Control Method and Related Device for Child Safety Seat

Abstract

The present application provides a control method and related device for a child safety seat, the method comprising: when a preset folding event occurs, detecting whether the child safety seat is used by a child; if the child safety seat is not used by the child, controlling a side wing drive mechanism to drive the side wing to rotate, and controlling a base drive mechanism to drive the base to rotate, so that the child safety seat is converted from an unfolded state to a folded state; if the child safety seat is used by the child, using a vehicle-mounted display screen to display a prompt message to prompt the user that the child is using the child safety seat. Under the premise of automatically folding the side wings and the base, the safety hazard of the child being pinched by the side wings and the base is reduced.

Description

Technical Field

[0001] This application relates to the technical fields of automotive safety seats, computer vision detection, and deep learning, and particularly relates to a control method and related device for child safety seats. Background Art

[0002] With the development of society, the probability of children taking private cars has increased sharply. Due to the differences between children and adults, the safety of children while riding in cars has gradually attracted people's attention. Automotive child safety seats are one type of child restraint system and play an important role in preventing children from dying or suffering serious injuries in collision accidents. Most ordinary automotive child safety seats are independent add-on devices that fix the child's body, prevent them from being injured by collisions during driving, and provide corresponding protection in the event of an accident.

[0003] Patent CN109747492A discloses a child safety seat wing opening and closing device, including wings, and also including a motor and a wing connection assembly. The wings are movably connected to the seat body. The motor is disposed on the seat body or the wings. The drive shaft of the motor is linked with the wing connection assembly, and the wing connection assembly is linked with the wings, such that when the motor acts, the drive shaft drives the wing connection assembly to displace, thereby driving the wings to unfold or retract. Although this kind of child safety seat can achieve automatic opening and closing of the wings, the safety of children is not considered during the process of retracting the wings. In case a child randomly bumps around on the safety seat and accidentally triggers the retraction switch while sitting, it may cause the body to be clamped by the wings, and in severe cases, it may pose a life-threatening danger.

[0004] Therefore, there is an urgent need to provide a control method and related device for child safety seats to improve the prior art. Summary of the Invention

[0005] The purpose of this application is to provide a control method and related device for child safety seats, which can reduce the potential safety hazard of children being clamped by the wings and the base on the premise of automatically retracting the wings and the base.

[0006] The purpose of this application is achieved by adopting the following technical solutions:

[0007] In a first aspect, this application provides a control method for a child safety seat, and the child safety seat includes:

[0008] A base;

[0009] A backrest, the lower part of the backrest is rotatably connected to the rear part of the base through a first pivot;

[0010] Wings, the wings are rotatably connected to the side of the backrest through a second pivot;

[0011] A connecting component, which is arranged at the lower part of the backrest and / or the rear part of the base and is used for connecting with an automobile seat;

[0012] The child safety seat has a deployed state and a folded state. When the child safety seat is converted from the deployed state to the folded state, the base rotates relative to the backrest around the first pivot axis, so that the upper surface of the base approaches the front surface of the backrest, and the side wings rotate relative to the backrest around the second pivot axis, so that the side wings approach the front surface of the backrest;

[0013] The method includes:

[0014] When a preset folding event occurs, detecting whether the child safety seat is being used by a child;

[0015] If the child safety seat is not being used by the child, controlling the side wing driving mechanism to drive the side wings to rotate and controlling the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from the deployed state to the folded state;

[0016] If the child safety seat is being used by the child, using an in-vehicle display screen to display a prompt message to prompt the user that the child is using the child safety seat;

[0017] The preset folding event includes at least one of the following:

[0018] Within a preset time period, the maximum value of the pressure received by the base is not greater than a preset pressure threshold, where the pressure received by the base is sensed by a pressure sensor arranged on the upper surface of the base;

[0019] The interaction component receives a folding trigger operation, and the interaction component includes at least one of the following: a voice recognition component, an image recognition component, an infrared receiving component, a button, and a touch screen.

[0020] The beneficial effects of this technical solution are as follows: Both the base and the side wings of the child safety seat can rotate. When a preset folding event occurs, it is necessary to detect whether the child safety seat is being used by a child, that is, to detect whether the child's body part is within the moving range of the child safety seat (the rotation range of the side wings and the rotation range of the base). On the one hand, the preset folding event can be that the maximum value of the pressure received by the base within a preset time duration is not greater than a preset pressure threshold. If the pressure value received by the base is small, it indicates that the child may have left the safety seat (it is not excluded that the child may put a hand or a foot on the base out of curiosity, so it is still necessary to detect whether the child safety seat is being used by a child). On the other hand, the preset event can be that the interaction component receives a folding trigger operation. The interaction component can be a voice recognition component to recognize the voice information of the people in the vehicle, or an image recognition component to recognize the body movements such as gestures of the people in the vehicle, or an infrared receiving component. The user can send an infrared signal to the infrared receiving component using an infrared remote control, or it can be a button or a touch screen, and the user triggers it manually.

[0021] When it is detected that the child safety seat is not being used by the child, the rotation of the side wings and the base will not pinch the child, and the corresponding driving mechanism can be used to achieve the automatic folding of the side wings and the base, saving the rear seat space; when it is detected that the child safety seat is being used by the child, the rotation of the side wings and / or the base may pinch the child. At this time, the side wings and the base are not folded, and the in-vehicle display screen is used to prompt the people in the vehicle that the child's body part is within the moving range of the child safety seat, avoiding the child being pinched.

[0022] In summary, the control method of the child safety seat takes into account the safety of the child. It detects whether the child safety seat is being used by the child on the premise of automatically folding the side wings and the base, reducing the safety hazard of the child being pinched by the side wings and the base. Even if there are few people in the vehicle (adults other than children), or even only the driver, there is no need to constantly watch the child, and there is no need to worry about the child accidentally triggering the folding mechanism and being pinched. The driver can drive with confidence, which improves the enthusiasm for using the child safety seat.

[0023] In some alternative embodiments, the detection of whether the child safety seat is being used by the child includes:

[0024] Using a first camera to photograph the child safety seat to obtain a seat image;

[0025] Based on the seat image, detecting whether the child safety seat is being used by the child.

