P’2 ],…,[S x .x P’n S y .y P’n S z .z P’n At this point, the corrected movement route information corresponds to the route between the estimated location information and the corrected new fixed-point location information (including one or more corrected new positioning points and their corrected new positioning coordinates).

[0065] For example, Figure 7 This is a schematic diagram of movement route information according to an embodiment of the present invention. Please refer to... Figure 7 The corrected reference route information P” is the set of the corrected new positioning points P”3, P”4, P”5 = [164.444100.36 120.4], [182.22145.36 118.25], [169.33 196.55 93.73].

[0066] Figure 8 This is a schematic diagram of movement route information according to an embodiment of the present invention. Please refer to... Figure 8 The autonomous mobile device 100 moves based on the reference route information formed by the corrected new positioning coordinates (i.e., the corrected new positioning points P”3, P”4, P”5), and is located at the actual points R3, R4, R5 respectively. Compared with the new positioning points P’3, P’4, P’5, the actual points R3, R4, R5 are closer to the corrected new positioning points P”3, P”4, P”5.

[0067] In one embodiment, the processor 130 can control the motion mechanism 110 to move along the modified movement path information. For example, the processor 130 can generate control commands based on the determined movement path information, causing the motion mechanism 110 to move forward, backward, rotate / turn, accelerate, decelerate, and / or stop according to the control commands.

[0068] In summary, the route correction method and autonomous mobile device of this invention perform positioning and route correction based on the deviation information between reference position information based on captured images and estimated position information based on motion information. This provides accurate and reliable positioning in environments where satellite positioning is impossible or difficult to achieve, and avoids the problem of further route deviation caused by continued driving, thereby improving mission efficiency.

[0069] Although the present invention has been disclosed above by way of embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be defined by the appended claims.

Claims

1. A route correction method, applicable to autonomous mobile devices, the route correction method comprising: It is determined to extract the reference position information corresponding to the reference point in the image, wherein the image is captured by taking a picture of the reference point in the environment; The deviation information between the estimated position information and the reference position information is determined, wherein the estimated position information is estimated based on the motion information of the autonomous mobile device; as well as The movement route information of the autonomous mobile device is corrected based on the deviation information, wherein the movement route information corresponds to the route between the estimated location information and the fixed location information.

2. The route correction method as described in claim 1, wherein the estimated location information includes a first estimated coordinate at the current time point, the reference location information includes a first reference coordinate at the current time point, the deviation information includes a deviation value between the first estimated coordinate and the first reference coordinate, and the step of correcting the movement route information of the autonomous mobile device based on the deviation information includes: Based on the deviation value, the positioning point corresponding to the fixed position information is corrected to become a new positioning point, wherein the new positioning point and the original positioning point have the deviation value.

3. The route correction method as described in claim 1, wherein the corrected movement route information includes multiple new positioning coordinates of multiple new positioning points, the reference position information includes multiple reference coordinates of multiple reference points respectively corresponding to the new positioning points, and the step of correcting the movement route of the autonomous mobile device based on the deviation information includes: The loss function is determined based on the distance between two of the new positioning points and the distance between two of the reference points, where the distance between two of the new positioning points is the difference between the corresponding two new positioning coordinates, the distance between two of the reference points is the difference between the corresponding two reference coordinates, and the difference between the two new positioning coordinates in the loss function corresponds to the scaling factor. as well as Minimize the loss function and determine the scaling factor, which is used to correct the two new positioning coordinates.

4. The route correction method as described in claim 3, wherein the loss function is the difference between a first value and a second value, the first value being the distance between two of the new positioning points, and the second value being the product of the distance between two of the reference points and the scaling factor.

5. The route correction method as described in claim 3, wherein the step of minimizing the loss function includes: The minimum value of the loss function is determined by gradient descent, where the minimum value corresponds to the scaling factor.

6. The route correction method as described in claim 3, further comprising: The product of the scaling factor and the new positioning coordinates is used as the corrected new positioning coordinates.

7. The route correction method as described in claim 1, further comprising: The estimated position information corresponding to the motion information of the autonomous mobile device is determined based on the initial position information, wherein the initial position information is known.

