A vehicle window anti-pinch control method and system fusing visual perception

Abstract

The application belongs to the technical field of vehicle safety, and particularly relates to a vehicle window anti-pinch control method and system fusing visual perception, the vehicle window anti-pinch control method comprising: a visual perception unit collecting in-vehicle image information and out-of-vehicle image information, and sending the in-vehicle image information and the out-of-vehicle image information to a cabin domain controller in real time; after the cabin domain controller receives window lifting information of any window, the cabin domain controller identifies an obstacle at the corresponding window according to the received in-vehicle image information and out-of-vehicle image information, and sends no-obstacle information or obstacle type information of the corresponding window to a body domain controller according to the identification result; the body domain controller selects a working mode of a glass lifting execution mechanism of the corresponding window according to the received no-obstacle information or obstacle type information of the corresponding window and collected window position information of the corresponding window, and controls the glass lifting execution mechanism of the corresponding window to execute the function of the selected working mode.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of vehicle safety, and particularly relates to a vehicle window anti-pinch control method and system fusing visual perception. BACKGROUND

[0002] The anti-pinch of the vehicle window is an important part of the humanization of the vehicle, and its main function is to recognize that the vehicle window is in a pinched state when the vehicle window is pinched to an obstacle after being lifted, and to make the vehicle window retreat to release the pinched object, so as to prevent the motor from being burned due to long-time blockage and prevent the vehicle occupant from being pinched.

[0003] The existing anti-pinch control method of the vehicle window usually realizes the anti-pinch function of the vehicle window by detecting the motor current through a current sensor or detecting the motor speed through a Hall sensor. Specifically, when the electronic control unit drives the motor to rotate, the electronic control unit receives real-time current information from the current sensor or real-time speed information from the Hall sensor. When the current value in the real-time current information reaches the anti-pinch current value, the electronic control unit drives the motor to reverse, or when the speed of the motor in the real-time speed information changes, the electronic control unit controls the motor to stop rotating (so that the vehicle window stops rising) or to reverse (so that the vehicle window descends), thereby realizing the anti-pinch function of the vehicle window.

[0004] However, the existing anti-pinch control method of the vehicle window mostly only has the anti-pinch function within the specified height range (i.e., the distance between the top end of the vehicle window glass and the sealing strip above it is within the preset distance range), so that when the distance between the top end of the vehicle window glass and the sealing strip above it is less than the minimum value of the preset distance range, the vehicle window usually loses the anti-pinch function, causing a high risk of pinching the limbs in the area above the vehicle window glass close to the sealing strip. In addition, since the anti-pinch force of the vehicle window in the existing anti-pinch control method of the vehicle window is usually a fixed value within the specification range, it is easy to cause the limbs of children to be pinched due to the too large anti-pinch force, or the anti-pinch action of the vehicle window to be too sensitive due to the too small anti-pinch force, thereby affecting the use experience. SUMMARY

[0005] In order to solve the defects of the prior art, the present application provides a vehicle window anti-pinch control method and system that fuses visual perception to solve the technical problems that the existing vehicle window anti-pinch control method mostly only has an anti-pinch function within a specified height range (i.e., when the distance between the top end of the window glass and the sealing strip above it is within a preset distance range), so that when the distance between the top end of the window glass and the sealing strip above it is less than the minimum value of the preset distance range, the window usually loses the anti-pinch function, causing a greater risk of pinching limbs above the window glass near the sealing strip area; and because the anti-pinch force of the window in the existing vehicle window anti-pinch control method is usually calibrated to a fixed value within the specification range, it is easy to cause the technical problems that the anti-pinch force is too large and easily pinches the limbs of children, or the anti-pinch force is too small and causes the window anti-pinch action to be too sensitive and affect the use experience.

[0006] In order to achieve the above purpose, the technical scheme of the present application is as follows:

[0007] A vehicle window anti-pinch control method that fuses visual perception, comprising:

[0008] The visual perception unit collects vehicle interior image information and vehicle exterior image information, and sends the vehicle interior image information and the vehicle exterior image information to the cabin domain controller in real time;

[0009] After the cabin domain controller receives the window lifting information of any window, the cabin domain controller identifies the obstacles at the corresponding window according to the received vehicle interior image information and vehicle exterior image information, and sends the obstacle-free information or obstacle type information of the corresponding window to the body domain controller according to the identification result;

[0010] The body domain controller selects the working mode of the glass lifting execution mechanism of the corresponding window according to the received obstacle-free information or obstacle type information of the corresponding window and the collected window position information of the corresponding window, and controls the glass lifting execution mechanism of the corresponding window to execute the function of the selected working mode.