[0026] The beneficial effects of this technical solution are as follows: The first camera can be used to photograph the child safety seat to obtain seat images from multiple angles, and whether there are body parts of a child in the seat images is detected to determine whether the child seat is being used by a child. This detection method is relatively simple and direct, and the detection results are relatively reliable.

[0027] In some alternative embodiments, detecting whether the child safety seat is being used based on the seat image includes:

[0028] Based on the seat image, detecting whether the body part of the child is within the rotation range of the side wing and detecting whether the body part of the child is within the rotation range of the base;

[0029] When the body part of the child is within the rotation range of the side wing or the rotation range of the base, it is determined that the child safety seat is being used by the child;

[0030] When the body part of the child is not within the rotation range of the side wing and not within the rotation range of the base, it is determined that the child safety seat is not being used by the child.

[0031] The beneficial effects of this technical solution are as follows: The activity spaces of the side wing and the base may not completely overlap. Artificial intelligence means (such as machine learning models, deep learning models, reinforcement learning models, etc.) can be used to analyze and process the seat image, detect whether the body part of the child is within the rotation range of the side wing, and detect whether the body part of the child is within the rotation range of the base. If the body part of the child is within the rotation range of the side wing or the base, it is determined that the child safety seat is being used by the child. If the body part of the child is neither within the rotation range of the side wing nor within the rotation range of the base, it is determined that the child safety seat is not being used. By detecting the spatial position relationship between the body part of the child and the seat, the risk of the child being clamped can be further determined, improving the accuracy of the detection results.

[0032] In some alternative embodiments, detecting whether the child safety seat is being used includes:

[0033] Inputting the seat image into a seat detection model to obtain a seat detection result, where the seat detection result is used to indicate whether the child safety seat is being used;

[0034] Among them, the training process of the seat detection model includes:

[0035] Obtaining a training set, where the training set includes a plurality of training data, and each training data includes a sample seat image and annotation data of the seat detection result corresponding to the sample seat image;

[0036] For each piece of training data in the training set, perform the following processing:

[0037] Input the sample seat image in the training data into a preset deep learning model to obtain prediction data of the seat detection result corresponding to the sample seat image;

[0038] Update the model parameters of the deep learning model based on the prediction data and annotation data of the seat detection result corresponding to the sample seat image;

[0039] Detect whether a preset training end condition is met; if so, use the trained deep learning model as the seat detection model; if not, continue to train the deep learning model using the next piece of training data.

[0040] The beneficial effect of this technical solution is that the essence of deep learning is to learn more useful features by constructing a machine learning model with many hidden layers and a large amount of training data, so as to improve the accuracy of classification or prediction. By designing and establishing an appropriate number of neuron computing nodes and a multi-layer operation hierarchy, selecting appropriate input and output layers, and through the learning and optimization of the network, a functional relationship from input to output is established. Although the functional relationship between input and output cannot be found 100%, it can approximate the real association relationship as much as possible. The seat detection model trained thereby can obtain the corresponding seat detection result based on the seat image, and the calculation result has high accuracy and high reliability.

[0041] In some optional embodiments, the method further includes:

[0042] During the process of the child safety seat transitioning from the deployed state to the retracted state, use the vehicle-mounted display screen to display the seat image captured in real time by the first camera.

[0043] The beneficial effect of this technical solution is that during the process of retracting the child safety seat, the seat image captured by the first camera can be displayed in real time on the vehicle-mounted display screen, allowing the vehicle occupants in the front row (the driver or the co-driver) to view in real time whether the child is touching the retracting child safety seat, and further avoiding the child from being pinched through an intelligent plus manual method.

[0044] In some optional embodiments, the child safety seat further has a restraint state;

[0045] In the deployed state, the included angle between the side wings and the backrest is a first included angle; in the restraint state, the included angle between the side wings and the backrest is a second included angle; in the retracted state, the side wings are attached to the backrest; the first included angle is greater than the second included angle, and the second included angle is an acute angle;

[0046] The method further includes:

[0047] Obtaining the vehicle speed of the vehicle;

[0048] When it is detected that the vehicle speed of the vehicle is not less than a preset speed threshold, controlling the flank driving mechanism to drive the flank to rotate, and controlling the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from an unfolded state to a restraint state.

[0049] The beneficial effect of this technical solution is that: in addition to the unfolded state and the folded state, the child safety seat also has a restraint state. In the restraint state, the second included angle between the flank and the backrest is smaller than the first included angle in the unfolded state; in the unfolded state, the first included angle can be 90°, 120° or 150°. In this way, the flank opens outwards, facilitating the child to enter the child safety seat. When the vehicle starts to drive at a high speed, the flank driving mechanism can be used to drive the flank to retract inwards, so that the included angle between the flank and the backrest is an acute angle. In this way, the flank can wrap the child, playing a role of buffering and protection to avoid the child being injured in a sudden accident.

[0050] In some alternative embodiments, the method further includes:

[0051] When it is detected that the vehicle speed of the vehicle is less than the preset speed threshold, controlling the flank driving mechanism to drive the flank to rotate, and controlling the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from a restraint state to an unfolded state.

[0052] The beneficial effect of this technical solution is that: when the vehicle starts to decelerate and prepare to stop, the flank driving mechanism can be used to drive the flank to automatically open outwards, facilitating the vehicle occupants to take the child out of the child safety seat.

[0053] In some alternative embodiments, the process of determining the second included angle includes:

[0054] Using a second camera to capture the rear seat area to obtain a rear seat image, and the display content of the rear seat image includes the child;

[0055] Based on the rear seat image, obtaining the body size information of the child;

[0056] Based on the body size information, determining the second included angle.

[0057] The beneficial effects of this technical solution are as follows: The second included angle can be determined according to the body size information of the child. When the rear seat is opened and the child gets on the vehicle, the rear seat area is photographed by the second camera to obtain a rear seat image, and the body size information of the child is obtained from the rear seat image. For a child with a relatively plump body, the second included angle can be set larger, such as 70° or 80°. For a child with a relatively thin body, the second included angle can be set smaller, such as 50° or 60°. This method of adaptively adjusting the included angle between the side wings and the backrest according to the body size of the child not only avoids the fatter child being too tightly bound by the side wings, but also avoids the thinner child not being bound by the side wings, greatly improving the user experience.

