Child car automatic following method and child car

Abstract

The application relates to the technical field of children's cars, and provides a children's car automatic following method and a children's car. The following method comprises the following steps: acquiring a following distance set by a user; acquiring first positioning information corresponding to a UWB component and second positioning information corresponding to a camera; taking the first positioning information as prior estimation of a Kalman filter, taking the second positioning information as observation value of the Kalman filter, and calculating target positioning information at a current time; and moving according to the target positioning information to keep the following distance from a remote controller. The application improves the reliability of children's car following.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of children's cars, in particular to a children's car automatic following method and a children's car. BACKGROUND

[0002] UWB (Ultra Wide Band, ultra-wideband) technology is a new type of wireless communication technology. Unlike traditional communication, UWB technology does not need a carrier to carry data transmission, but transmits extremely narrow pulses to transmit data. Therefore, the UWB signal has an extremely narrow time domain width, and can achieve high-precision positioning capability of centimeters.

[0003] Children's car automatic following refers to the movement of the wheels of the children's car driven by the motor of the children's car to follow the movement of adults, without the need for manual operation of the children's car by adults. The automatic following of the children's car requires more reliable and more stable following of adults. SUMMARY

[0004] The present application provides a children's car automatic following method and a children's car, aiming to improve the reliability of the children's car.

[0005] In a first aspect, the present application provides a children's car automatic following method, applicable to a children's car, the children's car comprising a UWB component, a camera, and a remote controller; wherein the remote controller is provided with a UWB tag, and the following method comprises:

[0006] obtaining a following distance set by a user;

[0007] calculating first positioning information of the remote controller according to raw positioning signals continuously collected by the UWB component, wherein the first positioning information is three-dimensional positioning information;

[0008] tracking a target object according to image signals continuously collected by the camera, and calculating second positioning information of the target object according to an area ratio and position information of the target object in the image signals; wherein the second positioning information is three-dimensional positioning information;

[0009] calculating target positioning information at the current time by taking the first positioning information as prior estimation of a Kalman filter and taking the second positioning information as observation value of the Kalman filter; and

[0010] moving according to the target positioning information to maintain the following distance to follow the remote controller.

[0011] In some embodiments, the calculation of the first positioning information of the remote controller according to the raw positioning signals continuously collected by the UWB component comprises:

[0012] The original positioning signals from the Nth moment to the N+Lth moment are acquired through at least 3 UWB base stations, wherein the N+Lth moment is the current moment;

[0013] The original positioning signals from the Nth moment to the N+Lth moment are smoothed to obtain target smoothed positioning signals from the Nth moment to the Nth moment, and the target smoothed positioning signal of the Nth moment is taken as the first positioning information of the Nth moment.

[0014] In some embodiments, the smoothing of the original positioning signals from the Nth moment to the N+Lth moment comprises:

[0015] (1) The Euclidean distance of the original positioning signal of the Mth moment and the original positioning signal of the M+1th moment is acquired;

[0016] (2) When the Euclidean distance is greater than a preset distance threshold, the interpolation weight of the original positioning signal of the Mth moment and the original positioning signal of the M+1th moment is confirmed according to the Euclidean distance, and the interpolation weight is in a positive correlation with the Euclidean distance;

[0017] (3) The original positioning signal of the Mth moment and the original positioning signal of the M+1th moment are interpolated according to the interpolation weight, and the interpolation result is taken as the intermediate smoothed positioning signal of the updated Mth moment, wherein the interpolation formula is slerp (original positioning signal of the Mth moment, original positioning signal of the M+1th moment, interpolation weight);

[0018] (4) For the case that N is less than or equal to L, the original positioning signals from the Nth moment to the N+Lth moment are used to iteratively execute steps (1), (2) and (3) for at least 3 times, to obtain the intermediate smoothed positioning signals from the Nth moment to the N+Lth moment, take the intermediate smoothed positioning signal of the Nth moment as the target smoothed positioning signal, and store the intermediate smoothed positioning signals from the N+1th moment to the N+Lth moment;

[0019] For the case that N is greater than L, when the original positioning signal of the N+L+1th moment is acquired, the stored intermediate smoothed positioning signals from the N+1th moment to the N+Lth moment are used to replace the original positioning signals from the N+1th moment to the N+Lth moment, and steps (1), (2) and (3) are iteratively executed according to the stored intermediate smoothed positioning signals from the N+1th moment to the N+Lth moment and the original positioning signal of the N+L+1th moment.