[0011] The visual perception unit is used to collect the image information inside and outside the vehicle, and then the cabin domain controller identifies the obstacles at the corresponding window according to the received image information inside and outside the vehicle, to identify whether there are obstacles blocking the window glass from rising at the corresponding window, and further identify the type of the obstacles when there are obstacles; and then the identification result is sent to the vehicle body domain controller, so that the vehicle body domain controller can select the working mode of the glass lifting execution mechanism of the corresponding window according to the identification result and the collected window position information of the corresponding window, and control the glass lifting execution mechanism of the corresponding window to execute the selected working mode function; so that the automobile window anti-pinch control method provided by the application can comprehensively judge the working mode of the glass lifting execution mechanism of the corresponding window according to the different heights of the window (i.e. the different distances between the top end of the window glass and the sealing strip above it) and the different types of obstacles blocking the window from rising, execute different anti-pinch strategies, so that the window will not lose the anti-pinch function at any height, and the anti-pinch strategy can be adjusted (the size of the anti-pinch force is adjusted, the rising speed of the window glass is adjusted) according to the type of the obstacle, which can effectively avoid the technical problems that children's limbs are easily pinched due to the too large anti-pinch force, or the window anti-pinch action is too sensitive due to the too small anti-pinch force, affecting the use experience.

[0012] Further, after the cabin domain controller receives the window rising information of any window, the cabin domain controller preliminarily identifies the obstacles at the corresponding window according to the received image information inside and outside the vehicle;

[0013] If the preliminary identification result is that there is no obstacle at the corresponding window, the cabin domain controller sends the no-obstacle information of the corresponding window to the vehicle body domain controller; if the preliminary identification result is that there is an obstacle at the corresponding window, the cabin domain controller identifies the type of the obstacle at the corresponding window according to the received image information inside and outside the vehicle, and sends the obstacle type information of the corresponding window to the vehicle body domain controller according to the type identification result;

[0014] The obstacle type information includes the type identification result, and the identification result includes one of adult limbs, child limbs and other obstacles.

[0015] The cabin domain controller identifies the type of the obstacle as an adult limb, a child limb or other obstacles, so that the vehicle body domain controller can determine the working mode of the glass lifting execution mechanism of the corresponding window according to the identification result and the different heights of the window (i.e. the different distances between the top of the window glass and the sealing strip above it), and execute the corresponding anti-pinch strategy when the obstacle is an adult limb, a child limb or other obstacles.

[0016] Further, the method for the cabin domain controller to identify the type of the obstacle at the corresponding window according to the received vehicle interior image information and vehicle exterior image information comprises the following steps:

[0017] Step 1: The cabin domain controller determines whether the obstacle at the corresponding window is a human limb according to the received vehicle interior image information and vehicle exterior image information. If yes, go to Step 2; otherwise, determine that the type identification result is other obstacles.

[0018] Step 2: The cabin domain controller determines whether the obstacle at the corresponding window is a child limb according to the received vehicle interior image information and vehicle exterior image information. If yes, determine that the type identification result is a child limb; otherwise, determine that the type identification result is an adult limb.

[0019] Further, the working mode of the glass lifting execution mechanism includes a low-sensitivity anti-pinch mode, a high-sensitivity anti-pinch mode, an emergency mode and an anti-stall mode.

[0020] The method for the vehicle body domain controller to control any glass lifting execution mechanism to execute the low-sensitivity anti-pinch mode comprises: the vehicle body domain controller controls the glass lifting execution mechanism of the window to drive the window glass of the window to rise at a first speed, and at the same time, the vehicle body domain controller sets the anti-pinch force of the glass lifting execution mechanism of the window at a first anti-pinch force value.

[0021] The method for the vehicle body domain controller to control any glass lifting execution mechanism to execute the high-sensitivity anti-pinch mode comprises: the vehicle body domain controller controls the glass lifting execution mechanism of the window to drive the window glass of the window to rise at a second speed, and at the same time, the vehicle body domain controller sets the anti-pinch force of the glass lifting execution mechanism of the window at a second anti-pinch force value.

