CN111497773A - Vehicle safety protection system and method - Google Patents
Vehicle safety protection system and method Download PDFInfo
- Publication number
- CN111497773A CN111497773A CN201910093102.5A CN201910093102A CN111497773A CN 111497773 A CN111497773 A CN 111497773A CN 201910093102 A CN201910093102 A CN 201910093102A CN 111497773 A CN111497773 A CN 111497773A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- falling object
- safety protection
- falling
- parameters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/02—Occupant safety arrangements or fittings, e.g. crash pads
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60N—SEATS SPECIALLY ADAPTED FOR VEHICLES; VEHICLE PASSENGER ACCOMMODATION NOT OTHERWISE PROVIDED FOR
- B60N2/00—Seats specially adapted for vehicles; Arrangement or mounting of seats in vehicles
- B60N2/24—Seats specially adapted for vehicles; Arrangement or mounting of seats in vehicles for particular purposes or particular vehicles
- B60N2/42—Seats specially adapted for vehicles; Arrangement or mounting of seats in vehicles for particular purposes or particular vehicles the seat constructed to protect the occupant from the effect of abnormal g-forces, e.g. crash or safety seats
- B60N2/427—Seats or parts thereof displaced during a crash
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R16/00—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for
- B60R16/02—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements
- B60R16/023—Electric or fluid circuits specially adapted for vehicles and not otherwise provided for; Arrangement of elements of electric or fluid circuits specially adapted for vehicles and not otherwise provided for electric constitutive elements for transmission of signals between vehicle parts or subsystems
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/02—Occupant safety arrangements or fittings, e.g. crash pads
- B60R21/11—Overhead guards, e.g. against loads falling down
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Aviation & Aerospace Engineering (AREA)
- Transportation (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a vehicle safety protection system, which comprises a detection system and a control system, wherein parameters of a falling object are detected, the parameters comprise the speed and the position of the falling object, and the control system determines emergency measures of a vehicle based on the detected parameters.
Description
Technical Field
The invention belongs to the field of automobile safety, and particularly relates to a system and a method for carrying out safety protection on an object falling from high altitude by a vehicle.
Background
With the rapid development of cities and the increasing popularity of vehicles, the urban road environment becomes more and more complex. Such as building sites near roads, unmanned aerial vehicles without installed billboards, uncontrolled unmanned aerial vehicles, branches without trimming for a long time, hailstones and other bad weather and other complex external environmental factors, may often cause the condition that a falling object impacts the running vehicle to occur, and may cause vehicle damage and casualties. Furthermore, with the development of technology, traffic will not be limited to ground traffic in the future, but will exhibit a three-dimensional traffic situation, i.e. different vehicles may travel at different heights, so that the danger from the top of the vehicle will become more and more.
Most of the ADAS ("Advanced Driver assistance systems") installed in vehicles today are often focused on the situation around the level of the vehicle in terms of vehicle safety, in particular in front of and to the side of the driving level of the vehicle, with few solutions to the danger of the vehicle overhead.
Disclosure of Invention
Based on the above, the invention provides a vehicle safety protection system and a protection method capable of dealing with objects falling from high altitude.
One aspect of the present invention provides a vehicle safety protection system, which includes: a detection system that detects parameters of a falling object, the parameters including a speed and a position of the falling object; and a control system that determines an emergency action to be taken by the vehicle based on the parameter.
According to one embodiment of the invention, the detection system determines the movement profile of the falling object based on the parameter. In one embodiment, the motion state comprises acceleration and/or a motion trajectory of the falling object. The detection system further detects a 3D model parameter of the falling object and determines a 3D characteristic of the falling object based on the 3D model parameter. In one embodiment, the 3D features include volume and/or shape parameters of the falling object.
According to one embodiment of the invention, the detection system is a combination of one or more of the following: laser radar, meter wave radar, millimeter wave radar and image acquisition device.
According to one embodiment of the invention, the detection system is mounted to the roof of the vehicle and/or to the surroundings of the vehicle, the detection area covering the entire space within a certain distance to and above the vehicle.
According to one embodiment of the invention, the millimeter wave radar or the meter wave radar detects the speed and position parameters of the falling object, and the lidar detects the 3D model parameters of the falling object and/or the speed and position parameters of the falling object. The millimeter wave radar or the meter wave radar continuously detects falling objects, and the laser radar is started when the falling objects enter the measuring range of the laser radar.
According to one embodiment of the invention, a falling object is identified as a suspicious dangerous object when the falling acceleration of the falling object is below a predetermined threshold. In one embodiment, the predetermined threshold is between 0.6g and 0.8 g.
According to one embodiment of the invention, the falling object is identified as a discrete motion state when the deviation between the actual motion trajectory of the falling object and the theoretical value is above a predetermined threshold. In one embodiment, a fall is identified as a discrete state of motion when the deviation between the actual position and the theoretical position of the fall at a particular time or at a particular location is above a predetermined threshold.
In one embodiment, the motion trajectory function is calculated with a motion formula of the falling object in at least one of the three-dimensional directions as a fitting function, the motion formula being:
height direction: h (i) ═ a + b + ti 2;
horizontal X direction: x (i) ═ a1 ti + b1 ti 2;
horizontal Y direction: y (i) ═ a2 ti + b2 ti 2;
wherein h (i), x (i), y (i) and ti are moving distances of the falling object in three directions at time ti, and a, a1, a2, b1 and b2 are unknown numbers to be solved.
According to one embodiment of the invention, the control system calculates the motion trajectory function based on at least one of the following methods: least squares, multiple linear regression, or gradient descent.
According to one embodiment of the invention, the control system determines the expected fall location and the fall time of the fall based on the calculated motion trajectory function.
According to one embodiment of the invention, the parameters of the falling object detected by the detection system further comprise the distance and angle of the falling object with respect to the vehicle.
According to one embodiment of the invention, the control system determines the risk factor in dependence on: the volume of the falling object, the speed at which it falls to a certain height and the area and location of the covered vehicle, and the adoption of emergency measures is determined on the basis of the risk factor. Further, the control system further determines the danger coefficient according to the condition of people in the vehicle and/or the strength of the roof structure, and determines the adoption of emergency measures based on the danger coefficient.
According to one embodiment of the invention, the control system performs weighted calculation on the factors according to a preset weight value to determine the risk coefficient. Further, the control system determines the adoption of the emergency measure based on the comparison between the danger coefficient of the emergency measure and the danger coefficient of the falling object.
According to one embodiment of the invention, the self risk factor of the emergency measure is determined based on one or more of the following factors:
whether a pedestrian or other vehicle is in front of the vehicle;
whether other vehicles follow behind the vehicle;
whether there is a pedestrian or other vehicle in the yaw direction; and
whether the angle of deflection reaches a rollover limit.
According to one embodiment of the invention, the emergency action comprises taking one or more of the following: the method comprises the steps of opening a car roof air bag, emergency braking, accelerating passing, deflecting running and safety protection in a cabin.