[0058] Second, the present application provides a child safety seat, which includes:

[0059] A base;

[0060] A backrest, the lower part of the backrest is rotatably connected to the rear part of the base through a first pivot;

[0061] Side wings, the side wings are rotatably connected to the side of the backrest through a second pivot;

[0062] A connecting component, the connecting component is arranged on the lower part of the backrest and / or the rear part of the base for connecting with the vehicle seat;

[0063] The child safety seat has a deployed state and a retracted state. When the child safety seat is converted from the deployed state to the retracted state, the base rotates relative to the backrest around the first pivot, so that the upper surface of the base approaches the front surface of the backrest, and the side wings rotate relative to the backrest around the second pivot, so that the side wings approach the front surface of the backrest;

[0064] The child safety seat further includes a memory and a processor. The memory stores a computer program, and the processor is configured to perform the following steps when executing the computer program:

[0065] When a preset retraction event occurs, detect whether the child safety seat is being used by a child;

[0066] If the child safety seat is not being used by the child, control the side wing driving mechanism to drive the side wings to rotate, and control the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from the deployed state to the retracted state;

[0067] If the child safety seat is being used by the child, use the vehicle-mounted display screen to display a prompt message to prompt the user that the child is using the child safety seat;

[0068] The preset retraction event includes at least one of the following:

[0069] Within a preset time duration, the maximum value of the pressure received by the base is not greater than a preset pressure threshold, where the pressure received by the base is sensed by a pressure sensor disposed on the upper surface of the base;

[0070] The interaction component receives a folding trigger operation, and the interaction component includes at least one of the following: a voice recognition component, an image recognition component, an infrared receiving component, a button, and a touch screen.

[0071] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to detect whether the child safety seat is being used by a child:

[0072] Use a first camera to capture the child safety seat to obtain a seat image;

[0073] Based on the seat image, detect whether the child safety seat is being used by a child.

[0074] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to detect whether the child safety seat is being used by a child:

[0075] Based on the seat image, detect whether the child's body part is within the rotation range of the side wing and detect whether the child's body part is within the rotation range of the base;

[0076] When the child's body part is within the rotation range of the side wing or the rotation range of the base, determine that the child safety seat is being used by the child;

[0077] When the child's body part is not within the rotation range of the side wing and not within the rotation range of the base, determine that the child safety seat is not being used by the child.

[0078] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to detect whether the child safety seat is being used by a child:

[0079] Input the seat image into a seat detection model to obtain a seat detection result, where the seat detection result is used to indicate whether the child safety seat is being used by a child;

[0080] Wherein, the training process of the seat detection model includes:

[0081] Obtain a training set, where the training set includes a plurality of training data, and each training data includes a sample seat image and annotation data of the seat detection result corresponding to the sample seat image;

[0082] For each piece of training data in the training set, perform the following processing:

[0083] Input the sample seat image in the training data into a preset deep learning model to obtain prediction data of the seat detection result corresponding to the sample seat image;

[0084] Update the model parameters of the deep learning model based on the prediction data and annotation data of the seat detection result corresponding to the sample seat image;

[0085] Detect whether a preset training end condition is met; if so, use the trained deep learning model as the seat detection model; if not, continue to train the deep learning model using the next piece of training data.

[0086] In some optional embodiments, the processor is further configured to implement the following steps when executing the computer program:

[0087] During the process of the child safety seat transitioning from the deployed state to the folded state, display the seat image captured in real time by the first camera on the in-vehicle display screen.

[0088] In some optional embodiments, the child safety seat further has a restraint state;

[0089] In the deployed state, the included angle between the side wings and the backrest is a first included angle; in the restraint state, the included angle between the side wings and the backrest is a second included angle; in the folded state, the side wings are attached to the backrest; the first included angle is greater than the second included angle, and the second included angle is an acute angle;

[0090] The processor is further configured to implement the following steps when executing the computer program:

[0091] Obtain the vehicle speed of the vehicle;

[0092] When it is detected that the vehicle speed of the vehicle is not less than a preset speed threshold, control the side wing drive mechanism to drive the side wings to rotate, and control the base drive mechanism to drive the base to rotate, so that the child safety seat transitions from the deployed state to the restraint state.

[0093] In some optional embodiments, the processor is further configured to implement the following steps when executing the computer program:

[0094] When it is detected that the vehicle speed of the vehicle is less than the preset speed threshold, control the side wing drive mechanism to drive the side wings to rotate, and control the base drive mechanism to drive the base to rotate, so that the child safety seat transitions from the restraint state to the deployed state.

[0095] In some alternative embodiments, when the processor is configured to execute the computer program, the second included angle is determined in the following manner:

[0096] The rear seat area is photographed by a second camera to obtain a rear seat image, and the display content of the rear seat image includes the child;

[0097] Based on the rear seat image, the body size information of the child is obtained;

[0098] Based on the body size information, the second included angle is determined.

[0099] In a third aspect, the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of any one of the above-described control methods for a child safety seat are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] The present application will be further described below in conjunction with the drawings and embodiments.

[0101] Figure 1 FIG. is a flowchart of a control method for a child safety seat provided by an embodiment of the present application.

[0102] Figure 2 FIG. is a structural diagram of a child safety seat provided by an embodiment of the present application.

[0103] Figure 3 FIG. is a structural block diagram of a child safety seat provided by an embodiment of the present application.

[0104] Figure 4 FIG. is a structural block diagram of a program product provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0105] Hereinafter, the present application will be further described in conjunction with the drawings and specific embodiments. It should be noted that, on the premise of no conflict, the following-described embodiments or technical features may be arbitrarily combined to form new embodiments.

[0106] In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B may be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" or its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c may represent: a, b, c, a and b, a and c, b and c, a and b and c, where a, b, and c may be single or multiple. It should be noted that "at least one" can also be interpreted as "one or more".

[0107] It should also be noted that in the embodiments of the present application, words such as "exemplary" or "for example" are used to give examples, illustrations, or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0108] See Figure 1 , Figure 1 is a schematic flowchart of a control method for a child safety seat provided by an embodiment of the present application.