8. The route correction method as described in claim 1, further comprising: Determine whether the deviation information meets the correction condition, wherein the correction condition is that the deviation value corresponding to the deviation information is greater than the deviation threshold value; If the deviation information meets the correction condition, the movement route information of the autonomous mobile device is corrected. as well as If the deviation information meets the correction conditions, the correction of the autonomous mobile device's movement route information is prohibited.

9. The route correction method as described in claim 1, wherein the step of determining the reference position information corresponding to the reference point in the captured image includes: The captured image is converted into a vector code, where the vector code is one-dimensional; as well as This determines the reference position information corresponding to the vector encoding.

10. The route correction method as described in claim 1, further comprising: The autonomous mobile device is controlled to move along the modified mobile route information, wherein the fixed position information corresponds to at least a certain point in the total route of the autonomous mobile device, and the total route is a set of routes passing through the at least certain point.

11. An autonomous mobile device, comprising: Memory, which stores program code; as well as The processor, coupled to the memory, loads the program code and executes it. It is determined to extract the reference position information corresponding to the reference point in the image, wherein the image is captured by taking a picture of the reference point in the environment; The deviation information between the estimated position information and the reference position information is determined, wherein the estimated position information is estimated based on the motion information of the autonomous mobile device; as well as The movement route information of the autonomous mobile device is corrected based on the deviation information, wherein the movement route information corresponds to the route between the estimated location information and the fixed location information.

12. The autonomous mobile device of claim 11, wherein the estimated position information includes a first estimated coordinate at the current time point, the reference position information includes a first reference coordinate at the current time point, the deviation information includes a deviation value between the first estimated coordinate and the first reference coordinate, and the processor further executes: Based on the deviation value, the positioning point corresponding to the fixed position information is corrected to become a new positioning point, wherein the new positioning point and the original positioning point have the deviation value.

13. The autonomous mobile device of claim 11, wherein the corrected mobile route information includes multiple new positioning coordinates of multiple new positioning points, the reference position information includes multiple reference coordinates of multiple reference points respectively corresponding to the new positioning points, and the processor further performs: The loss function is determined based on the distance between two of the new positioning points and the distance between two of the reference points, where the distance between two of the new positioning points is the difference between the corresponding two new positioning coordinates, and the distance between two of the reference points is the difference between the corresponding two reference coordinates. The difference between the two new positioning coordinates in the loss function corresponds to a scaling factor. Minimize the loss function and determine the scaling factor, which is used to correct the two new positioning coordinates.

14. The autonomous mobile device of claim 13, wherein the loss function is the difference between a first value and a second value, the first value being the distance between two of the new positioning points, and the second value being the product of the distance between two of the reference points and the scaling factor.

15. The autonomous mobile device of claim 13, wherein the processor further performs: The minimum value of the loss function is determined by gradient descent, where the minimum value corresponds to the scaling factor.

16. The autonomous mobile device of claim 13, wherein the processor further performs: The product of the scaling factor and the new positioning coordinates is used as the corrected new positioning coordinates.

17. The autonomous mobile device of claim 11, wherein the processor further performs: The estimated position information corresponding to the motion information of the autonomous mobile device is determined based on the initial position information, wherein the initial position information is known.

18. The autonomous mobile device of claim 11, wherein the processor further performs: Determine whether the deviation information meets the correction condition, wherein the correction condition is that the deviation value corresponding to the deviation information is greater than the deviation threshold value; If the deviation information meets the correction condition, the movement route information of the autonomous mobile device is corrected. as well as If the deviation information meets the correction conditions, the correction of the autonomous mobile device's movement route information is prohibited.

19. The autonomous mobile device of claim 11, wherein the processor further performs: The captured image is converted into a vector code, wherein the vector code is one-dimensional; and This determines the reference position information corresponding to the vector encoding.

20. The autonomous mobile device of claim 11, further comprising: A motion mechanism, coupled to the processor, and the processor also performs: The motion mechanism is controlled to move along the modified movement path information, wherein the fixed position information corresponds to at least a certain point in the total route of the autonomous moving device, and the total route is a set of routes passing through the at least certain point.

Images