According to one embodiment of the invention, the control system determines the emergency measure based on a BP neural network algorithm. In one embodiment, the model input nodes of the BP neural network algorithm comprise one or more of: the method comprises the following steps of (1) expected final speed of the falling object, expected falling position of the falling object, roof pressure bearing capacity, personnel condition in the vehicle, surrounding environment condition of the vehicle, covering area of the falling object, current vehicle speed, driving direction and theoretical braking distance.
According to one embodiment of the invention, the BP neural network algorithm adopts one or more of a Sigmoid function, a Re L U function and a Softmax function as an activation function of the hidden layer.
According to one embodiment of the invention, the BP neural network uses Softmax as an activation function, and the output node Y is calculated using the following formula:
Y＝Softmax(Relu(W1*X+B1)*W2+B2)，
where W1 and B1 are weight matrices and bias matrices between the input layer and the hidden layer, and W2 and B2 denote weight matrices and bias matrices between the hidden layer and the output layer.
According to one embodiment of the invention, the control system determines the emergency measure based on a local or cloud BP neural network algorithm.
According to one embodiment of the invention, the control system is based on an analysis of the results of the emergency action taken and is used to train and update the BP neural network algorithm.
According to one embodiment of the invention, the vehicle safety shield system further comprises a pressure sensor located on the roof to confirm the impact.
According to one embodiment of the invention, the vehicle safety shield system further comprises a first infrared beam-blocking sensor located at a first height within the vehicle. Further, the vehicle safety protection system further comprises a second infrared beam blocking sensor located at a second height in the vehicle, the second height being lower than the first height.
According to one embodiment of the invention, the control system calculates the current speed of the falling object based on the difference in height of the first infrared beam interruption sensor and the second infrared beam interruption sensor and the difference in sensing time.
According to one embodiment of the invention, the second infrared beam interruption sensors are respectively distributed at different seats so as to detect the falling position of the falling object.
In another aspect, the present invention provides a method for vehicle safety protection based on the vehicle safety protection system, including: detecting parameters of a falling object, the parameters including a speed and a position of the falling object; and determining emergency measures to be taken by the vehicle based on the parameters.
Yet another aspect of the invention provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of vehicle safety protection as described above.
In yet another aspect, the present invention provides an in-cabin safety protection system, comprising a hazard detection module including an infrared detection system including at least one infrared beam-blocking sensor for detecting a collision of a falling object with a vehicle.
In one embodiment, the at least one infrared beam-blocking sensor comprises a first infrared beam-blocking sensor and a second infrared beam-blocking sensor located at different heights.
In one embodiment, the at least one infrared beam-blocking sensor includes a transmitter and a receiver, which are respectively located at the front end and the rear end or the left end and the right end under the roof. In another embodiment, one of the transmitter or receiver is placed in a retractable headrest of the seat back and the other of the transmitter or receiver is placed in front or behind the horizontal. Further, the highest position of the infrared ray level in the retractable headrest is within a range of not more than 5cm near the top of the vehicle.
In one embodiment, the at least one infrared beam-blocking sensor is arranged to correspond to a position of each seat in the vehicle, respectively.
In one embodiment, the hazard detection module further comprises a pressure sensor located on the roof of the vehicle to detect a falling object.
In one embodiment, the hazard detection module further comprises an infrared ranging device to determine an actual impact location of the falling object.
In one embodiment, the in-cabin safety protection system includes a hazard handling module including an actuation device to drive a seat, a backrest, and/or a headrest. Further, the actuating device includes at least one of a motor, a cylinder, and a hydraulic cylinder.
Drawings
FIG. 1 is an architectural diagram of a vehicle safety shield system in accordance with one embodiment of the present invention;
FIG. 2 is a schematic view of a vehicle crash guard according to one embodiment of the present invention;
FIG. 3 is a schematic view of a coordinate system of a vehicle crash prevention object in accordance with one embodiment of the present invention;
FIG. 4 is a table of the ratings of various reference factors associated with a hazard in one embodiment of the present invention;
FIG. 5 is a reference table for calculating risk factors for a cockpit according to one embodiment of the present invention;
FIG. 6 is a reference table for calculating risk factors for non-cockpit according to one embodiment of the present invention;
FIG. 7 is a table of risk levels versus risk factors according to one embodiment of the present invention;
FIG. 8 is a schematic view of an in-vehicle infrared detection system in accordance with one embodiment of the present invention;
FIG. 9 is a flowchart of decision making based on the BP neural network algorithm according to an embodiment of the present invention;
FIG. 10 is a flowchart of decision making based on a BP neural network algorithm according to another embodiment of the present invention;
FIG. 11 is a model diagram of a BP neural network algorithm according to an embodiment of the present invention;
FIG. 12 is a flow chart of a vehicle safety protection method according to an embodiment of the invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather should be construed as broadly as the present invention is capable of modification in various respects, all without departing from the spirit and scope of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
FIG. 1 is an architectural diagram of a vehicle safety shield system 10 according to one embodiment of the present invention. Vehicle safety shield system 10 may be part of an ADAS system or may be a separate system in a vehicle and interact with an ADAS system. The vehicle safety protection system 10 includes a control system 11, and a detection system 12, an accident recording system 13, a vehicle positioning module 14, an infrared detection system 15, and an in-cabin safety protection system 16 in communication with the control system.
The detection system 12 is used to detect falling objects, such as falling rocks, building materials, waste, hail, and flying objects. The detection system 12 further comprises a first detection device 120 and a second detection device 121, each for detecting a different combination of parameters of the falling object. In one embodiment, the detection system is a radar system, which may be a subset of the ADAS system, or exist separately from the ADAS system. In one embodiment, the first detection device 120 is a meter-wave radar and the second detection device 121 is a lidar. The laser radar is suitable for middle and short distance detection, the detection precision is high, and the falling object can be imaged and 3D modeled through a DTM (digital terrestrial model). In other embodiments of the present invention, the first detection device 120 may be a millimeter wave radar, or both a millimeter wave radar and a millimeter wave radar may be used. The millimeter wave radar can be 77GHz Frequency Modulated Continuous Wave (FMCW) radar. The millimeter wave radar is suitable for remote detection and is less influenced by the environment. The lidar of the second detection device 121 may determine the type of the falling object and estimate the possible coverage area when the falling object falls based on the 3D features by modeling the falling object. The laser radar has different line number choices, such as 64 lines or 128 lines, and the detection result is more accurate when the number of laser lines is higher.
Referring to fig. 2, the meter-wave radar or the millimeter-wave radar is in a high-sensitivity working state to continuously monitor an object in the air. Specifically, the meter-wave radar or the millimeter-wave radar continuously detects parameters such as the position and the speed of the falling object. Once an object is airborne, it is quickly tracked, a first combination of parameters of the object, including but not limited to height of the falling object, distance and angle relative to the vehicle, acceleration, speed, volume, etc., is acquired and transmitted to the control system 11. The control system 11 determines the motion state of the object according to the following principle:
1) whether the falling acceleration of the object is close to the gravitational acceleration g (9.8 m/s)2) (ii) a Or
2) Whether the falling track of the falling object conforms to the motion state similar to a parabola or not is calculated according to the following formula:
X＝Vxt (equation 1)
H＝Vh*t+1/2*g*t2(formula 2)
Wherein X represents a moving distance in the horizontal direction, VxRepresenting the initial horizontal velocity, H representing the height of the fall, VhRepresenting the initial vertical velocity and t representing the movement time.