[0022] The method for the vehicle body domain controller to control any glass lifting execution mechanism to execute the emergency mode comprises: the vehicle body domain controller controls the glass lifting execution mechanism of the window to drive the window glass of the window to descend.

[0023] The method for controlling any glass lifting actuator to execute the anti-stall mode by the body domain controller comprises: the body domain controller controls the glass lifting actuator of the vehicle window to drive the vehicle window glass of the vehicle window to rise at a third speed, and simultaneously, the body domain controller detects the running current value of the motor of the glass lifting actuator of the vehicle window, and if the running current value is detected to be not less than the stall current value, the body domain controller controls the glass lifting actuator of the vehicle window to drive the vehicle window glass of the vehicle window to descend.

[0024] The first speed is greater than the second speed, and the first anti-pinch force value is greater than the second anti-pinch force value.

[0025] Further, the method for selecting the working mode of the glass lifting actuator of the corresponding vehicle window comprises:

[0026] If the body domain controller receives the obstacle type information of the vehicle window, and the collected vehicle window position information of the vehicle window is that the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above the vehicle window glass is greater than the maximum value of the preset distance range, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the anti-stall mode.

[0027] If the body domain controller receives the obstacle type information of the vehicle window, and the collected vehicle window position information of the vehicle window is that the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above the vehicle window glass is within the preset distance range, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the low-sensitivity anti-pinch mode or the high-sensitivity anti-pinch mode.

[0028] If the body domain controller receives the obstacle type information of the vehicle window, and the collected vehicle window position information of the vehicle window is that the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above the vehicle window glass is less than the minimum value of the preset distance range, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the low-sensitivity anti-pinch mode, the high-sensitivity anti-pinch mode or the emergency mode.

[0029] Further, when the body domain controller collects the vehicle window position information of a certain vehicle window, the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above the vehicle window glass is within the preset distance range.

[0030] If the body domain controller receives the result of type identification in the obstacle type information of the vehicle window as other obstacles, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the low-sensitivity anti-pinch mode.

[0031] If the type recognition result in the obstacle type information of the window received by the vehicle body domain controller is an adult limb or a child limb, the vehicle body domain controller selects the operation mode of the glass lifting actuator of the window as the high-sensitivity anti-pinch mode.

[0032] Further, when the window position information of a certain window collected by the vehicle body domain controller is that the distance between the top of the window glass of the window and the window sealing strip above it is less than the minimum value of the preset distance range;

[0033] If the type recognition result in the obstacle type information of the window received by the vehicle body domain controller is other obstacles, the vehicle body domain controller selects the operation mode of the glass lifting actuator of the window as the low-sensitivity anti-pinch mode.

[0034] If the type recognition result in the obstacle type information of the window received by the vehicle body domain controller is an adult limb, the vehicle body domain controller selects the operation mode of the glass lifting actuator of the window as the high-sensitivity anti-pinch mode.

[0035] If the type recognition result in the obstacle type information of the window received by the vehicle body domain controller is a child limb, the vehicle body domain controller selects the operation mode of the glass lifting actuator of the window as the emergency mode.

[0036] Preferably, the preset distance range is 4mm to 200mm.

[0037] Further, the method of selecting the operation mode of the glass lifting actuator corresponding to the window further comprises:

[0038] If the vehicle body domain controller receives the no-obstacle information of the window, the vehicle body domain controller selects the operation mode of the glass lifting actuator of the window as the anti-stall mode.

[0039] Based on the automobile window anti-pinch control method with fusion visual perception provided by the present application, the present application further provides an automobile window anti-pinch control system with fusion visual perception, which comprises a visual perception unit, a cabin domain controller, a vehicle body domain controller, and glass lifting actuators of each window of the automobile.

[0040] The visual perception unit is used to collect in-vehicle image information and out-of-vehicle image information, and send the in-vehicle image information and the out-of-vehicle image information to the cabin domain controller.

[0041] The cabin domain controller is used to identify the obstacles at each window according to the received in-vehicle image information and out-of-vehicle image information, and send the no-obstacle information or the obstacle type information of each window to the vehicle body domain controller according to the identification result.

[0042] The vehicle body domain controller is configured to collect the window position information of each window, and is further configured to select the working mode of the glass lifting actuator of any window according to the received unobstruction information or obstruction type information of the window and the collected window position information of the window, and control the glass lifting actuator of the window to perform the function of the selected working mode.