[0020] In some embodiments, the value of the interpolation weight is greater than or equal to 0.2 and less than or equal to 0.8; the positive correlation between the interpolation weight and the Euclidean distance is a nonlinear relationship, and the nonlinear relationship is that the greater the Euclidean distance, the smaller the ratio of the interpolation weight to the Euclidean distance.

[0021] In some embodiments, the following method further comprises:

[0022] According to the user set following strength and following distance, the interpolation adjustment coefficient is calculated, the interpolation adjustment coefficient is in positive correlation with the following strength, and the interpolation adjustment coefficient is in negative correlation with the tracking distance;

[0023] The interpolation of the original positioning signal at the Mth moment and the original positioning signal at the M+1th moment according to the interpolation weight comprises:

[0024] The interpolation adjustment coefficient is multiplied by the interpolation weight to obtain an updated interpolation weight;

[0025] The Mth moment positioning signal and the M+1th moment positioning signal are interpolated using the updated interpolation weight.

[0026] In some embodiments, the first positioning information is taken as the prior estimate of the Kalman filter, the second positioning information is taken as the observation value of the Kalman filter, and the target positioning information at the current moment is calculated, comprising:

[0027] The smoothed positioning signal at the Nth moment is taken as the prior estimate of the Kalman filter, and the second positioning information at the Nth moment is taken as the observation value of the Kalman filter, and the target positioning information at the Nth moment is calculated;

[0028] The target positioning information is used to follow the remote controller, comprising:

[0029] At the N+Lth moment, the target positioning information at the Nth moment is used to control the movement of the stroller to follow the remote controller.

[0030] In some embodiments, the time difference between the Nth moment and the N+Lth moment is less than or equal to 1.5 seconds.

[0031] In some embodiments, the target object is tracked according to the image signal, and the second positioning information of the target object and the stroller is calculated according to the area ratio and position of the target object in the image signal, comprising:

[0032] The camera is controlled to continuously collect image signals with the same shooting parameters;

[0033] The target object is tracked according to the continuous image signal;

[0034] The second positioning information of the target object and the stroller is calculated according to the area ratio and position of the target object in the image signal.

[0035] In a second aspect, the application provides a child stroller, comprising: a stroller body; a camera disposed on the stroller body; a UWB component disposed on the stroller body; a communication component disposed on the stroller body, configured to establish a communication connection with a remote controller; and a controller disposed on the stroller body, connected with the camera, the UWB component and the communication component respectively, and configured to execute a following program, which, when executed, implements the automatic following method of the child stroller.

[0036] The application obtains first positioning information by using the UWB component, second positioning information by using the area proportion and position information of the target object in the image signal collected by the single camera, and more reliable target positioning information by using a Kalman filter to fuse the first positioning information and the second positioning information, and then follows the remote controller based on the target positioning information, thereby improving the following safety and reliability of the child stroller that can automatically follow. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 a flowchart of an embodiment of the automatic following method of the child stroller of the application;

[0038] Figure 2 a flowchart of another embodiment of the automatic following method of the child stroller of the application;

[0039] Figure 3 a flowchart of still another embodiment of the automatic following method of the child stroller of the application;

[0040] Figure 4 a schematic diagram of the original positioning signal and the target smoothed positioning signal of an embodiment of the automatic following method of the child stroller of the application. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the application will be described clearly and completely in the description of the embodiments of the application in combination with the drawings in the embodiments of the application. Obviously, the described embodiments are only some of the embodiments of the application, but not all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the application.

[0042] In the description of the application, the terms “first” and “second” are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with “first” and “second” can explicitly or implicitly include one or more of the features. In the description of the application, the meaning of “multiple” is two or more, unless otherwise specifically limited.