If it is determined that the falling acceleration of the falling object is greater than the predetermined threshold value or the actual motion trajectory of the falling object is close to the motion state described in item 2 above, it indicates that the falling object has a substantial weight, and the air resistance is negligible at this time, and the falling object may pose a direct threat to the safety of the vehicle and passengers, for example, the falling object is an object having a certain weight such as a flower disc, a building material, hail, etc. falling from high altitude. The falling object can be identified as a suspicious dangerous object. On the contrary, if the motion track of the falling object is discrete or the falling acceleration is smaller than the preset threshold, the falling object is light in weight, the safety threat to the vehicle and passengers is small, and at the moment, the vehicle does not need to take special precautionary measures and only needs to continuously monitor the falling object. The predetermined threshold for acceleration may be empirically obtained, on the one hand, as in one embodiment, the predetermined threshold ranges from 0.6g to 0.8 g. On the other hand, the predetermined threshold value may be continuously and automatically updated based on the threshold value data collected in the past of the vehicle, the result of the corresponding generation, and the like.
In addition, taking a free fall motion formula as an example of a loss function of a motion trail, the magnitude of the variance of a function calculation value and a real observation height value h of a falling object is calculated by the following formula to judge the discrete motion state of falling:
wherein v is the initial speed of the falling object, hiIs tiThe actual falling height of the falling object at all times.
The meter-wave radar or the millimeter-wave radar can measure and record the parameters of the falling object, such as height values, at a certain frequency, and calculate the variance value of a specific time point based on the collected data. If the variance value is too large, for example, greater than a predetermined threshold value, it indicates that the actual height data has a large deviation from the theoretical height value of the free-fall, that the falling object is in a discrete motion, that it is greatly affected by air resistance, or that the falling object is in a controlled state, such as a flying bird, an aircraft, or the like. If the variance value is small, the falling object tends to move in a free falling body or belongs to an air parabola, the air resistance is small, and the threat to the safety of the vehicle is large.
In other embodiments, the loss function may also be a parabolic motion formula or other motion formula. For example, fitting functions may be further established in the horizontal two-dimensional directions, a variance between the actual horizontal movement distance and the theoretical movement distance may be determined, and when the variance is greater than a predetermined threshold, the falling object may be determined to be in a discrete motion state.
In one embodiment, if the millimeter wave radar or meter wave radar identifies a suspicious dangerous object by the above method, and the suspicious object enters the detection range of the lidar, the lidar may be used to monitor a second combination of parameters of the suspicious object, including but not limited to 3D model parameters, speed, position, angle, distance, acceleration, volume, etc. of the object, alone or in cooperation with the millimeter wave radar or meter wave radar. Wherein the lidar comprises a three-dimensional laser scanner and the three-dimensional laser scanner is used to build a 3D model of the object. Referring to fig. 3, the control system 11 receives the data of the laser radar, and establishes a three-dimensional coordinate system based on the traveling route of the vehicle and the second parameter characteristic of the falling object, such as establishing the coordinate system with the current position of the vehicle as the origin. More specifically, in one embodiment, the vehicle left-right center symmetry plane may be the Y-plane, which is also perpendicular to the Y-axis in FIG. 3; a plane which is perpendicular to the Y plane and passes through the plane of the longitudinal beam of the car body in parallel is taken as a Z plane, and the Z plane is perpendicular to the Z axis in the figure 3 at the same time; and finally, taking a plane which is perpendicular to the Y plane and the Z plane and passes through the center point of the longitudinal direction of the vehicle as an X plane. The intersection of the three planes serves as the origin of the coordinate system. In other embodiments, the three planes and the position of the origin of the coordinate system may be adjusted according to actual conditions. In addition to the three-dimensional coordinates described above, a global geographic coordinate system, a local coordinate system (e.g., northeast coordinate system), and a vehicle coordinate system (e.g., right-front coordinate system) may be established. The establishment of these coordinate systems is well known in the art and will not be described in further detail herein. The vehicle may optionally be equipped with a GPS ("Global Positioning System") position sensor, and to improve the Positioning accuracy, an Inertial Measurement Unit (IMU) may be further equipped. The inertial measurement unit measures three-axis attitude angles (or angular rates) of an object using an accelerometer and a gyroscope, and thus can provide more accurate object position information in a complex environment than a GPS position sensor.
The control system 11 obtains the final floor coverage area of the falling object according to the parameters of the 3D modeling. Based on the current speed and the running track of the vehicle and the occupied area of each part of the vehicle, the control system 11 calculates whether the coordinate interval of the falling object intersects with or is very close to the coordinate interval of the vehicle at the time T when the falling object reaches the ground or the height of the roof through the established three-dimensional coordinate system. If there is a danger, the control system 11 further calculates the danger coefficient and starts the early warning mode. For example, if the distance between the coordinates of the falling object and the coordinates of the vehicle at time T is smaller than the minimum distance required for the driver to avoid the danger by manually controlling the vehicle, it is determined that there is a danger in this case. In one example, it is predicted by the foregoing method that at time T, the falling object is located 5 meters ahead of the vehicle in the traveling direction, and the vehicle speed at this time is 60 km/h, and the falling object cannot be avoided in a condition of ensuring safety by manually taking normal emergency measures, so that such a condition can be determined as a dangerous condition. In an embodiment of the present invention, a way of computing by the cloud server and the local computing device is adopted, so as to further improve the computing efficiency.
Due to environmental influences, such as weather conditions like strong wind and rainfall, or due to air resistance, the falling of the object may deviate from a standard movement pattern, i.e. the landing position of the object cannot be calculated directly by using the principle of free fall or parabola. The determination of the actual fall position can now be made based on relevant parameters of the falling object, such as height, velocity, acceleration, horizontal position, etc. In one embodiment of the invention, the millimeter wave radar or the meter wave radar continuously collects the position and speed information of the falling object based on a certain frequency. The control system 11 calculates the time t as a parameter by using an algorithm such as a least square method, a multiple linear regression method, or a gradient descent methodiThe height data at the time is fitted to a function (loss function) to predict the landing position point with the smallest error.
Assuming that the height prediction fit function for the object fall is consistent with:
h(i)＝a*ti+b*ti 2(formula 4)
Wherein a and b are unknown numbers to be solved.
In an embodiment where the least squares method calculates the landing position of a falling object, the above h (i) function is substituted into the following equation:
wherein h isiIs tiThe actual drop height at the moment.