[0043] Further, the visual perception unit comprises a DMS camera, an OMS camera and a CMS camera.

[0044] The DMS camera and the OMS camera are both installed in a roof lamp assembly of the vehicle, and are collectively configured to collect the in-vehicle image information.

[0045] The CMS camera is installed in an electronic outside rearview mirror of the vehicle, and is configured to collect the out-of-vehicle image information. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. Obviously, the drawings described below are only some of the embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0047] Figure 1 is a flowchart of the vehicle window anti-pinch control method with fusion visual perception in embodiment 1;

[0048] Figure 2 is a system diagram of the vehicle window anti-pinch control system with fusion visual perception in embodiment 2. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all.

[0050] Embodiment 1:

[0051] As shown in Figure 1 Embodiment 1 provides a vehicle window anti-pinch control method with fusion visual perception, comprising:

[0052] The visual perception unit collects in-vehicle image information and out-of-vehicle image information, and sends the in-vehicle image information and the out-of-vehicle image information to the cabin domain controller in real time;

[0053] The cabin domain controller receives the window lifting information of any window, and identifies the obstacle at the corresponding window according to the received interior image information and exterior image information, and sends the obstacle-free information or obstacle type information of the corresponding window to the body domain controller according to the identification result;

[0054] The body domain controller selects the working mode of the glass lifting execution mechanism of the corresponding window according to the received obstacle-free information or obstacle type information of the corresponding window and the collected window position information of the corresponding window, and controls the glass lifting execution mechanism of the corresponding window to execute the function of the selected working mode.

[0055] The interior image information and exterior image information are collected by the visual perception unit, and the cabin domain controller identifies the obstacle at the corresponding window according to the received interior image information and exterior image information, to identify whether there is an obstacle blocking the window glass from rising at the corresponding window, and further identify the type of the obstacle when there is an obstacle. The identification result is sent to the body domain controller, so that the body domain controller can select the working mode of the glass lifting execution mechanism of the corresponding window according to the identification result and the collected window position information of the corresponding window, and control the glass lifting execution mechanism of the corresponding window to execute the function of the selected working mode. The fusion visual perception automobile window anti-pinch control method provided by the application can comprehensively judge the working mode of the glass lifting execution mechanism of the corresponding window according to the different heights of the window (i.e. the different distances between the top of the window glass and the sealing strip above it) and the different types of obstacles blocking the window from rising, execute different anti-pinch strategies, make the window not lose the anti-pinch function at any height, and adjust the anti-pinch strategy (adjust the size of the anti-pinch force, adjust the rising speed of the window glass) according to the type of the obstacle, which can effectively avoid the technical problems of easily pinching the child's limbs due to the too large anti-pinch force, or the too sensitive window anti-pinch action due to the too small anti-pinch force affecting the use experience.

[0056] The cabin domain controller receives the window lifting information of any window, and identifies the obstacle at the corresponding window according to the received interior image information and exterior image information, and sends the obstacle-free information or obstacle type information of the corresponding window to the body domain controller according to the identification result;

[0057] If the preliminary identification result is that there is no obstacle at the corresponding window, the cabin domain controller sends the obstacle-free information of the corresponding window to the body domain controller; if the preliminary identification result is that there is an obstacle at the corresponding window, the cabin domain controller identifies the type of the obstacle at the corresponding window according to the received interior image information and exterior image information, and sends the obstacle type information of the corresponding window to the body domain controller according to the type identification result;

[0058] The obstacle type information includes a type identification result, and the identification result includes one of an adult limb, a child limb, and other obstacles.

[0059] By identifying the type of the obstacle as an adult limb, a child limb, or other obstacles by the cabin domain controller, the vehicle body domain controller can determine the working mode of the glass lifting execution mechanism of the corresponding window according to the identification result and in combination with the different heights of the window (i.e., the different distances between the top end of the window glass and the sealing strip above the window glass), to execute the corresponding anti-pinch strategy when the obstacle is an adult limb, a child limb, or other obstacles.

[0060] Specifically, the method for identifying the type of the obstacle at the corresponding window by the cabin domain controller according to the received vehicle interior image information and vehicle exterior image information includes the following steps:

[0061] Step 1: The cabin domain controller determines whether the obstacle at the corresponding window is a human limb according to the received vehicle interior image information and vehicle exterior image information, and if so, proceeds to Step 2; otherwise, determines that the type identification result is other obstacles.