[0043] In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration." Any embodiment described as "for example" in the present application is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that the present application can be practiced without using these specific details. In other instances, well-known structures and processes are not elaborated upon in order to avoid unnecessary detail, which can obscure the description of the present application. Thus, the present application is not intended to be limited by the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0044] Embodiment 1

[0045] The present application provides a child car automatic following method, which is suitable for a child car, and the child car comprises a UWB component, a camera, and a remote controller, wherein the remote controller is provided with a UWB tag.

[0046] Reference Figure 1 In an embodiment, the following method comprises:

[0047] S100, acquiring a following distance set by a user;

[0048] The remote controller can be provided with a touch display or a button, and the user can set the distance of the child car following the remote controller by operating the touch display or the button. In actual application, selectable following distance options, such as 0.2 meters, 0.5 meters, 0.7 meters, 1.2 meters, 1.5 meters, etc., can be displayed on the display screen, and the user selects the corresponding following distance by selecting operation.

[0049] S200, calculating first positioning information of the remote controller according to the original positioning signals continuously collected by the UWB component, wherein the first positioning information is three-dimensional positioning information; the UWB component comprises a plurality of UWB base stations.

[0050] The child car is provided with a plurality of UWB base stations, and the UWB base stations can be four, wherein three UWB base stations are located on a first plane, and one UWB base station is located on a second plane, so that the four UWB base stations form a polyhedral layout, and better three-dimensional positioning of the positioning UWB tag is achieved.

[0051] The UWB tag is arranged on the remote controller and is powered by a lithium battery on the remote controller.

[0052] In this embodiment, the first positioning information is three-dimensional positioning information.

[0053] S300, track the target object according to the image signals continuously collected by the camera, and calculate second positioning information of the target object according to area proportion and position information of the target object in the image signals; wherein the second positioning information is three-dimensional positioning information.

[0054] The target object refers to an adult, that is, the object collected by the camera.

[0055] In this embodiment, a single image signal is used. Since the image collected by the single camera does not have depth information, only two-dimensional positioning can be realized, and the distance from the target object cannot be calculated. Therefore, the application uses a fixed zoom parameter to collect image signals, and then calculates the distance from the target object to the camera according to the area proportion of the target object in the image signals. Specifically, the greater the area proportion, the closer the target object is to the camera, and the smaller the area proportion, the farther the target object is from the camera. Based on the position of the target object in the image, two-dimensional positioning is confirmed, and only the relative position between the target object and a preset point (such as the center of the image or the corner of the image) in the image needs to be calculated to confirm the two-dimensional positioning.

[0056] The application uses the area proportion to realize three-dimensional positioning information of a single camera, matches the three-dimensional positioning of UWB, and fuses the first positioning information and the second positioning information with the same dimension.

[0057] S400, take the first positioning information as a priori estimation of the Kalman filter, take the second positioning information as an observation value of the Kalman filter, and calculate target positioning information at the current time.

[0058] The first positioning information and the second positioning information can solve the problem of inaccurate first positioning information caused by shielding of the UWB tag signal, and the problem of inaccurate second positioning information of the target object caused by the field of view of the single camera being blocked. The Kalman filter is used as a fusion algorithm, the first positioning information is used to update the system state of the Kalman filter, and then the second positioning information is used as an observation value in the same coordinate system as the first positioning information to calibrate the first positioning information, thereby avoiding serious deviation of the first positioning information.

[0059] S500, move according to the target positioning information to keep the following distance and follow the remote controller.

[0060] The target positioning information includes the relative position of the remote controller and the child car, so as long as the child car is controlled to move towards the remote controller, based on keeping the area proportion of the target object in the image signals and the constant distance collected by the UWB tag, the speed of the child car is dynamically adjusted, so as to realize following the remote controller at a fixed distance.

[0061] The present application obtains first positioning information by using a UWB component, obtains second positioning information of the area proportion and position information of the target object of the image signal collected by a single camera, then fuses the first positioning information and the second positioning information by using a Kalman filter to obtain more reliable target positioning information, and then follows the remote controller based on the target positioning information, thereby improving the following safety and reliability of the child car that can automatically follow.