When variance s2And when the minimum is needed, calculating partial derivatives of the a and the b, and enabling the two partial derivatives to be equal to zero, thereby calculating the a and the b and obtaining a correct motion track function. Based on the obtained motion trajectory function, the time required for the falling object to descend to a specific height, such as the ground or the roof, is calculated according to the height of the falling object. In the case of a falling object falling non-vertically, it is possible to follow the horizontal velocity andthe above calculated time directly calculates the moving distance of the falling object in the horizontal two-dimensional direction. Further, in the case of considering the air resistance, fitting functions are respectively established in the horizontal two-dimensional directions with reference to the above method:
X(i)＝a1*ti+b1*ti 2(formula 6)
Y(i)＝a2*ti+b2*ti 2(formula 7)
Wherein X (i) and Y (i) are at tiAt the moment the falling object moves a distance in the direction of the horizontal X, Y, a1、a2、b1And b2Respectively, unknown numbers to be solved.
And similarly, determining a fitting function and a motion track of the falling object in the two-dimensional direction based on a least square method, and combining the calculated time for the falling object to fall to a specific height, so as to determine the moving distance and the direction in the horizontal direction, and thus, calculating the estimated falling position of the falling object.
In addition, based on the established motion track in the three-dimensional direction, in combination with the method for determining the discrete state of the falling object in the height direction by the variance value described above, the discrete state of the falling object in the whole three-dimensional direction can be further determined. If the actual position of the falling object deviates greatly from the theoretical position in one or more three-dimensional directions at a certain moment or at a certain position, and if the deviation exceeds a preset threshold value, the falling object is judged to be in a discrete motion state. In one example, when the falling object falls to a height position of 10 meters above the vehicle, the difference between the actual three-dimensional position and the theoretical position at that time is calculated, such as a difference of 3 meters in the height direction, a difference of 4 meters in the horizontal X direction, and a difference of 1 meter in the Y direction, wherein the height direction and the horizontal X direction both reach or exceed a predetermined threshold of 3 meters for that height position, the falling object is considered to be in a discrete motion state.
In another embodiment of the invention, the fitting degree of the actual motion trail and the theoretical motion trail of the falling object is determined by utilizing an R-square algorithm. Taking the fitting of the motion trail in the height direction as an example, the following R-square algorithm is adopted to calculate the fitting degree of the motion trail:
wherein h isiIs tiThe actual falling height of the falling object at any moment,
represents the mean value of the actual height of the falling object at each time point, h (i) is tiThe fitting height at the moment.
R2The larger the value, i.e. the closer to 1, the better the fit, and the closer the theoretical motion trajectory is to the actual motion. R2The smaller the value, i.e., the closer to 0, the worse the fitness, the more the theoretical motion trajectory deviates from the actual motion. Also, based on similar principles, the degree of motion trajectory fit of the falling object in the direction of horizontal X, Y can be determined by an R-square algorithm. The detailed calculation method is not described herein.
In other embodiments of the invention, the fitting function of the falling object motion trajectory is calculated by using an algorithm such as a multiple linear regression method or a gradient descent method. The multiple linear regression method or the gradient descent method itself is an existing algorithm and will not be further illustrated in detail.
In the above embodiment, the detection system 12 employs a combination of the first detection device 120 and the second detection device 121, wherein the parameter for calculating the falling object trajectory is derived from the first detection device 120, such as millimeter wave radar or meter wave radar. In other embodiments of the present invention, the parameter may come from the second detection device 121, i.e. the lidar, or from both the lidar and the millimeter wave radar or the meter wave radar.
In other embodiments, detection system 12 may take the form of a single detection device or other combination, including but not limited to a combination of one or more of lidar, meter wave radar, millimeter wave radar, image acquisition devices. As in other embodiments of the present invention, the detection system 12 may use an image capturing Device, such as a CCD ("Charge Coupled Device", CCD image sensor), a CMOS ("Complementary Metal Oxide Semiconductor", CMOS), or the like, which may also enable detection of parameters of the falling object, and thus the control system 11 may be able to determine the motion state of the falling object and the 3D characteristics of the falling object based on the parameters, and determine emergency measures based on the motion state and the 3D characteristics. For example, U.S. patent application publication Nos. US20180350083A and US20180341263A, which each disclose methods and apparatus for object position and velocity detection using image sensing devices, are incorporated herein by reference.
The detection system 12 may be mounted to the roof of the vehicle and the periphery of the vehicle, with the detection range covering the entire range of motion of the falling object, such as the entire space within a certain distance to and above the location of the vehicle. The basic arrangement of detection systems is well known in the art, and for example, U.S. patent nos. US9904375B and US9625582B, and U.S. patent application publication nos. US20170364758A, US20180128922A, and US20180257582A, disclose the arrangement and specific application of radar and image sensing devices in vehicles or unmanned aerial vehicles. The present invention includes the contents of these patent documents by reference within the scope of the disclosure of the present invention.
When the vehicle detects danger and starts the early warning mode, the passenger is prompted through the buzzer or voice, the accident recording system 13 displays the falling track of the object through the display device 131, and the camera 130 is opened to record the image of the object falling process and the accident process. The accident recording system further transmits the falling process and the accident process of the object to the control system and the cloud end for corresponding analysis. According to one embodiment of the invention, the danger coefficient of the falling object is further calculated when the early warning mode is started. The risk factor depends on a comprehensive evaluation of a number of reference factors including, but not limited to, the area of the vehicle at risk of the falling, the roof material of the vehicle, the speed of the falling as it travels to the vehicle, the volume of the falling, the area of the vehicle covered by the falling, whether the area covered by the falling is occupied by a person, whether the area covered by the falling is occupied by a battery or a fuel tank, etc. The conditions of people in the vehicle can be detected and identified through an infrared sensor or a pressure sensor and the like. Fig. 4 is an example of ranking the above-mentioned various reference factors in one embodiment of the present invention. For example, in this example, the roof material may be divided into three classes, each representing a different roof structural strength.
Fig. 5 is a reference diagram for calculating a risk factor for a falling object hitting a cabin in an example, in which a covering area of the falling object, a roof material, whether a person is covered, and a potential energy of the falling object are distinguished, so that the risk factor is obtained by weighted calculation. The weight value can be preset according to experience, for example, the influence of each reference factor on final damage is analyzed according to the past simulation accident and actual accident situation, so that the corresponding preset weight value of each reference factor is determined. In addition, the control system 11 continuously updates the preset weight setting according to the data generated by the vehicle and the data acquired from the cloud, so that the preset weight setting is gradually accurate. According to an exemplary reference table shown in fig. 5, in the case that the coverage area is up to the super large standard, the roof material strength is high, there are people in the coverage area, and the potential energy is large, the risk coefficient is:
A＝20*1+25*0.2+30*1+25*0.8＝75。
FIG. 6 is a reference graph of the calculation of risk factors for a crash hitting a non-cabin area in one embodiment. The covering area of the falling object, whether the battery or the oil tank area is covered or not and the potential energy of the falling object are distinguished, so that the risk coefficient is obtained through weighting calculation. For example, according to the reference table shown in fig. 6, in the case of medium coverage, no coverage to the battery or the fuel tank, and a potential energy level of medium, the risk factor is:
A＝20*0.55+25*0.2+25*0.5＝28.5。
FIG. 7 further determines a risk level based on the risk factor. For example, a hazard value of 75 would represent a hazard level of "wounding" and a hazard value of 28.5 would represent a "hazard" level.