[0062] Step 2: The cabin domain controller determines whether the obstacle at the corresponding window is a child limb according to the received vehicle interior image information and vehicle exterior image information, and if so, determines that the type identification result is a child limb; otherwise, determines that the type identification result is an adult limb.

[0063] In this embodiment 1, the working modes of the glass lifting execution mechanism include a low-sensitivity anti-pinch mode, a high-sensitivity anti-pinch mode, an emergency mode, and an anti-stuck mode.

[0064] The method for the vehicle body domain controller to control any glass lifting execution mechanism to execute the low-sensitivity anti-pinch mode includes: the vehicle body domain controller controls the glass lifting execution mechanism of the window to drive the window glass of the window to rise at a first speed, and simultaneously, the vehicle body domain controller sets the anti-pinch force of the glass lifting execution mechanism of the window at a first anti-pinch force value.

[0065] The method for the vehicle body domain controller to control any glass lifting execution mechanism to execute the high-sensitivity anti-pinch mode includes: the vehicle body domain controller controls the glass lifting execution mechanism of the window to drive the window glass of the window to rise at a second speed, and simultaneously, the vehicle body domain controller sets the anti-pinch force of the glass lifting execution mechanism of the window at a second anti-pinch force value.

[0066] The method for the vehicle body domain controller to control any glass lifting execution mechanism to execute the emergency mode includes: the vehicle body domain controller controls the glass lifting execution mechanism of the window to drive the window glass of the window to descend.

[0067] The method for controlling any glass lifting actuator to execute the anti-stall mode by the body domain controller comprises: the body domain controller controls the glass lifting actuator of the vehicle window to drive the vehicle window glass of the vehicle window to rise at a third speed, and simultaneously, the body domain controller detects the running current value of the motor of the glass lifting actuator of the vehicle window, and if the detected running current value is not less than the stall current value, the body domain controller controls the glass lifting actuator of the vehicle window to drive the vehicle window glass of the vehicle window to descend.

[0068] Preferably, the first speed is greater than the second speed, and the first anti-pinch force value is greater than the second anti-pinch force value.

[0069] Specifically, the body domain controller sets the anti-pinch force of the glass lifting actuator of the vehicle window at the first anti-pinch force value, or sets the anti-pinch force of the glass lifting actuator of the vehicle window at the second anti-pinch force value, and then controls the glass lifting actuator of the vehicle window to drive the vehicle window glass of the vehicle window to descend or stop moving after the anti-pinch occurs.

[0070] Specifically, in the embodiment 1, the method for selecting the working mode of the glass lifting actuator corresponding to the vehicle window comprises:

[0071] If the body domain controller receives the obstacle type information of the vehicle window, and the collected vehicle window position information of the vehicle window is that the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above it is greater than the maximum value of the preset distance range, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the anti-stall mode.

[0072] If the body domain controller receives the obstacle type information of the vehicle window, and the collected vehicle window position information of the vehicle window is that the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above it is within the preset distance range, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the low-sensitivity anti-pinch mode or the high-sensitivity anti-pinch mode.

[0073] If the body domain controller receives the obstacle type information of the vehicle window, and the collected vehicle window position information of the vehicle window is that the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above it is less than the minimum value of the preset distance range, the body domain controller selects the working mode of the glass lifting actuator of the vehicle window as the low-sensitivity anti-pinch mode, the high-sensitivity anti-pinch mode or the emergency mode.

[0074] Preferably, in the embodiment 1, when the body domain controller collects the vehicle window position information of a certain vehicle window, the distance between the top of the vehicle window glass of the vehicle window and the vehicle window sealing strip above it is within the preset distance range.

[0075] If the vehicle body domain controller receives the obstacle type information of the window and the result of the type identification is other obstacles, the vehicle body domain controller selects the working mode of the window lifting actuator as low sensitivity anti-pinch mode.

[0076] If the vehicle body domain controller receives the obstacle type information for the window and the type identification result is an adult limb or a child limb, then the vehicle body domain controller selects the working mode of the window's glass lifting actuator as the high-sensitivity anti-pinch mode.