[0109] The child safety seat includes:

[0110] A base;

[0111] A backrest, the lower part of the backrest is rotatably connected to the rear part of the base through a first pivot;

[0112] Side wings, the side wings are rotatably connected to the side of the backrest through a second pivot;

[0113] A connection component, the connection component is arranged at the lower part of the backrest and / or the rear part of the base for connecting to an automobile seat;

[0114] The child safety seat has a deployed state and a folded state. When the child safety seat is converted from the deployed state to the folded state, the base rotates relative to the backrest around the first pivot, so that the upper surface of the base approaches the front surface of the backrest, and the side wings rotate relative to the backrest around the second pivot, so that the side wings approach the front surface of the backrest;

[0115] The method includes:

[0116] Step S101: When a preset folding event occurs, detect whether the child safety seat is being used by a child;

[0117] Step S102: If the child safety seat is not being used by the child, control the wing driving mechanism to drive the wings to rotate and control the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from the unfolded state to the folded state;

[0118] Step S103: If the child safety seat is being used by the child, use the in-vehicle display screen to display a prompt message to prompt the user that the child is using the child safety seat;

[0119] The preset folding event includes at least one of the following:

[0120] Within a preset time period, the maximum value of the pressure received by the base is not greater than a preset pressure threshold, where the pressure received by the base is sensed by a pressure sensor provided on the upper surface of the base;

[0121] The interaction component receives a folding trigger operation, and the interaction component includes at least one of the following: a voice recognition component, an image recognition component, an infrared receiving component, a button, and a touch screen.

[0122] Thus, both the base and the wings of the child safety seat can rotate. When a preset folding event occurs, it is detected whether the child safety seat is being used by a child, that is, it is necessary to detect whether the child's body part is within the activity range of the child safety seat (the rotation range of the wings and the rotation range of the base). On the one hand, the preset folding event can be that the maximum value of the pressure received by the base within a preset time period is not greater than the preset pressure threshold. If the pressure value received by the base is small, it indicates that the child may have left the safety seat (it is not excluded that the child may put a hand or a foot on the base out of curiosity, so it is still necessary to detect whether the child safety seat is being used by a child). On the other hand, the preset event can be that the interaction component receives a folding trigger operation. The interaction component can be a voice recognition component to recognize the voice information of the people in the vehicle, or an image recognition component to recognize the gestures and other body movements of the people in the vehicle, or an infrared receiving component, and the user can send an infrared signal to the infrared receiving component using an infrared remote control, or a button or a touch screen, and the user triggers it manually.

[0123] When it is detected that the child safety seat is not being used by the child, the rotation of the wings and the base will not pinch the child, and the corresponding driving mechanism can be used to realize the automatic folding of the wings and the base, saving the rear seat space; when it is detected that the child safety seat is being used by the child, the rotation of the wings and / or the base may pinch the child. At this time, the wings and the base are not folded, and the in-vehicle display screen is used to prompt the people in the vehicle that the child's body part is located within the activity range of the child safety seat, avoiding the child being pinched.

[0124] To sum up, the control method of the child safety seat takes the safety of children into consideration. It detects whether the child safety seat is used by a child while automatically folding the side wings and the base, thereby reducing the safety hazard of children being pinched by the side wings and the base. Even if there are fewer people in the car (adults other than children), or even only the driver, there is no need to keep an eye on the children all the time, and there is no need to worry about the children accidentally triggering the folding mechanism and being pinched. The driver can drive with peace of mind, which increases the enthusiasm for using child safety seats.

[0125] In some embodiments, the side wings may include a left wing disposed on the left side of the backrest and a right wing disposed on the right side of the backrest, and the folding and unfolding of the left wing and the right wing may be performed synchronously or asynchronously.

[0126] The base driving mechanism can be a motor or a cylinder, and the wing driving mechanism can be a motor or a cylinder.

[0127] When the child safety seat is converted from the unfolded state to the folded state, the base can be folded upward first, and then the side wings can be folded inward; or, the side wings can be folded inward first, and then the base can be folded upward.

[0128] The specific process of folding the base upwards first and then folding the side wings inwards is as follows:

[0129] First, the base driving mechanism is used to drive the base to rotate so that the upper surface of the base is close to the front surface of the backrest until the upper surface of the base is attached to the front surface of the backrest; then the side wing driving mechanism is used to drive the left wing and the right wing to rotate synchronously so that the left wing and the right wing are attached to the backrest surface.

[0130] The specific process of folding the side wings inwards first and then the base upwards is as follows:

[0131] First, the side wing driving mechanism is used to drive the left wing and the right wing to rotate synchronously so that the left wing and the right wing are attached to the front surface of the backrest; then the base driving mechanism is used to drive the base to rotate so that the upper surface of the base is close to the front surface of the backrest until the upper surface of the base is attached to the surface of the left wing and the right wing.

[0132] The embodiment of the present application does not limit the preset time length and the preset pressure threshold. The preset time length is, for example, 1 minute, 2 minutes or 3 minutes, and the preset pressure threshold is, for example, 1 kg, 2 kg or 3 kg.

[0133] The on-board display screen can be set next to the dashboard of the car, so that the driver and co-driver in the front row can view prompt information conveniently. The prompt information can be text, image or video information, and the prompt content can be "Children are using the seat, please do not fold the seat."

[0134] The interaction component can be set on other (adult) seats of the vehicle or on a child safety seat.

[0135] When the interaction component is a voice recognition component, the folding trigger operation can be a voice trigger operation.

[0136] The voice recognition component can include a microphone and a voice recognition chip. The microphone is used to collect the voice information of the vehicle occupants, and the voice recognition chip is used to recognize the voice information. When it is detected that the voice information contains a preset keyword, it is determined that the voice recognition component has received a voice trigger operation. The preset keyword can be "fold the seat", "collapse the seat", "stow the seat", etc.

[0137] When the interaction component is an image recognition component, the folding trigger operation can be an action trigger operation.

[0138] The image recognition component can include a third camera and an image recognition chip. The third camera is used to collect the image corresponding to the vehicle occupants, and the image recognition chip is used to detect whether the person in the image makes a preset action. When it is detected that the person in the image makes a preset action, it is determined that the image recognition component has received an action trigger operation. The preset action can be "close the palm inward", "bend the elbow", "swing the right hand", etc.

[0139] When the interaction component is an infrared receiving component, the folding trigger operation can be the operation of a vehicle occupant pressing an infrared remote control; when the interaction component is a button, the folding trigger operation can be the operation of pressing the button; when the interaction component is a knob, the folding trigger operation can be the operation of rotating the knob; when the interaction component is a touch screen, the folding trigger operation can be a touch operation, and the touch screen can be an in-vehicle display screen.