Once the risk factor and the risk level are evaluated, the control system 11 takes corresponding emergency measures. Emergency measures include, but are not limited to, emergency braking, acceleration of the vehicle through, steering of the vehicle, deployment of roof airbags, and cabin safety protection. Since these measures also present a certain risk in themselves, for example, the accelerated passage of a vehicle may lead to accidents with other vehicles or persons, it is a prerequisite that the risk factor of falling objects is much greater than the risk factor of carrying out the above emergency measures. The emergency action's own risk factor may be determined based on one or more of the following factors: whether there is a pedestrian or other vehicle in front of the vehicle, whether there is a pedestrian or other vehicle in the yaw direction, whether there is a following vehicle behind the vehicle, and whether the angle of yaw reaches a rollover limit. Specifically, different consideration factors are determined for different emergency measures. The method specifically comprises the following steps:
before deciding to take the emergency braking, the following factors are judged: 1) an orientation of an expected fall location of the fall with respect to the vehicle; 2) a reduced risk factor situation that is expected to be achievable by emergency braking; and 3) a collision with a rear vehicle that may be caused by emergency braking. Wherein the presence and speed of the rear vehicle can be detected by means of radar or image sensors arranged behind or on the roof of the vehicle. In one example, emergency braking action may be taken when the falling object is above and in front of the vehicle, the risk factor may be reduced by emergency braking, and at the same time emergency braking does not result in a collision with the vehicle behind or a slight probability of a collision. The judgment of whether to reduce the risk factor can refer to the calculation method of the risk factor described above. If the collision area between the vehicle and the falling object is changed from the expected roof position to the head position by the emergency braking, the risk factor is reduced.
Before deciding to take an accelerated pass, the following factors are judged: 1) pedestrian and other vehicle conditions ahead; 2) an orientation of an expected fall location of the fall with respect to the vehicle; 3) the risk factor reduction situation that is expected to be achievable is accelerated. The detection of the pedestrian or the vehicle in front of the vehicle may be performed by an infrared sensor, an image sensor, a radar device, or the like on the vehicle. In one example, a fall is expected to fall directly above the vehicle, the fall may be avoided by accelerating the pass, and the accelerating pass may not cause damage to the pedestrian and vehicle ahead, in which case the accelerating pass may be selected to avoid the hazard.
Before deciding to take a yaw ride, the following factors are judged: 1) pedestrian and other vehicle conditions in the yaw direction; 2) the expected achievable risk factor reduction situation is exercised by deflection; 3) whether the required yaw angle exceeds the rollover limit value. In one example, where there are no pedestrians or other vehicles in the direction of vehicle yaw, yaw play may reduce the risk factor and the desired yaw angle is insufficient to cause the vehicle to roll over, in which case yaw play may be selected to avoid the hazard. Wherein the rollover limit may be calculated by reference to the following example:
assuming a vehicle height of 1.5 meters, a width of 1.8 meters, a total vehicle weight Mg of 1500 kilograms, a center-of-gravity-to-ground distance L1 of 0.5 meters, and a tire edge distance L2 of 0.8 meters (i.e., arm of force), according to the moment formula F × L1 > Mg × L2, where F is centrifugal force, rollover will occur when F > 24000N.
The safety protection in the cabin is performed by a safety protection system 16 in the cabin, and the safety protection system 16 in the cabin is composed of a danger detection module and a danger coping module. Even small errors can be life threatening due to errors in the detection of the radar. Therefore, in one embodiment, a danger detection module is added in the cabin for secondary confirmation so as to ensure the safety of the passengers. Referring also to FIG. 8, in one embodiment, the hazard detection module includes a pressure sensor (not shown) located on the roof of the vehicle and an infrared detection system 15. The infrared detection system 15 further comprises a first infrared beam-blocking sensor 150 located inside the roof of the vehicle, and a second infrared beam-blocking sensor 151 located at the head of the seat back. Wherein the vehicle roof pressure sensor is disposed at the vehicle interior roof end for confirming the occurrence of an actual impact. The first infrared beam interrupter sensor 150 and the second infrared beam interrupter sensor 151 include a transmitter and a receiver, respectively. For the first infrared beam interruption sensor 150, the transmitter and the receiver may be located at the front and rear ends or the left and right ends, respectively, below the roof of the vehicle. For the second infrared beam blocking sensor 151, one of the transmitter or receiver may be placed in a retractable back head of a seat back, such as a headrest, and the other of the transmitter or receiver may be placed in front of or behind the substantially horizontal. In one embodiment, the backrest head's telescoping criteria are: the infrared level is at least 5cm above the head of the highest occupant in the vehicle and/or at most within 5cm of the vehicle roof. The danger detection module further comprises an infrared distance measuring device (not shown), when a danger occurs, the falling object or other vehicle parts trigger infrared ray interruption, and the infrared ray returns to the receiving device to realize distance measurement, so that the actual direction of the falling object is further determined.
In one embodiment, first infrared beam break sensor 150 and second infrared beam break sensor 151 are located at different heights between the roof and the passenger. When the first infrared beam blocking sensor 150 senses that light is blocked, but the second infrared beam blocking sensor 151 located below does not sense that light is blocked, there is a possibility that the first infrared beam blocking sensor is triggered by a passenger by mistake or the roof of the vehicle is impacted by a falling object, but the deformation of the roof is small so that the second infrared beam blocking sensor 151 is not triggered, and the safety of people in the vehicle is generally not affected. When the two lights are intercepted, the instantaneous speed of the falling object can be calculated through the height difference of the two lights and the sensed time difference, and therefore relevant protective measures can be taken according to the speed. In one embodiment of the present invention, the second infrared beam blocking sensor 151 is disposed for each seat in the vehicle, so as to detect and protect the falling object for each seat, and identify the position of the falling object. The second ir beam blocking sensor 151 of each seat may be one, or a plurality of sensors may be provided side by side.
In other embodiments, the first infrared beam blocking sensor 150 and the second infrared beam blocking sensor 151 may be implemented separately, but not necessarily in combination, and may have various features as described above, so that the falling object detection and the falling object direction judgment can be performed in the same manner.
And the danger coping module starts safety protection in the cabin according to the dangerous condition. In one embodiment, if all seats have infrared devices blocked, indicating that the vehicle body is heavily crushed, the vehicle's overall frame is damaged. At this point the back of the seat is lowered rapidly while the person is simultaneously tipped backwards with the assistance of the harness to reduce the height. At the moment, the weight is born only by the frame of the vehicle body, and people are prevented from being directly injured by falling objects or roof parts. If only the infrared ray of the front seat is intercepted, and meanwhile, no person exists in the corresponding rear seat, the seat is forcibly pulled to the rear, and the pulling speed is enough to enable the passenger to avoid the injury of falling objects. If the driver seat is hit, the emergency brake is triggered at the same time. If the falling object is small and is at the front end position, the seat can be overturned to avoid the impact. If only the outside infrared light of the front seat is intercepted, the danger can be avoided by rotating the seat angle. Further, in one embodiment, a retractable head restraint is provided to support the roof or fall. When the vehicle prejudges or detects the danger of falling objects according to the mode, the headrest is upwards popped out by a certain height to play a role in supporting the roof and the falling objects, so that the safety of people in the vehicle is protected. The above adjustments of the seat in the vehicle interior, such as the lowering of the backrest, the movement or tilting of the seat, the ejection of the headrest, etc., can be performed by the control system 11 controlling the corresponding actuating devices, which can include motors, hydraulic and pneumatic cylinders, etc. The amount of movement of the seat is such that the safety of the occupant is ensured, for example by lowering the back so that the maximum height of the occupant, when laid down, is below the minimum height of the damaged B-pillar of the vehicle.