[0062] With reference to Figure 2 In some embodiments, the adult may suddenly stop or suddenly accelerate during the process of guiding the child car. Taking the sudden stop as an example, at this time, the child car has been moving at a certain speed, and the sudden stop may cause the child car to collide with the adult or the child car to overturn.

[0063] To solve the above problems, step S200 of the present application comprises the following steps:

[0064] S201, obtaining the original positioning signal from the Nth moment to the N+Lth moment through at least three UWB base stations; wherein the N+Lth moment is the current moment.

[0065] S202, performing smoothing processing on the original positioning signal from the Nth moment to the N+Lth moment to obtain the target smoothed positioning signal from the Nth moment to the Nth moment, and taking the target smoothed positioning signal of the Nth moment as the first positioning information of the Nth moment.

[0066] With reference to Figure 3 In some embodiments, S202, the smoothing processing on the original positioning signal from the Nth moment to the N+Lth moment comprises the following steps:

[0067] S2021, (1) obtaining the Euclidean distance between the original positioning signal of the Mth moment and the original positioning signal of the M+1th moment.

[0068] S2022, (2) when the Euclidean distance is greater than a preset distance threshold, confirming the interpolation weight of the original positioning signal of the Mth moment and the original positioning signal of the M+1th moment according to the Euclidean distance, and the interpolation weight is in a positive correlation with the Euclidean distance.

[0069] S2023, (3) interpolating the original positioning signal of the Mth moment and the original positioning signal of the M+1th moment according to the interpolation weight, and taking the interpolation result as the updated intermediate smoothed positioning signal of the Mth moment, wherein the interpolation formula is: slerp (the original positioning signal of the Mth moment, the original positioning signal of the M+1th moment, the interpolation weight); the greater the interpolation weight, the closer the output of slerp to the original positioning signal of the M+1th moment.

[0070] S2024、(4)For cases where N is less than or equal to L, the original positioning signal from time N to time N+L is used. After iteratively executing steps (1)(2)(3) at least 3 times, the intermediate smooth positioning signal from time N to time N+L is obtained. The intermediate smooth positioning signal from time N is used as the target smooth positioning signal, and the intermediate smooth positioning signal from time N+1 to time N+L is stored.

[0071] For cases where N is greater than L, when obtaining the original positioning signal at time N+L+1, the stored intermediate smooth positioning signal from time N+1 to time N+L is used to replace the original positioning signal from time N+1 to time N+L. Based on the stored intermediate smooth positioning signal from time N+1 to time N+L and the original positioning signal at time N+L+1, steps (1), (2), and (3) are executed iteratively.

[0072] In other words, the movement of the children's vehicle in this application has a certain lag and predictability relative to the collected first and second positioning information.

[0073] This application uses the original positioning signal from time N to time N+L to confirm the target smooth positioning signal at time N. With this setting, when the motion state changes rapidly (suddenly stops or suddenly accelerates) between time N and time N+L, the Euclidean distance between the original positioning signals of two adjacent time points will suddenly increase. We determine the interpolation weight based on the Euclidean distance, and the interpolation weight is positively correlated with the Euclidean distance. Therefore, the stroller will decelerate in advance.

[0074] Where pm = slerp(pm, pm+1, W) = (1-w)*pm + W*pm+1. Slerp can be found in existing function libraries. Here, pm represents the original positioning signal (intermediate smoothed positioning signal) at time M, pm+1 represents the original positioning signal (intermediate smoothed positioning signal) at time M+1, and W represents the interpolation weight.

[0075] Reference Figure 4 Taking the case where N is greater than L, with N being 6 and L=5, that is, the motion state changes rapidly at time N+5. Of course, this is just for the sake of explanation. In reality, the value of L is much greater than 5, such as 30, 60, or 120. The specific value is related to the frame rate of the image signal acquisition and the acquisition frequency of the UWB positioning signal. As long as the lag time is less than 1.5 seconds, it is acceptable. Taking an image acquisition frame rate of 30 frames as an example, then L is less than or equal to 45.