[0140] In some alternative embodiments, the detecting whether the child safety seat is being used by a child includes:

[0141] Using a first camera to photograph the child safety seat to obtain a seat image;

[0142] Based on the seat image, detecting whether the child safety seat is being used by a child.

[0143] Thus, a first camera can be used to photograph the child safety seat to obtain seat images from multiple angles, and whether a child's body part appears in the seat images can be detected to determine whether the child seat is being used by a child. This detection method is relatively simple and direct, and the detection result is relatively reliable.

[0144] In some other embodiments, the child safety seat is provided with a first set of optoelectronic switches on the backrest and the base (the transmitting unit is on the backrest and the receiving unit is on the base, or the transmitting unit is on the base and the receiving unit is on the backrest), and a second set of optoelectronic switches is provided on the left wing and the right wing (the transmitting unit is on the left wing and the receiving unit is on the right wing, or the transmitting unit is on the right wing and the receiving unit is on the left wing);

[0145] Detecting whether the child safety seat is used by a child includes:

[0146] Using the first set of optoelectronic switches to detect whether the child's body part is within the rotation range of the base, and using the second set of optoelectronic switches to detect whether the child's body part is within the rotation range of the wing;

[0147] When the child's body part is within the rotation range of the wing or the rotation range of the base, it is determined that the child safety seat is used by the child;

[0148] When the child's body part is not within the rotation range of the wing and not within the rotation range of the base, it is determined that the child safety seat is not used by the child.

[0149] An optoelectronic switch is short for an optoelectronic proximity switch. The transmitting unit and the receiving unit of the optoelectronic switch are arranged opposite to each other. When there is no object blocking, the optical signal emitted by the transmitting unit can be received by the receiving unit. When there is an object blocking, the optical signal emitted by the transmitting unit cannot be received by the receiving unit. By using the blocking or reflection of the optical signal by the object, it is detected whether the child's body part is within the activity space of the child safety seat.

[0150] The rotation range of the wing refers to the area passed through during the rotation of the wing, and the rotation range of the base refers to the area passed through during the rotation of the base.

[0151] In some alternative embodiments, detecting whether the child safety seat is used by a child based on the seat image includes:

[0152] Based on the seat image, detecting whether the child's body part is within the rotation range of the wing and detecting whether the child's body part is within the rotation range of the base;

[0153] When the child's body part is within the rotation range of the wing or the rotation range of the base, it is determined that the child safety seat is used by the child;

[0154] When the child's body part is not within the rotation range of the wing and not within the rotation range of the base, it is determined that the child safety seat is not used by the child.

[0155] Thus, the movement space of the side wings and the movement space of the base may not completely overlap. Artificial intelligence means (such as machine learning models, deep learning models, reinforcement learning models, etc.) can be used to analyze and process the seat image to detect whether the child's body part is within the rotation range of the side wings and to detect whether the child's body part is within the rotation range of the base. If the child's body part is within the rotation range of the side wings or the base, it is determined that the child safety seat is being used by the child. If the child's body part is neither within the rotation range of the side wings nor within the rotation range of the base, it is determined that the child safety seat is not being used by the child. By detecting the spatial position relationship between the child's body part and the seat, the risk of the child being pinched can be further determined, improving the accuracy of the detection result.

[0156] The child's body part can be a partial body, such as the head, hands, feet, buttocks, etc.

[0157] In some embodiments, detecting whether the child's body part is within the rotation range of the side wings and detecting whether the child's body part is within the rotation range of the base based on the seat image includes:

[0158] Input the seat image into a target detection model to obtain the first relative position relationship between the child's body part and the side wings and the second relative position relationship between the child's body part and the base;

[0159] Judge whether the child's body part is within the rotation range of the side wings according to the first relative position relationship;

[0160] Judge whether the child's body part is within the rotation range of the base according to the second relative position relationship.

[0161] In some alternative embodiments, detecting whether the child safety seat is being used by the child includes:

[0162] Input the seat image into a seat detection model to obtain a seat detection result, and the seat detection result is used to indicate whether the child safety seat is being used by the child;

[0163] Wherein, the training process of the seat detection model includes:

[0164] Obtain a training set, the training set includes a plurality of training data, and each training data includes a sample seat image and the annotation data of the seat detection result corresponding to the sample seat image;

[0165] For each training data in the training set, perform the following processing:

[0166] Input the sample seat images in the training data into a preset deep learning model to obtain prediction data of the seat detection results corresponding to the sample seat images;

[0167] Update the model parameters of the deep learning model based on the prediction data and annotation data of the seat detection results corresponding to the sample seat images;

[0168] Detect whether a preset training end condition is satisfied; if so, use the trained deep learning model as the seat detection model; if not, continue to train the deep learning model with the next piece of training data.

[0169] Thus, the essence of deep learning is to learn more useful features by constructing a machine learning model with many hidden layers and a large amount of training data, so as to improve the accuracy of classification or prediction. By designing and establishing an appropriate number of neuron computing nodes and a multi-layer operation hierarchy, selecting appropriate input and output layers, and through the learning and optimization of the network, a functional relationship from input to output is established. Although the functional relationship between input and output cannot be found 100%, it can approximate the real correlation relationship as much as possible. The seat detection model trained in this way can obtain the corresponding seat detection results based on the seat images, and the calculation results are highly accurate and reliable.

[0170] Different from the above method of using the object detection model to judge whether a child safety seat is being used by a child (the output result has two, that is, two relative position relationships, and the final conclusion: whether the child safety seat is being used by a child needs to be further analyzed based on the output result), the method of using the deep learning model for detection is more direct. The preset deep learning model is trained with a large amount of training data to obtain a seat detection model, and the output result is obtained in one step, directly obtaining the seat detection result, that is, whether the child safety seat is being used by a child.

[0171] In some alternative embodiments, the method further includes:

[0172] During the process of the child safety seat being converted from the unfolded state to the folded state, use the in-vehicle display screen to display the seat images captured in real time by the first camera.

[0173] Thus, during the process of folding the child safety seat, the seat images captured by the first camera can be displayed in real time on the in-vehicle display screen, allowing the vehicle occupants in the front row (the driver or the co-pilot) to view in real time whether the child is touching the folding child safety seat, and further avoiding the child from being pinched through the intelligent plus manual method.