[0077] Preferably, in this embodiment 1, when the window position information of a certain window collected by the vehicle body domain controller is such that the distance between the top of the window glass and the window sealing strip above it is less than the minimum value of a preset distance range;

[0078] If the vehicle body domain controller receives the obstacle type information of the window and the result of the type identification is other obstacles, the vehicle body domain controller selects the working mode of the window lifting actuator as low sensitivity anti-pinch mode.

[0079] If the vehicle body domain controller receives the obstacle type information of the window and the result of the type identification is an adult limb, then the vehicle body domain controller selects the working mode of the window lifting actuator as the high-sensitivity anti-pinch mode.

[0080] If the vehicle body domain controller receives the obstacle type information for the window and the result of the type identification is a child's limb, then the vehicle body domain controller selects the working mode of the window's glass lifting actuator as emergency mode.

[0081] In this embodiment 1, preferably, the preset distance range is 4mm to 200mm.

[0082] In this embodiment 1, preferably, the method for selecting the working mode of the glass lifting actuator corresponding to the vehicle window further includes:

[0083] If the vehicle body domain controller receives accessibility information for the window, it selects the window lift actuator's operating mode as anti-blocking mode.

[0084] Example 2:

[0085] like Figure 2 As shown, based on the vehicle window anti-pinch control method with fusion visual perception provided in Embodiment 1, Embodiment 2 provides a vehicle window anti-pinch control system with fusion visual perception, including a visual perception unit, a cockpit domain controller, a body domain controller, and glass lifting actuators for each window of the vehicle.

[0086] The visual perception unit is configured to collect the in-vehicle image information and the out-of-vehicle image information, and transmit the in-vehicle image information and the out-of-vehicle image information to the cabin domain controller;

[0087] The cabin domain controller is configured to identify the obstacle at each vehicle window according to the received in-vehicle image information and out-of-vehicle image information, and transmit the obstacle-free information or the obstacle type information of each vehicle window to the body domain controller according to the identification result;

[0088] The body domain controller is configured to collect the vehicle window position information of each vehicle window, and further configured to select the working mode of the glass lifting execution mechanism of any vehicle window according to the received obstacle-free information or obstacle type information of the vehicle window and the collected vehicle window position information of the vehicle window, and control the glass lifting execution mechanism of the vehicle window to execute the selected working mode.

[0089] The visual perception unit and the cabin domain controller interact through a serial bus, the cabin domain controller and the body domain controller interact through a CAN bus, and the body domain controller and the glass lifting execution mechanism interact through a hardwire.

[0090] The main function of the visual perception unit is to collect the image information of the obstacle during the glass lifting process, the appearance and position of the person inside and outside the vehicle, and form the in-vehicle image information and the out-of-vehicle image information.

[0091] Preferably, as shown in the embodiment 2, the visual perception unit includes a DMS camera, an OMS camera and a CMS camera. Figure 2

[0092] The DMS camera and the OMS camera can adopt an integrated arrangement scheme or a distributed arrangement scheme.

[0093] The DMS camera and the OMS camera adopting the integrated arrangement scheme are both installed in the roof lamp assembly of the vehicle, and the DMS camera and the OMS camera are jointly configured to collect the in-vehicle image information.

[0094] The CMS camera is installed in the electronic outside rearview mirror of the vehicle, and the CMS camera is configured to collect the out-of-vehicle image information.

[0095] The automobile window anti-pinch control method and system provided by the present application have at least the following technical effects or advantages:

[0096] ​1. The method comprises the following steps: collecting the image information inside and outside the vehicle by using the visual perception unit; identifying the obstacle at the corresponding window by the cabin domain controller according to the received image information inside and outside the vehicle, to identify whether there is an obstacle blocking the window glass from rising at the corresponding window, and further identify the type of the obstacle when there is an obstacle; and sending the identification result to the body domain controller, so that the body domain controller can select the working mode of the glass lifting actuator of the corresponding window according to the identification result and the window position information of the corresponding window collected, and control the glass lifting actuator of the corresponding window to execute the selected working mode. The method provided by the application can comprehensively judge the working mode of the glass lifting actuator of the corresponding window according to the different heights of the window (i.e. the different distances between the top end of the window glass and the sealing strip above it) and the different types of the obstacle blocking the window from rising, execute different anti-pinch strategies, so that the window will not lose the anti-pinch function at any height, and the anti-pinch strategy can be adjusted (the size of the anti-pinch force is adjusted, and the rising speed of the window glass is adjusted) according to the type of the obstacle, which can effectively avoid the technical problems that the child's limbs are easily pinched due to the too large anti-pinch force, or the use experience is affected due to the too sensitive anti-pinch action caused by the too small anti-pinch force.