[0076] Reference Figure 4 The solid line represents the original positioning signal, and the dashed line represents the target smooth positioning signal.

[0077] The interpolation result of the original positioning signal at the 10th moment and the intermediate smoothing positioning signal at the 9th moment is closer to the intermediate smoothing positioning signal at the 10th moment;

[0078] The interpolation result of the intermediate smoothing positioning signal at the 9th moment and the intermediate smoothing positioning signal at the 8th moment is closer to the intermediate smoothing positioning signal at the 9th moment;

[0079] The interpolation result of the intermediate smoothing positioning signal at the 8th moment and the intermediate smoothing positioning signal at the 7th moment is closer to the intermediate smoothing positioning signal at the 8th moment;

[0080] The interpolation result of the intermediate smoothing positioning signal at the 7th moment and the intermediate smoothing positioning signal at the 6th moment is closer to the intermediate smoothing positioning signal at the 7th moment;

[0081] In summary, the interpolation result of the intermediate smoothing positioning signal at the 6th moment and the intermediate smoothing positioning signal at the 5th moment is closer to the intermediate smoothing positioning signal at the 6th moment, that is, the stroller starts to accelerate in response to the rapid change of the motion state at the 5th moment, and the scene principle of the sudden stop is the same, and the stroller will decelerate in advance.

[0082] In some embodiments, the interpolation weight is greater than or equal to 0.2 and less than or equal to 0.8; the positive correlation between the interpolation weight and the Euclidean distance is a nonlinear relationship.

[0083] When the interpolation weight changes between 0.2 and 0.8, the speed of the stroller changes. In practical applications, it is better to set the relationship between the interpolation weight and the Euclidean distance as a nonlinear relationship, specifically, the nonlinear relationship is: the larger the Euclidean distance, the larger the ratio of the interpolation weight to the Euclidean distance, and thus the situation of rapid change of the motion state can be responded faster. Exemplarily, the relationship between the interpolation weight and the Euclidean distance can be an inverse proportional function relationship, an exponential function relationship.

[0084] In practical applications, the relationship between the interpolation weight and the Euclidean distance can be obtained using a mapping table, first confirming the value range of the Euclidean distance, and then obtaining the corresponding interpolation weight based on the value range.

[0085] In other words, the interpolation weight and the value range are in a one-to-one correspondence, and are in a nonlinear relationship.

[0086] In some embodiments, the following method further comprises:

[0087] According to the following intensity and the following distance set by the user, calculate the interpolation adjustment coefficient, the interpolation adjustment coefficient is positively correlated with the following intensity, and the interpolation adjustment coefficient is negatively correlated with the tracking distance;

[0088] S2023、(3) according to the interpolation weight, the original positioning signal of the M moment and the original positioning signal of the M+1 moment are interpolated, including:

[0089] The interpolation adjustment coefficient is multiplied by the interpolation weight to obtain an updated interpolation weight;

[0090] The positioning signal of the M moment and the positioning signal of the M+1 moment are interpolated using the updated interpolation weight.

[0091] In actual application, in some scenarios, the child car needs to closely follow the adult, for example, in the vegetable market and the shopping mall, while in some scenarios, the following distance and the following strength can be appropriately released, for example, in the community downstairs and the park.

[0092] In the embodiment, the user can select the following strength as "strong", "moderate" and "weak" according to the actual scene, and of course, the gears can be increased between the adjacent two gears.

[0093] In some embodiments, S400, the first positioning information is taken as the prior estimation of the Kalman filter, the second positioning information is taken as the observation value of the Kalman filter, and the target positioning information at the current moment is calculated, including:

[0094] The smoothed positioning signal at the N moment is taken as the prior estimation of the Kalman filter, and the second positioning information at the N moment is taken as the observation value of the Kalman filter, to calculate the target positioning information at the N moment;

[0095] S500, according to the target positioning information, follow the remote controller, including:

[0096] At the N+L moment, the child car is controlled to move according to the target positioning information at the N moment, so as to follow the remote controller.

[0097] The time difference between the N moment and the N+L moment is less than or equal to 1.5 seconds.