In one embodiment of the invention, the safety protection in the cabin may be opened based on the above described detection of a falling object and further determination of the pressure sensor and/or the infrared detection system in the vehicle. In other embodiments of the invention, the in-cabin safety protection may be activated first based solely on the detection of a falling object by a detection system, such as a radar system, without waiting for further detection by the in-vehicle pressure sensor and/or the infrared detection system. For example, when the radar system detects that the speed or the volume of the falling object is large, and the danger caused by the falling object cannot be avoided or reduced by taking other emergency measures such as emergency braking and the like, or the danger caused by other emergency measures is too high, the safety protection in the cabin can be directly opened, such as quickly adjusting the seat structure, the position and the like of people in the vehicle, so that the injury of the falling object to the people is avoided.
In another embodiment of the invention, the in-cabin safety protection may be implemented separately from the application scenario in which the detection system 11 detects a falling object. That is, the falling object is detected solely based on the pressure sensor and/or the infrared detection system, and safety protection measures in the vehicle are started based on the detection result, such as the above-described movement control of the seat, the seat back, the headrest, and the like in the vehicle.
The invention is described above, the corresponding preset weight of the risk factor is determined manually or mechanically, the risk coefficient of the falling object is calculated, and the risk coefficient of the emergency measure is determined at the same time, so that the emergency measure to be finally taken is determined. In addition, another embodiment of the present invention calculates various external dangerous characteristic data and implementation measures by using a bp (back propagation) neural network algorithm, thereby determining the corresponding measures. Under the complex environment that the actual vehicle encounters objects falling from high altitude, an optimal countermeasure can be scientifically obtained by a computer through an algorithm.
Referring to fig. 9 to 11, fig. 9 and 10 are flowcharts of different embodiments of the present invention for making vehicle protection decision based on the BP neural network algorithm. FIG. 11 is a diagram of a model of a BP neural network algorithm according to an embodiment of the present invention, the model including an input layer, a hidden layer, and an output layer. In the figure, X1 to X10 represent input nodes, i.e., input parameters, H1 to H5 represent nodes of the hidden layer, and O1 to O10 represent output nodes, i.e., emergency measures to be taken in the present invention. The basic principles of the BP neural network algorithm are well known in the art, and reference may be made to the descriptions of the BP neural network disclosed in U.S. patent application publication No. US20160071010A and U.S. patent application publication No. US20130204818A, which are incorporated herein by reference. The invention is described below only with respect to the specific application of the BP neural network algorithm in vehicle protection decision making.
According to the embodiment shown in fig. 9, in step S10, corresponding data is collected for falling object tests performed in different complex scenes by using a detection system equipped with the vehicle, including but not limited to laser radar, millimeter wave radar, image capturing device, and the like. The data format collected contains two aspects: dangerous characteristics and degree of injury. The dangerous characteristics refer to all data which influence the decision-making and response measures of the vehicle system, such as various data of falling objects, current data of vehicles, external environment data and the like, and include but are not limited to: the method comprises the following steps of (1) expected final speed of the falling object, expected falling position of the falling object, roof pressure bearing capacity, personnel condition in the vehicle, surrounding environment condition of the vehicle, covering area of the falling object, current vehicle speed, driving direction and theoretical braking distance. Each feature data corresponds to each input node in the BP neural network input layer. In these features, the expected final speed of the falling object and the expected falling position of the falling object can be determined by the algorithm described above, and the conditions of the person in the vehicle and the conditions of the environment around the vehicle can be obtained by the sensing devices such as the radar, the infrared sensor, and the image sensor, which are provided in the vehicle. The degree of injury refers to the degree of danger caused by taking different measures when encountering the same or similar test environment, such as the degree of danger caused by emergency braking, accelerated passage, deflected driving, safety protection in a cabin and the like.
In step S11, the collected receipts are processed, including data cleansing and data marking. In the data cleaning process, the collected abnormal data is screened and deleted, for example, invalid data, data with an excessive standard deviation, illegal data and the like are removed. For example, if the speed of capturing a falling object in a sample is zero, the data is invalid and the record needs to be deleted. There are a number of methods for data cleansing, and in one embodiment, data cleansing is performed using, for example, box elimination in statistics. Data tags refer to being in the same set of environmentsUnder the characteristic value, n times of tests are carried out in total, and m avoidance strategies S are adopted1-SmAnd selecting a strategy with the lowest harm to record in the corresponding environment characteristic value sample according to the sample data of the test and the safety evaluation of the vehicle. For example, in three similar experiments, emergency braking S was used1Is minimized, the corresponding avoidance maneuver marking is performed on the set of dangerous characteristics.
In step S12, a BP neural network is established using a three-layer neural network architecture, including an input layer, a hidden layer, and an output layer, and additionally, functions including, but not limited to, a Sigmoid function, a Re L U function, and a Softmax function are employed as activation functions for the hidden layer of the neural network.
Y＝Softmax(Relu(W1*X+B1)*W2+B2) (formula 9)
Wherein, W1And B1Is a weight matrix and a bias matrix between the input layer and the hidden layer for calculating the hidden layer matrix. W2And B2Representing a weight matrix and a bias matrix between the hidden layer and the output layer for computing a matrix for the output layer.
The individual nodes of the input layer of the neural network are derived from the danger features as mentioned above. The nodes of the output layer include, but are not limited to, emergency braking, acceleration passing, and yaw driving.
In step S13, a portion of the sample data is used for training of the neural network model, and the remaining data is used for testing of the model.
In step S14, a model evaluation of the neural network is performed. Through the selection of a plurality of different hyper-parameters, a plurality of algorithm models can be trained. At this time, the model needs to be verified by using test data, and the learning rate, the recognition accuracy and whether the overfitting condition occurs are checked.
An algorithm capable of realizing the minimum injury is established according to the previous verification condition in step 15, and the probability of taking a certain countermeasure is calculated according to the parameters of the falling object, such as the final falling speed, the vehicle speed, the number of passengers on the vehicle and the like in step 16, wherein the decision with the highest probability is the optimal strategy.
For each actual accident and simulated accident, the vehicle control system 11 records relevant parameters in the accident process, including but not limited to, relevant parameters of falling objects in the accident process, such as speed, motion trajectory, volume, falling position and the like, and relevant parameters and conditions of the vehicle, such as speed, damage condition of the vehicle, emergency measures taken in the accident process and the like, and stores and transmits the data to the cloud. And analyzing the data locally or in a cloud, and updating and adjusting related algorithm models and related parameter settings of the vehicle protection strategy according to the data. The updating process is carried out circularly, and different vehicle data are shared through the cloud end, so that corresponding algorithms can be rapidly and continuously perfected according to a large amount of data about vehicle emergency precaution, and the automatic driving safety level of the vehicle is improved.