[0174] For example, a family of three, namely A, B, and their child C, are in a car. A is the driver, B and C are sitting in the back row. C is not sitting in a child safety seat but playing on the back seat. B is swiping their phone. At this time, A controls the retraction of the child safety seat through voice. C becomes curious and puts their hand on the rotating side wing. A sees this through the seat image displayed on the in-vehicle display and quickly reminds B to stop C's behavior.

[0175] In some embodiments, an emergency stop component (button or key) is provided beside the in-vehicle display or on the surface of the child safety seat to emergently stop the retraction process of the child safety seat.

[0176] In some alternative embodiments, the child safety seat further has a restraint state;

[0177] In the deployed state, the angle between the side wing and the backrest is a first angle; in the restraint state, the angle between the side wing and the backrest is a second angle; in the retracted state, the side wing adheres to the backrest; the first angle is greater than the second angle, and the second angle is an acute angle;

[0178] The method further includes:

[0179] Obtaining the vehicle speed of the car;

[0180] When it is detected that the vehicle speed of the car is not less than a preset speed threshold, controlling the side wing driving mechanism to drive the side wing to rotate and controlling the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from the deployed state to the restraint state.

[0181] Thus, in addition to the deployed state and the retracted state, the child safety seat further has a restraint state. In the restraint state, the second angle between the side wing and the backrest is smaller than the first angle in the deployed state; in the deployed state, the first angle can be 90°, 120°, or 150°. In this way, the side wing opens outward, facilitating the child to enter the child safety seat. When the car starts to drive at a high speed, the side wing driving mechanism can be used to drive the side wing to retract inward, making the angle between the side wing and the backrest an acute angle. In this way, the side wing can wrap around the child, playing a role in buffering and protection to prevent the child from being injured in a sudden accident.

[0182] The embodiments of the present application do not limit the preset speed threshold, and the preset speed threshold is, for example, 5 km / h, 10 km / h, or 20 km / h. The first included angle can adopt the default recommended value, or the user can set it by himself within a given range. Similarly, the second included angle can adopt the default recommended value, or the user can set it by himself within a given range. The default recommended value of the first included angle is greater than the default recommended value of the second included angle. The given range of the first included angle is, for example, 90° to 150°, and the given range of the second included angle is, for example, 50° to 80°.

[0183] In some alternative embodiments, the method further includes:

[0184] When it is detected that the vehicle speed of the car is less than the preset speed threshold, control the flank driving mechanism to drive the flank to rotate, and control the base driving mechanism to drive the base to rotate, so that the child safety seat is transformed from the restrained state to the unfolded state.

[0185] Thus, when the car starts to decelerate and prepare to stop, the flank driving mechanism can be used to drive the flank to automatically open outwards, facilitating the vehicle occupants to take the child out of the child safety seat.

[0186] In some alternative embodiments, the process of determining the second included angle includes:

[0187] Use the second camera to capture the rear seat area to obtain a rear seat image, and the display content of the rear seat image includes the child;

[0188] Based on the rear seat image, obtain the body size information of the child;

[0189] Based on the body size information, determine the second included angle.

[0190] Thus, the second included angle can be determined according to the body size information of the child. When the rear seat is opened and the child gets into the car, the second camera can be used to capture the rear seat area to obtain a rear seat image, and the body size information of the child can be obtained from the rear seat image. For a child with a relatively plump body, the second included angle can be set larger, such as 70° or 80°. For a child with a relatively thin body, the second included angle can be set smaller, such as 50° or 60°. This way of adaptively adjusting the included angle between the flank and the backrest according to the child's body size not only avoids the fatter child being too tightly restrained by the flank, but also avoids the thinner child not being restrained by the flank, greatly improving the user experience.

[0191] In some embodiments, the body size information may include the chest circumference. The second camera may be a rotatable camera, and its shooting angle may include facing the side of the rear seat occupant. The chest circumference of the child can be obtained from the rear seat image captured by the second camera, and the second included angle is determined according to the chest circumference.

[0192] The first camera, the second camera, and the third camera in the embodiments of the present application are all disposed inside the vehicle, and the first camera, the second camera, and the third camera may be the same camera.

[0193] See Figure 2 , Figure 2 which is a schematic structural diagram of a child safety seat provided by the embodiments of the present application.

[0194] The child safety seat includes:

[0195] A base 10;

[0196] A backrest 20, the lower part of the backrest 20 is rotatably connected to the rear part of the base 10 through a first pivot;

[0197] Side wings 30, the side wings 30 are rotatably connected to the side of the backrest 20 through a second pivot;

[0198] A connection component, the connection component is disposed on the lower part of the backrest 20 and / or the rear part of the base 10 for connecting to an automotive seat;

[0199] The child safety seat has a deployed state and a retracted state. When the child safety seat is converted from the deployed state to the retracted state, the base 10 rotates relative to the backrest 20 around the first pivot, so that the upper surface of the base 10 approaches the front surface of the backrest 20, and the side wings 30 rotate relative to the backrest 20 around the second pivot, so that the side wings 30 approach the front surface of the backrest 20;

[0200] The child safety seat further includes a memory and a processor. The memory stores a computer program, and the processor is configured to perform the following steps when executing the computer program:

[0201] When a preset retraction event occurs, detect whether the child safety seat is being used by a child;

[0202] If the child safety seat is not being used by the child, control the side wing 30 driving mechanism to drive the side wings 30 to rotate, and control the base 10 driving mechanism to drive the base 10 to rotate, so that the child safety seat is converted from the deployed state to the retracted state;

[0203] If the child safety seat is being used by the child, use the in-vehicle display screen to display a prompt message to prompt the user that the child is using the child safety seat;

[0204] The preset retraction event includes at least one of the following:

[0205] Within a preset time duration, the maximum value of the pressure received by the base 10 is not greater than a preset pressure threshold, where the pressure received by the base 10 is sensed by a pressure sensor disposed on the upper surface of the base 10;

[0206] The interaction component receives a folding trigger operation, and the interaction component includes at least one of the following: a voice recognition component, an image recognition component, an infrared receiving component, a button, and a touch screen.

[0207] In some embodiments, the child safety seat may further include a headrest 40 and a head pillow. The headrest 40 is disposed on the backrest 20, and the head pillow is connected to the headrest 40 by a magic tape. The height of the headrest 40 is adjustable.