[0098] As described above, at the N+L moment, the child car is controlled to move according to the target positioning information at the N moment, so that the child car has a certain predictability, but also causes the child car to have a certain lag, and the embodiment limits the lag time to be less than or equal to 1.5 seconds, balancing safety and reliability.

[0099] In some embodiments, the target object is tracked according to the image signal, and the second positioning information of the target object and the child car is calculated according to the area ratio and the position of the target object in the image signal, including:

[0100] The camera is controlled to continuously collect image signals with the same shooting parameters;

[0101] tracking the target object according to the continuous image signals;

[0102] calculating second positioning information of the target object according to the area ratio and the position of the target object in the image signals.

[0103] The embodiment adopts the mode of following first and then confirming the area ratio and the position. Compared with the mode of identifying the target object frame by frame and then confirming the area ratio and the position, the embodiment saves the computing power of the stroller.

[0104] Embodiment 2

[0105] The application provides a stroller, comprising: a vehicle body; a camera arranged on the vehicle body; a UWB component arranged on the vehicle body; a communication component arranged on the vehicle body, used to establish a communication connection with a remote controller; and a controller arranged on the vehicle body, the remote controller being connected with the camera, the UWB component and the communication component respectively, and the controller being configured to execute a following program, when the following program is executed, a following method of the stroller is realized.

[0106] obtaining a following distance set by a user;

[0107] calculating first positioning information of the remote controller according to original positioning signals continuously collected by the UWB component, wherein the first positioning information is three-dimensional positioning information;

[0108] tracking a target object according to image signals continuously collected by the camera, and calculating second positioning information of the target object according to an area ratio and position information of the target object in the image signals, wherein the second positioning information is three-dimensional positioning information;

[0109] calculating target positioning information at a current time by taking the first positioning information as prior estimation of a Kalman filter and taking the second positioning information as observation value of the Kalman filter; and

[0110] moving according to the target positioning information to keep the following distance to follow the remote controller.

[0111] In some embodiments, the calculation of the first positioning information of the remote controller according to the original positioning signals continuously collected by the UWB component comprises:

[0112] obtaining original positioning signals from the Nth time to the N+Lth time through at least three UWB base stations, wherein the N+Lth time is the current time;

[0113] Smooth the original positioning signals at the Nth moment to the N+Lth moment to obtain target smoothed positioning signals at the Nth moment, and the target smoothed positioning signal at the Nth moment is taken as the first positioning information at the Nth moment.

[0114] In some embodiments, the smoothing of the original positioning signals at the Nth moment to the N+Lth moment comprises:

[0115] (1) obtaining the Euclidean distance of the original positioning signal at the Mth moment and the original positioning signal at the M+1th moment;

[0116] (2) when the Euclidean distance is greater than a preset distance threshold, confirming the interpolation weight of the original positioning signal at the Mth moment and the original positioning signal at the M+1th moment according to the Euclidean distance, and the interpolation weight is in a positive correlation with the Euclidean distance;

[0117] (3) interpolating the original positioning signal at the Mth moment and the original positioning signal at the M+1th moment according to the interpolation weight, and taking the interpolation result as the intermediate smoothed positioning signal at the Mth moment, wherein the interpolation formula is slerp (original positioning signal at the Mth moment, original positioning signal at the M+1th moment, interpolation weight);

[0118] (4) for the case that N is less than or equal to L, performing steps (1), (2) and (3) iteratively at least 3 times using the original positioning signals at the Nth moment to the N+Lth moment, to obtain the intermediate smoothed positioning signals at the Nth moment to the N+Lth moment, taking the intermediate smoothed positioning signal at the Nth moment as the target smoothed positioning signal, and storing the intermediate smoothed positioning signals at the N+1th moment to the N+Lth moment;

[0119] For the case that N is greater than L, when the original positioning signal at the N+L+1th moment is obtained, the stored intermediate smoothed positioning signals at the N+1th moment to the N+Lth moment are used instead of the original positioning signals at the N+1th moment to the N+Lth moment, and steps (1), (2) and (3) are iteratively performed according to the stored intermediate smoothed positioning signals at the N+1th moment to the N+Lth moment and the original positioning signal at the N+L+1th moment.