FIG. 10 is a flowchart of decision making based on BP neural network algorithm according to another embodiment of the present invention. Unlike the embodiment represented in fig. 9, the vehicle makes decisions based on existing algorithms without the need to build new algorithmic models. Specifically, in step S20, data collection is performed, and optionally step S21 is simultaneously performed to store and record corresponding data. In step S22, it is checked whether the network interconnecting the vehicle and the cloud is good, and if the network is poor, the existing algorithm local to the vehicle is used for analysis, and a decision is made in step S25 to select a correct emergency measure. If the network is good, the latest algorithm in the cloud may be called in step S24, and then a decision may be made in step S25 to select the correct emergency measure. The results of the decision, such as whether the decision is appropriate and the actual circumstances of the resulting accident, are analyzed and judged in step S26, and an accident recording is performed in step S27. Based on the decision taken and the various data generated by the corresponding accident, training learning of the algorithm is performed in step 28, and the generated new algorithm is updated locally and in the cloud in step S29. The above processes can be performed in a circulating manner under each dangerous state of the vehicle so as to continuously improve the algorithm. Referring to fig. 11 in combination, one embodiment of the present invention provides a vehicle safety protection method. In step S30, the first detecting device 120 continuously detects whether there is a falling object, such as a millimeter wave radar or a meter wave radar, and if there is a falling object, in step S31 transmits parameters of the detected falling object, such as a distance, an angle, a speed, an acceleration, and a volume of the falling object from the vehicle, to the control system 11. Based on the above parameters, the control system 11 can determine the state of motion of the falling object, for example, determine whether the falling acceleration of the falling object is greater than a predetermined threshold (e.g., 0.6g), or whether the falling object motion trajectory approaches a parabolic state of motion. Meanwhile, in step S32, it is determined whether the falling object enters the detection range of the second detection means 121 such as the laser radar. And if the falling object does not enter the detection range of the laser radar, continuously detecting the millimeter wave radar or the meter wave radar. Once the falling object enters the detection range of the lidar, the lidar is turned on to detect the falling object, including the position, angle, and 3D model parameters of the falling object, and transmit these parameters to the control system 11 to calculate the 3D characteristics of the falling object in step S33. In step S34, the control system 11 determines an emergency measure based on the 3D characteristics and the motion state of the falling object.
The above flow of the vehicle protection method is only one embodiment of the present invention, and those skilled in the art will understand that the spirit of the present invention is not limited to the above specific steps, and can be modified accordingly. For example, in one embodiment, the lidar may be continuously turned on for high-altitude falling object detection without having to determine in advance whether the falling object enters the detection range.
In the above embodiment, the control system 11 calculates the motion state and 3D characteristics of the falling object by receiving the parameters about the falling object from the first detecting means 120 and the second detecting means 121. In other embodiments, the first detection device 120 and the second detection device 121 integrate the functions of data calculation and analysis, so that the final result or intermediate result can be transmitted to the control system 11 after data analysis, and the control system 11 finally determines the motion state and 3D characteristics of the falling object and finally confirms the emergency measures to be taken.
It should be noted that, as one of ordinary skill in the art would understand, all or part of the processes in the above method may be implemented by a computer program and then the computer program instructs related hardware to complete the processes, and the program may be stored in a computer readable storage medium. The above described programs, when executed, may comprise the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM) or a Random Access Memory (RAM). An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for vehicle safety protection as described above, and specific functions are not described herein again.
Furthermore, the control system 11 according to the present invention may comprise various aspects in practical applications, such as it is integrated in an ADAS system as part of the ADAS system. The control system 11 may also be present as a relatively independent system, for example constituting a roof safety decision system of a vehicle. Further, the control system 11 should not be limited to a certain independent device such as a CPU or the like, but includes a general term for all hardware and software related to calculation and control. Meanwhile, the hardware and software can be centralized or dispersed in each module in the whole vehicle safety protection system.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (20)
1. A vehicle safety shield system comprising:
a detection system that detects parameters of a falling object, the parameters including a speed and a position of the falling object; and
a control system that determines an emergency action to be taken by the vehicle based on the parameter.
2. The vehicle safety protection system of claim 1, wherein the detection system determines a motion profile of the falling object based on the parameter.
3. The vehicle safety protection system of claim 1, wherein the detection system further detects 3D model parameters of the drop and determines 3D characteristics of the drop based on the 3D model parameters.
4. The vehicle safety protection system of claim 1, wherein the detection system is a combination of one or more of: laser radar, meter wave radar, millimeter wave radar and image acquisition device.
5. The vehicle safety protection system of claim 4, wherein the millimeter wave radar or the meter wave radar detects speed and position parameters of a falling object, and the lidar detects 3D model parameters of a falling object and/or speed and position parameters of a falling object.
6. The vehicle safety protection system of claim 1, wherein the control system determines the hazard coefficient based on: the volume of the falling object, the speed when the falling object falls to a specific height, and the area and the position of the covered vehicle, and the adoption of emergency measures is determined based on the risk coefficient, and the control system performs weighted calculation on the factors according to a preset weight value to determine the risk coefficient.
7. The vehicle safety protection system of claim 6, wherein the control system determines adoption of the emergency measure based on a comparison of the emergency measure's own risk factor to a risk factor of the falling object, the emergency measure comprising one or more of: the method comprises the steps of opening a car roof air bag, emergency braking, accelerating passing, deflecting running and safety protection in a cabin.
8. The vehicle safety protection system of claim 1, wherein the control system determines the emergency measure based on a BP neural network algorithm, model input nodes of the BP neural network algorithm comprising one or more of: the method comprises the following steps of (1) expected final speed of the falling object, expected falling position of the falling object, roof pressure bearing capacity, personnel condition in the vehicle, surrounding environment condition of the vehicle, covering area of the falling object, current vehicle speed, driving direction and theoretical braking distance.
9. The vehicle safety protection system of claim 8, wherein the control system is configured to perform training and updating of the BP neural network algorithm based on analysis of results from taking emergency action.
10. A method for vehicle safety protection based on the vehicle safety protection system of any one of claims 1-9, comprising:
detecting parameters of a falling object, the parameters including a speed and a position of the falling object; and
an emergency action to be taken by the vehicle is determined based on the parameter.
11. The utility model provides an under-deck safety protection system, includes the danger detection module, the danger detection module includes infrared detection system, infrared detection system includes at least one infrared beam and hides disconnected formula inductor for detect the collision of falling object and vehicle.
12. The under-cabin safety shield system according to claim 11, wherein the at least one infrared beam-blocking sensor comprises a first infrared beam-blocking sensor and a second infrared beam-blocking sensor located at different heights.
13. The under-deck safety shield system according to claim 11, wherein said at least one infrared beam-blocking sensor comprises a transmitter and a receiver, said transmitter and receiver being located at either or both of the front and rear ends of the underside of the roof.