[0208] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to detect whether the child safety seat is used by a child:

[0209] Use a first camera to capture the child safety seat to obtain a seat image;

[0210] Based on the seat image, detect whether the child safety seat is used by a child.

[0211] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to detect whether the child safety seat is used by a child:

[0212] Based on the seat image, detect whether the child's body part is within the rotation range of the side wing 30 and detect whether the child's body part is within the rotation range of the base 10;

[0213] When the child's body part is within the rotation range of the side wing 30 or the rotation range of the base 10, determine that the child safety seat is used by the child;

[0214] When the child's body part is not within the rotation range of the side wing 30 and not within the rotation range of the base 10, determine that the child safety seat is not used by the child.

[0215] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to detect whether the child safety seat is used by a child:

[0216] Input the seat image into a seat detection model to obtain a seat detection result, where the seat detection result is used to indicate whether the child safety seat is used by a child;

[0217] Wherein, the training process of the seat detection model includes:

[0218] Obtain a training set, where the training set includes a plurality of training data, and each piece of training data includes a sample seat image and annotation data corresponding to the seat detection result of the sample seat image;

[0219] For each piece of training data in the training set, perform the following processing:

[0220] Input the sample seat image in the training data into a preset deep learning model to obtain prediction data corresponding to the seat detection result of the sample seat image;

[0221] Update the model parameters of the deep learning model based on the prediction data and annotation data corresponding to the seat detection result of the sample seat image;

[0222] Detect whether a preset training end condition is met; if so, use the trained deep learning model as the seat detection model; if not, continue to train the deep learning model with the next piece of training data.

[0223] In some alternative embodiments, the processor is further configured to implement the following steps when executing the computer program:

[0224] During the process of the child safety seat transitioning from the deployed state to the folded state, display the seat image captured in real time by the first camera on the in-vehicle display screen.

[0225] In some alternative embodiments, the child safety seat further has a restraint state;

[0226] In the deployed state, the included angle between the side wing 30 and the backrest 20 is a first included angle; in the restraint state, the included angle between the side wing 30 and the backrest 20 is a second included angle; in the folded state, the side wing 30 adheres to the backrest 20; the first included angle is greater than the second included angle, and the second included angle is an acute angle;

[0227] The processor is further configured to implement the following steps when executing the computer program:

[0228] Obtain the vehicle speed of the vehicle;

[0229] When it is detected that the vehicle speed of the vehicle is not less than a preset speed threshold, control the side wing 30 driving mechanism to drive the side wing 30 to rotate, and control the base 10 driving mechanism to drive the base 10 to rotate, so that the child safety seat transitions from the deployed state to the restraint state.

[0230] In some alternative embodiments, the processor is further configured to implement the following steps when executing the computer program:

[0231] When the vehicle speed of the car is detected to be less than the preset speed threshold, control the wing 30 driving mechanism to drive the wing 30 to rotate, and control the base 10 driving mechanism to drive the base 10 to rotate, so that the child safety seat is transformed from the restraint state to the deployed state.

[0232] In some alternative embodiments, when the processor is configured to execute the computer program, the following method is adopted to determine the second included angle:

[0233] Use the second camera to capture the rear seat area to obtain a rear seat image, and the display content of the rear seat image includes the child;

[0234] Based on the rear seat image, obtain the body size information of the child;

[0235] Based on the body size information, determine the second included angle.

[0236] See Figure 3 , Figure 3 is a structural block diagram of a child safety seat provided by an embodiment of the present application.

[0237] The child safety seat includes at least one memory 210 and at least one processor 220. The child safety seat may further include a bus 230 connecting different platform systems.

[0238] The memory 210 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 211 and / or a cache memory 212, and may further include a read-only memory (ROM) 213.

[0239] Among them, the memory 210 also stores a computer program, and the computer program can be executed by the processor 220, so that the processor 220 realizes the steps of any of the above methods. The specific implementation manner is the same as the implementation manner and the achieved technical effects recorded in the above method embodiments, and some contents will not be repeated.

[0240] The memory 210 may further include a utility tool 214 having at least one program module 215. Such program modules 215 include but are not limited to: an operating system, one or more application programs, other program modules, and program data. The implementation of a network environment may be included in each or some combination of these examples.

[0241] Correspondingly, the processor 220 can execute the above computer program and can also execute the utility tool 214.

[0242] The processor 220 may employ one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0243] The bus 230 may be one or more representing several types of bus architectures, including a memory bus or a memory autonomous vehicle, a peripheral bus, a graphics acceleration port, a processor, or a local area bus of any bus architecture using multiple bus architectures.

[0244] The child safety seat may also communicate with one or more external devices 240 such as a keyboard, a pointing device, a Bluetooth device, etc., and may also communicate with one or more devices capable of interacting with the child safety seat, and / or communicate with any device (such as a router, a modem, etc.) that enables the child safety seat to communicate with one or more other computing devices. Such communication may be carried out through the input / output interface 250. Moreover, the child safety seat may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 260. The network adapter 260 may communicate with other modules of the child safety seat through the bus 230. It should be understood that although not shown in the figure, other hardware and / or software modules may be used in combination with the child safety seat, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0245] The embodiments of the present application also provide a computer-readable storage medium for storing a computer program, and when the computer program is executed, the steps of any of the above methods are implemented. The specific implementation manner is the same as the implementation manner and the achieved technical effects recorded in the above method embodiments, and some content will not be elaborated again.

[0246] See Figure 4 , Figure 4 which shows a schematic structural diagram of a program product provided by the embodiments of the present application.

[0247] The program product is used to implement any of the above methods. The program product may be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this application, the readable storage medium may be any tangible medium that contains or stores a program, and the program can be used by or in conjunction with an instruction execution system, apparatus, or device. The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0248] The computer-readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, in which the readable program code is carried. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium may also be any readable medium that can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above. The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as C language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0249] This application is described from the perspectives of purpose of use, efficacy, progressiveness, and novelty, and already meets the functional enhancement and usage requirements emphasized by the Patent Law. The above description and accompanying drawings of this application are only preferred embodiments of this application and do not limit this application thereto. Therefore, all those that are similar or identical to the structure, device, features, etc. of this application, that is, all equivalent substitutions or modifications made according to the scope of the patent application of this application, shall fall within the scope of protection of the patent application of this application.