[0120] In some embodiments, the interpolation weight is greater than or equal to 0.2 and less than or equal to 0.8, and the positive correlation between the interpolation weight and the Euclidean distance is a nonlinear relationship, and the nonlinear relationship is that the greater the Euclidean distance, the smaller the ratio of the interpolation weight to the Euclidean distance.

[0121] In some embodiments, the following steps are further included in the following method:

[0122] According to the following strength and the following distance set by the user, an interpolation adjustment coefficient is calculated, the interpolation adjustment coefficient is in a positive correlation with the following strength, and the interpolation adjustment coefficient is in a negative correlation with the tracking distance;

[0123] The interpolation of the original positioning signal at the Mth moment and the original positioning signal at the M+1th moment according to the interpolation weight comprises:

[0124] The interpolation adjustment coefficient is multiplied by the interpolation weight to obtain an updated interpolation weight;

[0125] The Mth moment and the M+1th moment of the positioning signal are interpolated using the updated interpolation weight.

[0126] In some embodiments, the first positioning information is taken as a priori estimation of the Kalman filter, the second positioning information is taken as an observation value of the Kalman filter, and the target positioning information at the current moment is calculated, comprising:

[0127] The smoothed positioning signal at the Nth moment is taken as a priori estimation of the Kalman filter, and the second positioning information at the Nth moment is taken as an observation value of the Kalman filter, and the target positioning information at the Nth moment is calculated;

[0128] The target positioning information is used to follow the remote controller, comprising:

[0129] At the N+Lth moment, the target positioning information at the Nth moment is used to control the movement of the stroller to follow the remote controller.

[0130] In some embodiments, the time difference between the Nth moment and the N+Lth moment is less than or equal to 1.5 seconds.

[0131] In some embodiments, the target object is tracked according to the image signal, and the second positioning information of the target object and the stroller is calculated according to the area ratio and the position of the target object in the image signal, comprising:

[0132] The camera is controlled to continuously capture image signals with the same shooting parameters;

[0133] The target object is tracked according to the continuous image signal;

[0134] The second positioning information of the target object and the stroller is calculated according to the area ratio and the position of the target object in the image signal.

[0135] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0136] Those skilled in the art will appreciate that embodiments of the present application can be devised for a variety of applications. It is intended that the present application be limited only by the scope of the appended claims, and it is intended that various modifications and alterations made by those skilled in the art be considered as within the scope of the present application. The embodiments of the present application will be described with reference to the attached drawings identified below.

[0137] The present application is described in reference to the drawings using a flowchart and / or a block diagram of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagram, and combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0138] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0140] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those skilled in the art once they learn of the basic inventive concepts. Therefore, the appended claims are intended to cover all such modifications and variations as fall within the scope of the present application.

[0141] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for automatic following of a children's stroller, characterized in that, Applicable to children's vehicles, the children's vehicle includes: a UWB component, a single camera, and a remote control; wherein the remote control is equipped with a UWB tag, and the following method includes: Get the user-defined follow distance; Based on the raw positioning signals continuously acquired by the UWB component, the first positioning information of the remote controller is calculated, wherein the first positioning information is three-dimensional positioning information; the raw positioning signals from time N to time N+L are smoothed to obtain the target smoothed positioning signal at time N, and the target smoothed positioning signal at time N is used as the first positioning information at time N; wherein, the smoothing of the raw positioning signals from time N to time N+L includes: (1) Obtain the Euclidean distance between the original positioning signal at time M and the original positioning signal at time M+1; (2) When the Euclidean distance is greater than the preset distance threshold, the interpolation weights of the original positioning signal at time M and the original positioning signal at time M+1 are determined according to the Euclidean distance. The interpolation weights are positively correlated with the Euclidean distance. (3) Interpolate the original positioning signal at time M and the original positioning signal at time M+1 according to the interpolation weight, and use the interpolation result as the updated intermediate smooth positioning signal at time M, wherein the interpolation formula is: slerp(original positioning signal at time M, original positioning signal at time M+1, interpolation weight). (4) For the case where N is less than or equal to L, the original positioning signal from time N to time N+L is used to iterate and execute steps (1), (2) and (3) at least 3 times to obtain the intermediate smooth positioning signal from time N to time N+L. The intermediate smooth positioning signal from time N is used as the target smooth positioning signal, and the intermediate smooth positioning signal from time N+1 to time N+L is stored. For cases where N is greater than L, when obtaining the original positioning signal at time N+L+1, the stored intermediate smooth positioning signal from time N+1 to time N+L is used to replace the original positioning signal from time N+1 to time N+L. Based on the stored intermediate smooth positioning signal from time N+1 to time N+L and the original positioning signal at time N+L+1, steps (1), (2), and (3) are executed iteratively. The target object is tracked based on the image signals continuously acquired by the camera, and the second positioning information of the target object is calculated based on the area ratio and position information of the target object in the image signal; wherein, the second positioning information is three-dimensional positioning information; Using the first positioning information as a priori estimate of the Kalman filter and the second positioning information as the observation value of the Kalman filter, the target positioning information at the current time is calculated; and Move according to the target positioning information to maintain the following distance and follow the remote control; Here, the N+Lth time is the current time.