[0097] 2. The type of the obstacle identified by the cabin domain controller is adult limbs, child limbs or other obstacles, so that the body domain controller can comprehensively judge the working mode of the glass lifting actuator of the corresponding window according to the identification result and the different heights of the window (i.e. the different distances between the top end of the window glass and the sealing strip above it), and execute the corresponding anti-pinch strategy when the obstacle is adult limbs, child limbs or other obstacles.

[0098] The above is only a specific application example of the application, and does not constitute any limitation on the protection scope of the application. Any technical solution formed by equivalent transformation or equivalent replacement falls within the protection scope of the application.

Claims

1. A method for preventing pinching of a car window with fusion visual perception, characterized in that, The method comprises the following steps: The visual perception unit collects image information inside and outside the vehicle and sends the image information inside and outside the vehicle to the cabin domain controller in real time; After receiving the window lifting information of any window, the cabin domain controller identifies the obstacle at the corresponding window according to the received image information inside and outside the vehicle, and sends the obstacle-free information or the obstacle type information of the corresponding window to the body domain controller according to the identification result; The body domain controller selects the working mode of the glass lifting execution mechanism of the corresponding window according to the received obstacle-free information or obstacle type information of the corresponding window and the collected window position information of the corresponding window, and controls the glass lifting execution mechanism of the corresponding window to execute the function of the selected working mode; After receiving the window lifting information of any window, the cabin domain controller performs preliminary identification of the obstacle at the corresponding window according to the received image information inside and outside the vehicle; If the preliminary identification result is that there is no obstacle at the corresponding window, the cabin domain controller sends the obstacle-free information of the corresponding window to the body domain controller; If the preliminary identification result is that there is an obstacle at the corresponding window, the cabin domain controller performs type identification of the obstacle at the corresponding window according to the received image information inside and outside the vehicle, and sends the obstacle type information of the corresponding window to the body domain controller according to the type identification result; The obstacle type information includes the type identification result, and the identification result includes one of adult limbs, child limbs, and other obstacles; The working modes of the glass lifting execution mechanism include a low-sensitivity anti-pinch mode, a high-sensitivity anti-pinch mode, an emergency mode, and an anti-stuck mode; The method for selecting the working mode of the glass lifting execution mechanism of the corresponding window comprises: If the body domain controller receives the obstacle type information of the window and the collected window position information of the window is that the distance between the top of the window glass of the window and the window sealing strip above it is greater than the maximum value of the preset distance range, the body domain controller selects the working mode of the glass lifting execution mechanism of the window as the anti-stuck mode; If the body domain controller receives the obstacle type information of the window and the collected window position information of the window is that the distance between the top of the window glass of the window and the window sealing strip above it is within the preset distance range, the body domain controller selects the working mode of the glass lifting execution mechanism of the window as the low-sensitivity anti-pinch mode or the high-sensitivity anti-pinch mode; If the body domain controller receives the obstacle type information of the window and the collected window position information of the window is that the distance between the top of the window glass of the window and the window sealing strip above it is less than the minimum value of the preset distance range, the body domain controller selects the working mode of the glass lifting execution mechanism of the window as the low-sensitivity anti-pinch mode, the high-sensitivity anti-pinch mode, or the emergency mode.

2. The method of claim 1, wherein: The method for identifying the type of the obstacle at the corresponding window by the cabin domain controller according to the received vehicle interior image information and the vehicle exterior image information comprises the following steps: Step 1: The cabin domain controller determines whether the obstacle at the corresponding window is a human limb according to the received vehicle interior image information and the vehicle exterior image information, and if yes, enters step 2; otherwise, determines that the result of the type identification is other obstacles; Step 2: The cabin domain controller determines whether the obstacle at the corresponding window is a child limb according to the received vehicle interior image information and the vehicle exterior image information, and if yes, determines that the result of the type identification is a child limb; otherwise, determines that the result of the type identification is an adult limb.

3. The automobile window anti-pinch control method of the fusion visual perception according to claim 1, characterized in that: The method for the vehicle body domain controller to control any glass lifting actuator to execute the low-sensitivity anti-pinch mode comprises: the vehicle body domain controller controls the glass lifting actuator of the window to drive the window glass of the window to rise at a first speed, and simultaneously, the vehicle body domain controller sets the anti-pinch force of the glass lifting actuator of the window to a first anti-pinch force value; The method for the vehicle body domain controller to control any glass lifting actuator to execute the high-sensitivity anti-pinch mode comprises: the vehicle body domain controller controls the glass lifting actuator of the window to drive the window glass of the window to rise at a second speed, and simultaneously, the vehicle body domain controller sets the anti-pinch force of the glass lifting actuator of the window to a second anti-pinch force value; The method for the vehicle body domain controller to control any glass lifting actuator to execute the emergency mode comprises: the vehicle body domain controller controls the glass lifting actuator of the window to drive the window glass of the window to descend; The method for the vehicle body domain controller to control any glass lifting actuator to execute the anti-stall mode comprises: the vehicle body domain controller controls the glass lifting actuator of the window to drive the window glass of the window to rise at a third speed, and simultaneously, the vehicle body domain controller detects the running current value of the motor of the glass lifting actuator of the window, and if the detected running current value is not less than the stall current value, the vehicle body domain controller controls the glass lifting actuator of the window to drive the window glass of the window to descend; Wherein, the first speed is greater than the second speed, and the first anti-pinch force value is greater than the second anti-pinch force value.

4. The method of claim 3, wherein: When the vehicle body domain controller collects the window position information of a certain window, and the distance between the top of the window glass of the window and the window sealing strip above it is within the preset distance range; If the vehicle body domain controller receives the type identification result of the obstacle type information of the window as other obstacles, the vehicle body domain controller selects the working mode of the glass lifting actuator of the window as the low-sensitivity anti-pinch mode; If the vehicle body domain controller receives the type identification result of the obstacle type information of the window as the limb of a minor or a child, the vehicle body domain controller selects the working mode of the glass lifting actuator of the window as the high-sensitivity anti-pinch mode.

5. The method of claim 3, wherein: when the vehicle body domain controller collects the window position information of a certain window, and the distance between the top of the window glass of the window and the window sealing strip above it is less than the minimum value of the preset distance range; if the result of type identification in the obstacle type information of the window received by the vehicle body domain controller is other obstacles, the vehicle body domain controller selects the working mode of the glass lifting actuator of the window as the low-sensitivity anti-pinch mode; if the result of type identification in the obstacle type information of the window received by the vehicle body domain controller is adult limbs, the vehicle body domain controller selects the working mode of the glass lifting actuator of the window as the high-sensitivity anti-pinch mode; if the result of type identification in the obstacle type information of the window received by the vehicle body domain controller is child limbs, the vehicle body domain controller selects the working mode of the glass lifting actuator of the window as the emergency mode.

6. The method of claim 3, wherein: The method for selecting the working mode of the glass lifting actuator of a corresponding window further comprises: if the vehicle body domain controller receives the no-obstacle information of the window, the vehicle body domain controller selects the working mode of the glass lifting actuator of the window as the anti-stall mode.

7. A vehicle window anti-pinch control system that fuses visual perception, characterized by: The system comprises a visual perception unit, a cabin domain controller, a vehicle body domain controller, and glass lifting actuators of each window of the vehicle; The visual perception unit is configured to collect in-cabin image information and out-of-cabin image information, and send the in-cabin image information and the out-of-cabin image information to the cabin domain controller; The cabin domain controller is configured to identify obstacles at each window according to the received in-cabin image information and out-of-cabin image information, and send no-obstacle information or obstacle type information of each window to the vehicle body domain controller according to the identification result; The vehicle body domain controller is configured to collect window position information of each window, and further configured to select the working mode of the glass lifting actuator of any window according to the received no-obstacle information or obstacle type information of the window and the collected window position information of the window, and control the glass lifting actuator of the window to execute the selected working mode; The vehicle window anti-pinch control system with fused visual perception is configured to perform the steps in the vehicle window anti-pinch control method with fused visual perception according to any one of claims 1-6.

8. The fusion visual perception based car window anti-pinch control system of claim 7, wherein: The visual perception unit comprises a DMS camera, an OMS camera, and a CMS camera; The DMS camera and the OMS camera are both installed in the roof lamp assembly of the vehicle, and are collectively configured to collect the in-cabin image information; The CMS camera is installed in the electronic outside rearview mirror of the vehicle, and is configured to collect the out-of-cabin image information.

Images