Claims

1. A control method for a child safety seat, characterized in that, The child safety seat includes: A base; A backrest, the lower part of the backrest is rotatably connected to the rear part of the base through a first pivot; Side wings, the side wings are rotatably connected to the side of the backrest through a second pivot; A connection component, the connection component is arranged on the lower part of the backrest and / or the rear part of the base for connecting with a vehicle seat; The child safety seat has a deployed state and a retracted state. When the child safety seat is converted from the deployed state to the retracted state, the base rotates relative to the backrest around the first pivot, so that the upper surface of the base approaches the front surface of the backrest, and the side wings rotate relative to the backrest around the second pivot, so that the side wings approach the front surface of the backrest; The method includes: When a preset retraction event occurs, detecting whether the child safety seat is being used by a child; If the child safety seat is not being used by the child, controlling a side wing driving mechanism to drive the side wings to rotate, and controlling a base driving mechanism to drive the base to rotate, so as to convert the child safety seat from the deployed state to the retracted state; If the child safety seat is being used by the child, using an in-vehicle display screen to display a prompt message to prompt the user that the child is using the child safety seat; The preset retraction event includes at least one of the following: Within a preset time period, the maximum value of the pressure received by the base is not greater than a preset pressure threshold, where the pressure received by the base is sensed by a pressure sensor arranged on the upper surface of the base; An interaction component receives a retraction trigger operation, and the interaction component includes at least one of the following: a voice recognition component, an image recognition component, an infrared receiving component, a button, and a touch screen.

2. The control method of the child safety seat according to claim 1, wherein The detecting whether the child safety seat is being used by a child includes: Using a first camera to capture the child safety seat to obtain a seat image; Based on the seat image, detecting whether the child safety seat is being used by a child.

3. The control method of the child safety seat according to claim 2, wherein The detecting whether the child safety seat is being used by a child based on the seat image includes: Based on the seat image, detecting whether the body part of the child is within the rotation range of the side wings and detecting whether the body part of the child is within the rotation range of the base; When the body part of the child is within the rotation range of the side wings or the rotation range of the base, determining that the child safety seat is being used by the child; When the body part of the child is not within the rotation range of the side wings and not within the rotation range of the base, determining that the child safety seat is not being used by the child.

4. The control method of the child safety seat according to claim 2, characterized in that, The detecting whether the child safety seat is being used by a child includes: Inputting the seat image into a seat detection model to obtain a seat detection result, and the seat detection result is used to indicate whether the child safety seat is being used by a child; Wherein, the training process of the seat detection model includes: Obtaining a training set, the training set includes a plurality of training data, and each training data includes a sample seat image and annotation data of the seat detection result corresponding to the sample seat image; For each training data in the training set, performing the following processing: Input the sample seat images in the training data into a preset deep learning model to obtain prediction data of the seat detection results corresponding to the sample seat images; Update the model parameters of the deep learning model based on the prediction data and annotation data of the seat detection results corresponding to the sample seat images; Detect whether a preset training end condition is satisfied; if so, use the trained deep learning model as the seat detection model; if not, continue to train the deep learning model using the next piece of training data.

5. The control method of the child safety seat according to claim 2, wherein, The method further includes: During the process of the child safety seat transitioning from the deployed state to the retracted state, use the in-vehicle display screen to display the seat image captured in real time by the first camera.

6. The control method of the child safety seat according to claim 1, characterized in that, The child safety seat further has a restraint state; In the deployed state, the included angle between the side wings and the backrest is a first included angle; in the restraint state, the included angle between the side wings and the backrest is a second included angle; in the retracted state, the side wings are attached to the backrest; The first included angle is greater than the second included angle, and the second included angle is an acute angle; The method further includes: Obtain the vehicle speed of the vehicle; When it is detected that the vehicle speed of the vehicle is not less than a preset speed threshold, control the side wing drive mechanism to drive the side wings to rotate, and control the base drive mechanism to drive the base to rotate, so that the child safety seat transitions from the deployed state to the restraint state.

7. The control method of the child safety seat according to claim 6, characterized in that The method further includes: When it is detected that the vehicle speed of the vehicle is less than the preset speed threshold, control the side wing drive mechanism to drive the side wings to rotate, and control the base drive mechanism to drive the base to rotate, so that the child safety seat transitions from the restraint state to the deployed state.

8. The control method of the child safety seat according to claim 6, wherein, The process of determining the second included angle includes: Use a second camera to capture the rear seat area to obtain a rear seat image, and the display content of the rear seat image includes the child; Based on the rear seat image, obtain the body size information of the child; Based on the body size information, determine the second included angle.

9. A child safety seat, characterized in that, The child safety seat includes: A base; A backrest, the lower part of the backrest is rotatably connected to the rear part of the base through a first pivot; Side wings, the side wings are rotatably connected to the side of the backrest through a second pivot; A connection component, the connection component is arranged on the lower part of the backrest and / or the rear part of the base for connecting with the vehicle seat; The child safety seat has a deployed state and a retracted state. When the child safety seat transitions from the deployed state to the retracted state, the base rotates relative to the backrest around the first pivot, so that the upper surface of the base approaches the front surface of the backrest, and the side wings rotate relative to the backrest around the second pivot, so that the side wings approach the front surface of the backrest; The child safety seat further includes a memory and a processor, the memory stores a computer program, and the processor is configured to implement the following steps when executing the computer program: When a preset retraction event occurs, detect whether the child safety seat is being used by a child; If the child safety seat is not used by the child, control the wing driving mechanism to drive the wings to rotate, and control the base driving mechanism to drive the base to rotate, so that the child safety seat is converted from the deployed state to the retracted state; If the child safety seat is used by the child, use the in-vehicle display screen to display a prompt message to prompt the user that the child is using the child safety seat; The preset retraction event includes at least one of the following: Within a preset time period, the maximum value of the pressure received by the base is not greater than a preset pressure threshold, where the pressure received by the base is sensed by a pressure sensor provided on the upper surface of the base; The interaction component receives a retraction trigger operation, and the interaction component includes at least one of the following: a voice recognition component, an image recognition component, an infrared receiving component, a button, and a touch screen.

10. A computer-readable storage medium storing a computer program, which when executed by a processor implements the steps of the control method of the child safety seat according to any one of claims 1-8.

Images