2. The automatic following method for a children's stroller as described in claim 1, characterized in that, The step of calculating the first positioning information of the remote controller based on the raw positioning signals continuously collected by the UWB component includes: The original positioning signals from time N to time N+L are obtained using at least three UWB base stations.

3. The automatic following method for a children's stroller as described in claim 1, characterized in that, The interpolation weight is greater than or equal to 0.2 and less than or equal to 0.

8.

4. The automatic following method for a children's stroller as described in claim 1, characterized in that, The positive correlation between the interpolation weights and the Euclidean distance is a non-linear relationship. The non-linear relationship is that the larger the Euclidean distance, the smaller the ratio of the interpolation weights to the Euclidean distance.

5. The automatic following method for a children's vehicle as described in claim 1, characterized in that, Also includes: Based on the user-set following intensity and following distance, an interpolation adjustment coefficient is calculated. The interpolation adjustment coefficient is positively correlated with the following intensity and negatively correlated with the following distance. The step of interpolating the original positioning signal at time M and the original positioning signal at time M+1 according to the interpolation weight includes: Multiply the interpolation adjustment coefficient by the interpolation weight to obtain the updated interpolation weight; The updated interpolation weights are used to interpolate the positioning signal at time M and the positioning signal at time M+1.

6. The automatic following method for a children's vehicle as described in claim 1, characterized in that, The step of using the first positioning information as a priori estimate of the Kalman filter and the second positioning information as an observation of the Kalman filter to calculate the target positioning information at the current moment includes: The smoothed positioning signal at time N is used as the prior estimate of the Kalman filter, and the second positioning information at time N is used as the observation value of the Kalman filter to calculate the target positioning information at time N. The step of following the remote control based on the target positioning information includes: At time N+L, the stroller is controlled to move according to the target positioning information at time N, so as to follow the remote control.

7. The automatic following method for a children's vehicle as described in claim 6, characterized in that, The time difference between time N and time N+L is less than or equal to 1.5 seconds.

8. The automatic following method for a children's stroller as described in any one of claims 1-7, characterized in that, The step of tracking the target object based on the image signal and calculating the second positioning information of the target object and the stroller based on the area ratio and position of the target object in the image signal includes: Control the camera to continuously acquire image signals with the same shooting parameters; The target object is tracked based on continuous image signals; The second positioning information of the target object and the stroller is calculated based on the area ratio and position of the target object in the image signal.

9. A children's stroller, characterized in that, include: Body; A single camera is installed on the vehicle body; UWB components are installed on the vehicle body; A communication component, installed on the vehicle body, is used to establish a communication connection with the remote controller; A controller is disposed on the vehicle body, and the remote controller is connected to the camera, the UWB component and the communication component respectively. The controller is configured to execute a follow program, and when the follow program is executed, it implements the automatic following method for children's vehicles as described in any one of claims 1-8.

Images