14. The under-cabin safety shield system of claim 11, wherein the at least one infrared beam-blocking sensor comprises a transmitter and a receiver, one of the transmitter or the receiver being disposed in a retractable headrest of a seat back, the other of the transmitter or the receiver being disposed in front of or behind horizontal.
15. The under-cabin safety shield system according to claim 14, wherein an uppermost position of the infrared level in the retractable headrest is within no more than 5cm of the vehicle roof.
16. An intra-cabin safety protection system according to claim 11, wherein the at least one infrared beam-blocking sensor is arranged corresponding to the position of each seat in the vehicle.
17. The under-cabin safety shield system of claim 11, wherein the hazard detection module further comprises a pressure sensor located on the roof of the vehicle to detect a falling object.
18. The under-cabin safety protection system of claim 11, wherein the hazard detection module further comprises an infrared ranging device to determine an actual impact location of the falling object.
19. An intra-cabin safety shield system according to claim 11, wherein the intra-cabin safety shield system comprises a hazard management module comprising an actuation device to drive a seat, backrest and/or headrest.
20. The under-cabin safety shield system according to claim 19, wherein the actuation device comprises at least one of a motor, a cylinder, and a hydraulic cylinder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910093102.5A CN111497773A (en) | 2019-01-30 | 2019-01-30 | Vehicle safety protection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910093102.5A CN111497773A (en) | 2019-01-30 | 2019-01-30 | Vehicle safety protection system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111497773A true CN111497773A (en) | 2020-08-07 |
Family
ID=71865136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910093102.5A Pending CN111497773A (en) | 2019-01-30 | 2019-01-30 | Vehicle safety protection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111497773A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112009416A (en) * | 2020-08-11 | 2020-12-01 | 广州汽车集团股份有限公司 | Vehicle protection method and system and automobile |
| CN112606833A (en) * | 2020-12-21 | 2021-04-06 | 长沙理工大学 | Model prediction control-based intelligent vehicle motion control method for protecting human-ground collision damage |
| CN112731369A (en) * | 2020-12-25 | 2021-04-30 | 上海为彪汽配制造有限公司 | High-altitude parabolic early warning method and system |
| CN112918419A (en) * | 2021-03-18 | 2021-06-08 | 东风汽车集团股份有限公司 | Vehicle-mounted method and device for preventing falling object from high altitude |
| CN113548042A (en) * | 2021-05-17 | 2021-10-26 | 中国第一汽车股份有限公司 | Anti-smashing emergency avoidance control method and system for automobile and automobile |
| CN114715068A (en) * | 2022-04-06 | 2022-07-08 | 东风汽车集团股份有限公司 | Protection method and protection system based on external airbag of vehicle body and vehicle |
| CN115082864A (en) * | 2022-07-25 | 2022-09-20 | 青岛亨通建设有限公司 | Building construction safety monitoring system |
| CN116279113A (en) * | 2023-01-12 | 2023-06-23 | 润芯微科技(江苏)有限公司 | Extreme weather early warning method, system and storage medium |
| US11919479B2 (en) | 2021-05-18 | 2024-03-05 | Ford Global Technologies, Llc | Systems and methods for providing security to a vehicle |
-
2019
- 2019-01-30 CN CN201910093102.5A patent/CN111497773A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112009416A (en) * | 2020-08-11 | 2020-12-01 | 广州汽车集团股份有限公司 | Vehicle protection method and system and automobile |
| CN112606833A (en) * | 2020-12-21 | 2021-04-06 | 长沙理工大学 | Model prediction control-based intelligent vehicle motion control method for protecting human-ground collision damage |
| CN112731369A (en) * | 2020-12-25 | 2021-04-30 | 上海为彪汽配制造有限公司 | High-altitude parabolic early warning method and system |
| CN112731369B (en) * | 2020-12-25 | 2024-03-01 | 上海为彪汽配制造有限公司 | High-altitude parabolic early warning method and system |
| CN112918419A (en) * | 2021-03-18 | 2021-06-08 | 东风汽车集团股份有限公司 | Vehicle-mounted method and device for preventing falling object from high altitude |
| CN113548042A (en) * | 2021-05-17 | 2021-10-26 | 中国第一汽车股份有限公司 | Anti-smashing emergency avoidance control method and system for automobile and automobile |
| US11919479B2 (en) | 2021-05-18 | 2024-03-05 | Ford Global Technologies, Llc | Systems and methods for providing security to a vehicle |
| CN114715068A (en) * | 2022-04-06 | 2022-07-08 | 东风汽车集团股份有限公司 | Protection method and protection system based on external airbag of vehicle body and vehicle |
| CN115082864A (en) * | 2022-07-25 | 2022-09-20 | 青岛亨通建设有限公司 | Building construction safety monitoring system |
| CN116279113A (en) * | 2023-01-12 | 2023-06-23 | 润芯微科技(江苏)有限公司 | Extreme weather early warning method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111497773A (en) | Vehicle safety protection system and method | |
| US11370541B2 (en) | Accident monitoring using remotely operated or autonomous aerial vehicles | |
| US20230069346A1 (en) | Methods and Systems for Detecting Weather Conditions using Vehicle Onboard Sensors | |
| JP4760715B2 (en) | Vehicle motion control device | |
| US9360556B2 (en) | Methods and systems for detecting weather conditions including fog using vehicle onboard sensors | |
| US9632210B2 (en) | Methods and systems for detecting weather conditions using vehicle onboard sensors | |
| US20180284789A1 (en) | Vehicle control system, vehicle control method, and vehicle control program | |
| US20230013136A1 (en) | System and Method for Adjusting an Interior Configuration of a Vehicle in Response to a Vehicular Accident | |
| EP2473388B1 (en) | Vehicle or traffic control method and system | |
| CN110494339B (en) | Vehicle control system, vehicle control method, and storage medium | |
| US20220314839A1 (en) | Adjusting interior configuration of a vehicle based on external environment data | |
| KR20200044157A (en) | Methods and systems for detecting weather conditions using vehicle onboard sensors | |
| CN106463059A (en) | Obstacle-information-managing device | |
| CN110461677B (en) | Vehicle control system, vehicle control method, and storage medium | |
| CN109466488A (en) | A kind of vehicle collision automatic help alarm system | |
| CN110065494A (en) | A kind of vehicle collision avoidance method based on wheel detection | |
| US11807257B2 (en) | Sensing interactions with unpermitted components within a vehicle | |
| CN103373350B (en) | Large animal vehicle collision safety apparatus and method | |
| US20210323444A1 (en) | Adjusting interior configuration of a vehicle based on vehicle operation | |
| US20220089063A1 (en) | Dynamic limiting of interior vehicle configurations | |
| CN209534979U (en) | Security protection system in cabin | |
| US20220388464A1 (en) | Sensing the ingress of water into a vehicle | |
| CN115273035A (en) | Method, device and equipment for detecting air floating object and storage medium | |
| JP2020140445A (en) | Vehicle control system, vehicle control method, and program | |
| JP2021011152A (en) | Vehicle control device, vehicle control method and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |