Raster map updating method and device, medium, driving assistance system and vehicle

Abstract

The invention discloses a grid map updating method, electronic equipment, a storage medium, an automatic driving assistance device and a vehicle, and the grid map updating method comprises the steps: obtaining the position information of a target point and the obstacle attribute of the target point, the target point being a point in a grid map corresponding to the field angle range of an ultrasonic sensor; obtaining a probability updating parameter of the target point according to the position information and the obstacle attribute; updating the obstacle judgment probability of the target point according to the probability updating parameter; wherein when an obstacle exists in the field angle range of the ultrasonic sensor, the updating efficiency of the probability updating parameters in the direction where the obstacle is located is the highest, and the updating efficiency of the probability updating parameters is gradually reduced in sequence on the two sides deviating from the direction where the obstacle is located. According to the method, the edge of the obstacle can be effectively prevented from being amplified, and the obstacle detection precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a grid map updating method, and also to an electronic device, a non-volatile readable storage medium, an automatic driving assistance system and a vehicle. Background Art

[0002] In the prior art, in the ultrasonic mapping method, it is necessary to update the grid within the range covered by the ultrasound in the grid map, that is, to update the probability of each point in the grid being judged as an obstacle. When a frame of ultrasonic rays passes through the probability grid, the arc where the obstacle is located can be determined according to the field of view of the ultrasonic sensor and the distance to the obstacle. The points on the line connecting the ultrasonic sensor to the arc are all considered to be miss points, and the points on the arc are all considered to be hit points. The hit points are all updated according to the same update parameter p. hit This will cause some points on the edge of the obstacle to be considered as obstacle points, causing the obstacle edge to be enlarged and reducing the detection accuracy. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, one object of the present invention is to provide a grid map updating method, which can effectively limit the enlargement of obstacle edges and improve obstacle detection accuracy.

[0004] A second objective of the present invention is to provide an electronic device.

[0005] The third object of the present invention is to provide a non-volatile readable storage medium.

[0006] The fourth objective of the present invention is to provide an automatic driving assistance system.

[0007] A fifth object of the present invention is to provide a vehicle.

[0008] In order to achieve the above-mentioned purpose, a grid map updating method of an embodiment of the first aspect of the present invention includes: obtaining position information of a target point and obstacle attributes of the target point, the target point being a point in a grid map corresponding to the field of view angle range of an ultrasonic sensor; obtaining a probability update parameter of the target point according to the position information and the obstacle attributes; updating the obstacle judgment probability of the target point according to the probability update parameter; wherein, when there is an obstacle within the field of view angle range of the ultrasonic sensor, the update efficiency of the probability update parameter is the highest in the direction where the obstacle is located, and the update efficiency of the probability update parameter decreases successively on both sides of the direction where the obstacle is located.

[0009] According to the grid map updating method of the embodiment of the present invention, different probability update parameters are obtained according to the position information of the target point in the grid map and the obstacle attributes, which can affect the efficiency of its probability update, that is, when there is an obstacle within the field of view of the ultrasonic sensor, the update efficiency of obstacle points at different positions is different. The direction where the obstacle is located has a high update efficiency and can be quickly determined as an obstacle point. At the edge of the obstacle, as the vehicle moves, the grid where the obstacle point in the direction of the obstacle is located will be quickly blackened due to the higher update efficiency, while the grids on both sides will not be blackened. Therefore, the enlargement of the obstacle edge can be effectively limited, thereby improving the accuracy of obstacle detection.

[0010] In some embodiments, when the obstacle does not exist within the field of view of the ultrasonic sensor, the update efficiency of the probability update parameter in the direction of the ultrasonic sensor is the highest, and the update efficiency of the probability update parameter decreases successively on both sides away from the direction of the ultrasonic sensor.

[0011] In some embodiments, obtaining the probability update parameter of the target point according to the position information and the obstacle attribute information includes:

[0012] The probability update parameter of the target point is obtained by a normal distribution probability model, the type of the probability update parameter is determined according to the obstacle attribute of the target point, the normal distribution probability model takes the target point angle as input, the target point angle is the angle between the target point line and the upper radial line of the ultrasonic sensor, and the target point line is the line between the target point and the ultrasonic sensor;

[0013] Wherein, the target point angle is determined according to the position information.

[0014] In some embodiments, the normal distribution probability model includes a normal distribution equation and a first mapping relationship, wherein the first mapping relationship is a mapping relationship between the target point angle and the abscissa value of the normal distribution equation;

[0015] Obtaining the probability update parameter of the target point corresponding to the obstacle attribute through the normal distribution probability model includes: obtaining the horizontal coordinate value of the target point angle corresponding to the normal distribution equation according to the first mapping relationship.

[0016] In some embodiments, when the target point angle is greater than the reference angle, the first mapping relationship is expressed as:

[0017]

[0018] Wherein, X is the horizontal coordinate value of the normal distribution equation, L α is the target point angle, and the value range of X is from 0 to X βBetween, L α The value of is between α and β. When X = 0, L α =α, X = X β Time α =β, α is the reference angle, β is the half angle of the field of view of the ultrasonic sensor, X β is the abscissa value of the normal distribution equation corresponding to the half angle of the field of view.

[0019] In some embodiments, when the target point angle is smaller than the reference angle, the first mapping relationship is expressed as:

[0020]

[0021] Wherein, X is the horizontal coordinate value of the normal distribution equation, L α is the target point angle, and the value range of X is from 0 to (-X β ), L α The value of is between α and (-β), when X = 0, L α =α, X = -X β Time α =-β, α is the reference angle, β is the half angle of the field of view of the ultrasonic sensor, X β is the abscissa value of the normal distribution equation corresponding to the half angle of the field of view.

[0022] In some embodiments, the reference angle is an angle between a radial line toward which the ultrasonic sensor is directed and an obstacle line, and the obstacle line is a line from the ultrasonic sensor to the obstacle.

[0023] In some embodiments, the orientation of the ultrasonic sensor and the position of the obstacle are determined based on vehicle environment visual information collected by an onboard visual device.

[0024] In some embodiments, obtaining the probability update parameter of the target point corresponding to the obstacle attribute through the normal distribution probability model also includes: obtaining the ordinate value of the normal distribution equation through the normal distribution equation according to the abscissa value of the normal distribution equation.

[0025] In some embodiments, the normal distribution probability model further includes a second mapping relationship, wherein the second mapping relationship is a correspondence between the ordinate value of the normal distribution equation and the probability update parameter;

[0026] Obtaining the probability update parameter of the target point corresponding to the obstacle attribute through the normal distribution probability model also includes: searching the second mapping relationship according to the ordinate value of the normal distribution equation to obtain the probability update parameter of the target point corresponding to the obstacle attribute.

[0027] In some embodiments, the second mapping relationship is expressed as:

[0028]

[0029] Wherein, P is the probability update parameter, y is the ordinate value of the normal distribution equation, P B Update the parameter y for the probability of setting the known point B. max P B The ordinate value of the corresponding normal distribution equation, y α is the ordinate value of the normal distribution equation corresponding to the half angle of the field of view.

[0030] In some embodiments, the obstacle attributes include obstacle points and non-obstacle points; when point B is the obstacle point, the P B is greater than 0.5, or when the point B is the non-obstacle point, the P B Less than 0.5.

[0031] In some embodiments, the target point is any point in the region of interest within the field of view of the corresponding ultrasonic sensor.

[0032] In some embodiments, the area of ​​interest includes a first grid area and a second grid area; the first grid area is a plurality of arcs centered on the ultrasonic sensor within the field of view of the ultrasonic sensor and with a distance at which the ultrasonic sensor can detect obstacles as a radius; the second grid area is the area between the ultrasonic sensor and each of the arcs.

[0033] In some embodiments, the obstacle attributes include obstacle points and non-obstacle points; the obstacle points correspond to points in the first grid area, and the non-obstacle points correspond to points in the second grid area.

[0034] In some embodiments, the plurality of arcs in the first grid region are arranged at intervals along the radial direction of the field of view of the ultrasonic sensor, and the arc length of the arc at the middle position of the plurality of arcs in the radial direction is longer than the arc length of the arcs at the two end positions.

[0035] In order to achieve the above-mentioned purpose, the second aspect of the present invention proposes an electronic device, comprising: at least one processor; a memory communicatively connected to the at least one processor; the memory stores a computer program executable by the at least one processor, and the at least one processor implements the grid map updating method when executing the computer program.

[0036] According to the electronic device of the embodiment of the present invention, at least one processor executes the grid map updating method of the above embodiment, and the grid points at various positions adopt different probability update parameters, which can effectively limit the enlargement of the obstacle edge and improve the obstacle detection accuracy.

[0037] In order to achieve the above-mentioned purpose, a third aspect of the present invention provides a non-volatile readable storage medium having a computer program stored thereon, and the computer program implements the grid map updating method when executed.

[0038] In order to achieve the above-mentioned objectives, an embodiment of the fourth aspect of the present invention proposes an automatic driving assistance system, comprising: an ultrasonic sensor for collecting obstacle information within a field of view angle; a controller, the controller being connected to the ultrasonic sensor, for constructing a grid map based on the obstacle information, and updating the grid map according to the grid map updating method.

[0039] According to the automatic driving assistance system of the embodiment of the present invention, a grid map is constructed based on ultrasound, and the controller performs grid probability update according to the grid map update method of the above embodiment. Different probability update parameters are used for the grid points at each position, which can effectively limit the enlargement of the obstacle edge and improve the obstacle detection accuracy.

[0040] In some embodiments, the automatic driving assistance system also includes: an on-board visual device, connected to the controller, for obtaining visual information of the vehicle environment; the controller is also used to determine the direction of the ultrasonic sensor and the position of the obstacle based on the visual information.

[0041] In order to achieve the above-mentioned purpose, a fifth aspect of the present invention proposes a vehicle, which includes the electronic device; or, the vehicle includes the automatic driving assistance system.

[0042] According to the vehicle of the embodiment of the present invention, the grid map is updated through the electronic device or the automatic driving assistance system of the above embodiment, and the grid points at each position use different probability update parameters, which can effectively limit the enlargement of the obstacle edge and improve the obstacle detection accuracy.

[0043] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0045] Figure 1 is a flowchart of a grid map updating method according to an embodiment of the present invention;

[0046] Figure 2 is a schematic diagram of the field of view angle range of an ultrasonic sensor according to an embodiment of the present invention;

[0047] Figure 3 is a schematic diagram of the field of view angle range of an ultrasonic sensor according to another embodiment of the present invention;

[0048] Figure 4 is a schematic diagram of a normal distribution equation according to an embodiment of the present invention;

[0049] Figure 5 is a schematic diagram of an arc in the field of view angle range of an ultrasonic sensor according to an embodiment of the present invention;

[0050] Figure 6 is a diagram showing the effect of left and right ultrasonic sensors mapping when α=0 according to an embodiment of the present invention;

[0051] Figure 7 is a diagram showing the effect of left and right ultrasonic sensors mapping when α<0 according to another embodiment of the present invention;

[0052] Figure 8 is a schematic diagram of ultrasonic mapping results according to an embodiment of the present invention;

[0053] Fig. 9 is a schematic diagram of ultrasonic mapping results according to another embodiment of the present invention;

[0054] Fig.10 is a block diagram of an electronic device according to an embodiment of the present invention;

[0055] Fig.11 is a block diagram of an automatic driving assistance system according to an embodiment of the present invention;

[0056] Fig.12 is a block diagram of a vehicle according to one embodiment of the present invention;

[0057] Fig.13 is a block diagram of a vehicle according to another embodiment of the present invention. DETAILED DESCRIPTION

[0058] Embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. Embodiments of the present invention are described in detail below.

[0059] Reference below Figure 1-Figure 9 A grid map updating method according to an embodiment of the present invention is described.

[0060] Figure 1 is a flowchart of a grid map updating method according to an embodiment of the present invention. Figure 1As shown, the grid map updating method of the embodiment of the present invention includes the following steps S1-S3.

[0061] S1, obtain the location information of the target point and the obstacle attributes of the target point.

[0062] Specifically, in the ultrasonic mapping method, the grid within the range covered by the ultrasound in the grid map, that is, the field of view of the ultrasonic sensor, is updated, that is, the probability of each point in the grid being judged as an obstacle is updated. In an embodiment of the present invention, the probability of each point in the grid being judged as an obstacle can be referred to as the obstacle judgment probability.

[0063] The target point may be a point in a grid map corresponding to the field of view of the ultrasonic sensor. The obstacle attribute may determine whether the point is an obstacle. For example, the obstacle attribute may include an obstacle point and a non-obstacle point.

[0064] For example, when updating the obstacle judgment probability of a single grid, when a frame of ultrasonic rays passes through the grid, a series of grid coordinates of hit points and miss points can be calculated, wherein the hit point, i.e., the obstacle point, can be set as the hit point, and the miss point, i.e., the non-obstacle point, can be set as the miss point.

[0065] S2, obtain the probability update parameters of the target point according to the location information of the target point and the attributes of the obstacle.

[0066] S3, updating the obstacle judgment probability of the target point according to the probability update parameter.

[0067] The position information of the target point may be the coordinates of the target point, such as any point within the field of view of the ultrasonic sensor, in the sensor coordinate system.

[0068] The probability update parameter may include a probability update parameter corresponding to the obstacle, for example, set to p hit The probability update parameter corresponding to the non-obstacle is set to p miss , determine the probability update parameters of the corresponding type according to the obstacle attributes of the target point.

[0069] Furthermore, in the embodiment of the present invention, the probability update parameter of the target point is related to the position information of the target point, and the same fixed probability update parameter is not used. The mapping relationship between the probability update parameter and the position information can be calibrated and saved. When the grid map is updated, the corresponding probability update parameter can be directly obtained based on the position information and attribute information of the target point, and the obstacle judgment probability of the target point can be updated according to the probability update parameter.

[0070] In an embodiment, when there is an obstacle within the field of view of the ultrasonic sensor, the update efficiency of the probability update parameter is the highest in the direction of the obstacle, and the update efficiency of the probability update parameter decreases successively on both sides of the direction of the obstacle. That is, for the obstacle, the probability update parameter of the target point is normally distributed. In this way, the update efficiency of the point at the obstacle is faster than that at the edge, which can avoid the enlargement of the obstacle edge.

[0071] In other embodiments, within the field of view angle range of the ultrasonic sensor, the update efficiency of the probability update parameters in the direction of the ultrasonic sensor is the highest, and the update efficiency of the probability update parameters decreases successively on both sides of the direction of the ultrasonic sensor. That is, within the corresponding field of view angle range of the ultrasonic sensor, with the direction of the ultrasonic sensor as the center, the update efficiency of the probability update parameters is normally distributed.

[0072] According to the grid map updating method of the embodiment of the present invention, when a vehicle passes in front of an obstacle and the ultrasonic sensor scans the obstacle, different probability update parameters are obtained according to position information and obstacle attributes at different positions within the scanning range of the ultrasonic sensor, i.e., within the field of view, which can affect the efficiency of its probability update. For obstacle points, the update efficiency of obstacle points at different positions is different. The direction where the obstacle is located has a high update efficiency and can be quickly determined as an obstacle point. At the edge of the obstacle, as the vehicle moves, the grid where the obstacle point in the direction of the obstacle is located will be quickly blackened due to the higher update efficiency, while the grids on both sides will not be blackened. Therefore, the enlargement of the obstacle edge can be effectively limited, thereby improving the accuracy of obstacle detection.

[0073] In some embodiments, the target point may be any point in the region of interest within the field of view of the corresponding ultrasonic sensor, wherein the region of interest may be determined according to the scanning characteristics of the ultrasonic sensor or set according to demand, thereby ensuring the accuracy of scanning obstacles and reducing the amount of data processing for grid updates.

[0074] Further, in an embodiment, the region of interest may include a first grid region and a second grid region, wherein the first grid region is a plurality of arcs within the field of view of the ultrasonic sensor with the ultrasonic sensor as the center and the distance at which the ultrasonic sensor can detect obstacles as the radius, and the second grid region is the region between the ultrasonic sensor and each arc.

[0075] Specifically, during the ultrasonic mapping process, the ultrasonic sensor scans obstacles within a certain range, that is, within the field of view. When an obstacle is detected, the specific location of the obstacle cannot be determined, but the obstacle can be determined to exist at a certain distance from the ultrasonic sensor. For example, the arc where the obstacle is located can be determined based on the field of view of the sensor and the distance from the detected obstacle. Therefore, the points on the arc are considered to be obstacle points, and the points on the line connecting the ultrasonic sensor to the arc are considered to be non-obstacle points. That is to say, in the embodiment, the obstacle points correspond to the points in the first grid area, the non-obstacle points correspond to the points in the second grid area, the probability update parameters of the first grid area and the second grid area are different, and the probability update parameters of different positions in the first grid area and the second grid area are also different.

[0076] Furthermore, in an embodiment, multiple arcs in the first grid area in the region of interest are arranged at intervals along the radial direction of the field of view of the ultrasonic sensor, and the arc length of the middle position of the multiple arcs in the radial direction is longer than the arc length of the arcs at the two end positions. In this way, the region of interest can be set based on the characteristics of the ultrasonic sensor in detecting the distance of obstacles, which can not only ensure the accuracy of obstacle detection but also reduce the data processing amount of grid update, thereby improving the update efficiency.

[0077] For example, Figure 2 and Figure 3 As shown in the figure, the outermost contour is the set ultrasonic shape, and the arc is determined according to the set compensation. According to different detection distances, the coordinates of all points on the corresponding arc can be calculated. For any arc, the point on the arc is considered to be an obstacle point, i.e., the hit point, and the point between the origin, i.e., the ultrasonic sensor position, and the arc is considered to be a non-obstacle point, i.e., the miss point.

[0078] Among them, the larger the probability update parameter of the hit point, the higher the update efficiency of the corresponding grid point, and the smaller the probability update parameter of the miss point, the higher the update efficiency of the corresponding grid point.

[0079] The following is a description of the probability update parameters for obtaining the target point according to the embodiment of the present invention.

[0080] In an embodiment, the probability update parameters of the target point can be obtained through a normal distribution probability model. The target point can be any point in the area of ​​interest in the field of view of the ultrasonic sensor. The probability update parameters of the target point can correspond to the probability update parameters of the obstacle point or the probability update parameters of the non-obstacle point. The type of the probability update parameters of the target point is determined according to the obstacle attribute of the target point. For example, if the target point is a point on an arc within the field of view of the ultrasonic sensor, the probability update parameters of the target point correspond to the probability update parameters of the obstacle point; or, if the target point is a point on a line connecting the origin and the arc within the field of view of the ultrasonic sensor, the probability update parameters of the target point correspond to the probability update parameters of the non-obstacle point.

[0081] In some embodiments, the normal distribution probability model uses the target point angle as input, the target point angle is the angle between the target point line and the upper radial line of the ultrasonic sensor direction, and the target point line is the line connecting the target point and the ultrasonic sensor. The target point angle can be determined according to the position information of the target point, for example, the ultrasonic sensor is the origin, the ultrasonic sensor direction is determined, and the target point angle can be determined according to the position information of the target point.

[0082] Furthermore, the normal distribution probability model may include a normal distribution equation and a first mapping relationship, wherein the first mapping relationship is a mapping relationship between the target point angle and the horizontal coordinate value of the normal distribution equation. After obtaining the target point angle, the horizontal coordinate value of the normal distribution equation corresponding to the target point angle may be obtained according to the first mapping relationship.

[0083] For example, Figure 4 As shown, we can establish an expectation of 0 and a variance of σ 2 The normal distribution equation can be expressed as follows:

[0084]

[0085] The equation is x α and σ are input values, the ordinate value f1(x α )=f1(-x α )=y α , f1(0)=y max .

[0086] like Figure 5 As shown, for each set of points, c1 and c2 are the two edge points of the arc, B is the center point of the arc, A is any point on the arc, and 2β is the angle of the sector, i.e., the field of view of the ultrasonic sensor, i.e., ∠c1oc2. Taking point A as the target point, L α It is the angle between AO and OB, that is, the target point angle. The left deviation of OB can be set to negative and the right deviation to positive.

[0087] Furthermore, we can establish Lα The mapping relationship with the horizontal coordinate value of the normal distribution equation is the first mapping relationship. The angle α between the radial line of the ultrasonic sensor and the obstacle line is taken as the reference angle. Here, relative to the ultrasonic sensor direction, the left deviation is negative and the right deviation is positive. For each point on the arc, that is, the obstacle point, the corresponding L can be calculated. α , if L α is greater than α, then according to (α, 0) and (β, x α ) Establish L α The mapping relationship with the abscissa value of the normal distribution equation is the first mapping relationship.

[0088] In some embodiments, at the target point angle L α When it is greater than the reference angle, the first mapping relationship can be expressed as:

[0089]

[0090] Among them, X is the horizontal coordinate value of the normal distribution equation, L α is the target point angle, the value range of X is 0 to X β Between, L α The value of is between α and β. When X = 0, L α =α, X = X β Time α =β, α is the reference angle, β is the half angle of the ultrasonic sensor’s field of view, X β is the horizontal coordinate value of the normal distribution equation corresponding to the half angle of the field of view.

[0091] In the embodiment, the reference angle may be the angle between the radial line of the ultrasonic sensor and the obstacle line, where the obstacle line is the line between the ultrasonic sensor and the obstacle position. Of course, the reference angle may also be other set angles.

[0092] In other embodiments, if L α is less than α, then according to (α, 0) and (-β, -x α ) establishes a first mapping relationship. For example, when the target point angle is smaller than the reference angle, the first mapping relationship can be expressed as:

[0093]

[0094] Among them, X is the horizontal coordinate value of the normal distribution equation, L α is the target point angle, the value range of X is 0 to (-X β ), L α The value of is between α and (-β), when X = 0, L α =α, X = -X β Time α=-β, α is the reference angle, β is the half angle of the ultrasonic sensor's field of view, X β is the abscissa value of the normal distribution equation corresponding to the half angle of the ultrasonic sensor field of view.

[0095] In the embodiment, the orientation of the ultrasonic sensor and the position of the obstacle can be determined based on the visual information of the vehicle environment collected by the on-board visual device. For example, the position of the obstacle can be determined based on the image of the vehicle surrounding environment collected by the on-board camera, and when the relative positions of the camera and the ultrasonic sensor are known, the line connecting the ultrasonic sensor and the obstacle can be determined, and then the reference angle can be obtained. Of course, the obstacle position and the orientation of the ultrasonic sensor can also be determined based on the data of other on-board visual devices.

[0096] Further, after obtaining the abscissa value of the normal distribution equation, the ordinate value of the normal distribution equation can be obtained by calculating the normal distribution equation according to the abscissa value of the normal distribution equation. α Substituting into the above normal distribution equation can obtain the corresponding ordinate value.

[0097] Furthermore, in an embodiment, the normal distribution probability model may further include a second mapping relationship, which is a correspondence between the ordinate value of the normal distribution equation and the probability update parameter. After obtaining the ordinate value of the normal distribution equation, the probability update parameter of the target point corresponding to the obstacle attribute can be obtained by searching the second mapping relationship according to the ordinate value of the normal distribution equation.

[0098] In the embodiment, for establishing the second mapping relationship, for example, referring to Figure 5 As shown, assuming that the probability update parameters corresponding to c1 and c2 are 0.5, in some embodiments, the probability update parameter of the unupdated grid can be set to 0.5 by default, and the probability update parameter corresponding to point B can be set to p B , for the hit point, p B is a number greater than 0.5. For the miss point p B is a number less than 0.5, which can be input manually and can be calculated based on (y max , p B ) and (y α , 0.5) to establish a second mapping relationship.

[0099] For example, in some embodiments, the second mapping relationship may be expressed as follows:

[0100]

[0101] Among them, P is the probability update parameter, y is the ordinate value of the normal distribution equation, P BUpdate the parameter y for the probability of setting the known point B. max P B The ordinate value of the corresponding normal distribution equation, y α is the ordinate value of the normal distribution equation corresponding to the half angle of the field of view.

[0102] Further, in the embodiment, the obstacle attributes include obstacle points and non-obstacle points; when point B is an obstacle point, P B If it is greater than 0.5, the probability of judging the grid point as an obstacle will increase with each update of the corresponding grid point, and the maximum value shall not exceed 1. Alternatively, when point B is a non-obstacle point, P B If it is less than 0.5, the probability of judging the grid point as an obstacle will decrease each time the corresponding grid point is updated, and the minimum value shall not be less than 0.

[0103] From the above, the probability update parameters corresponding to the target point can be obtained, and then the obstacle judgment probability of the target point can be updated according to the probability update parameters.

[0104] In summary, for any point in the field of view of the ultrasonic sensor, the corresponding L can be calculated based on the coordinates of the point in the sensor coordinate system. α . Then, according to L α The first mapping relationship with the abscissa of the normal distribution equation is used to calculate the angle L α The horizontal coordinate value of the normal distribution equation corresponding to the normal distribution equation is calculated; the corresponding vertical coordinate value is calculated according to the normal distribution equation; and, according to the second mapping relationship between the probability update parameter and the vertical coordinate value of the normal distribution equation, the probability update parameter corresponding to any point in the field of view of the ultrasonic sensor can be calculated. For example, it can be p hit or miss .

[0105] Specifically, when updating the obstacle probability of the grid points, for any point within the field of view of the ultrasonic sensor, the corresponding probability update parameter p is obtained according to its coordinates in the sensor coordinate system and its obstacle attributes, that is, whether the point is a hit point or a miss point. point (p hit 、p miss ), and then can be updated as follows:

[0106]

[0107] M new (x) = clamp(odds -1 (odds(M old (x))·odds(p point )));

[0108] Among them, ppoint For p hit or miss , according to the obstacle attribute of the point, determine that the point is an obstacle point. For example, if the point is located on the arc, then p point For p hit , otherwise, p point For p miss . M new (x) is the final result of the update at this point, M old (x) is the result of the last update of this point. In the embodiment, the final result of the update (the probability of judging it as an obstacle) is not more than 1 at most and not less than 0 at least.

[0109] Based on the grid map updating method of the above embodiment, Figure 6 The following is the effect diagram of ultrasonic sensor mapping when α=0. Figure 7 The figure shows the effect of ultrasonic sensor mapping when α<0. It can be clearly seen that the update effect in the direction of the obstacle is more obvious. Figure 8 and 9 The figure shows the mapping result, and the black area is the obstacle outline.

[0110] Specifically, when the vehicle is driving slowly, a local map is constructed based on the ultrasonic data, and the obstacle boundary is extracted based on the constructed map. The direction of the ultrasound and the angle between the ultrasound and the obstacle can be determined based on the visual detection results, and the p of each point within the ultrasonic field of view can be calculated based on the normal distribution probability model. hit and p miss , and the update efficiency of all points on the arc where the obstacle point is located is different. The smaller the angle, the higher the update efficiency, that is, the probability update efficiency of the obstacle direction is the highest. When updating the grid probability, the corresponding probability update parameter (p hit 、p miss ), used for probability update.

[0111] In the embodiment of the present invention, the probability update parameters in the mapping process are calculated based on the model, rather than the default same value, and the update efficiency of different positions is different. It is considered that the probability update parameter of the direction the sensor is facing is the largest (hit) or the smallest (miss), that is, the update efficiency is the highest, and the two sides are reduced (hit) or increased (miss) in turn based on the normal distribution model.

[0112] Therefore, when the vehicle passes in front of an obstacle and the ultrasonic sensor scans the obstacle, different probability update parameters can be obtained through the normal distribution probability model at different positions within the scanning range of the ultrasonic sensor, thereby affecting the efficiency of its probability update, that is, the update efficiency of the ultrasonic sensor facing the direction is the highest, and the update efficiency on both sides decreases successively. For the hit point, the update efficiency of all points on the arc where it is located is different. The update efficiency in the direction of the obstacle is high and can be quickly determined as the hit point. At the edge of the obstacle, as the vehicle moves, the grid where the hit point in the direction of the obstacle is located will be quickly blackened due to the higher update efficiency, while the two sides will not, so it can effectively limit the enlargement of the obstacle edge and improve the accuracy of obstacle detection. In addition, the probability update effect can be adjusted by modifying the parameters to improve adaptability.

[0113] Based on the grid map updating method of the above embodiment, a second aspect of the present invention provides an electronic device.

[0114] Fig.10 is a block diagram of an electronic device according to an embodiment of the present invention, such as Fig.10 As shown, the electronic device 10 includes at least one processor 11 and a memory 12 .

[0115] The memory 12 is in communication connection with at least one processor 11. The memory 12 stores a computer program executable by at least one processor 11. When at least one processor executes the computer program, the grid map updating method of the above embodiment is implemented.

[0116] In an embodiment, the electronic device 10 may be a vehicle-mounted controller such as a vehicle controller or a domain controller or a vehicle-mounted electronic device or a mobile device or an electronic terminal, and may be other devices that need to update a grid map based on ultrasonic mapping, without specific limitation herein.

[0117] According to the electronic device 10 of the embodiment of the present invention, at least one processor 11 executes the grid map updating method of the above embodiment, and the grid points at various positions use different probability update parameters, which can effectively limit the enlargement of the obstacle edge and improve the obstacle detection accuracy.

[0118] A third aspect of the present invention provides a non-volatile readable storage medium having a computer program stored thereon. When the computer program is executed, the grid map updating method of the above embodiment can be implemented.

[0119] In an embodiment, the non-volatile readable storage medium may be a non-transitory storage medium, for example, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes, or may be a transient storage medium.

[0120] Based on the grid map updating method of the above embodiment, the fourth aspect of the present invention provides an automatic driving assistance system, such as Fig.11 As shown, the automatic driving assistance system 100 includes an ultrasonic sensor 101 and a controller 102 .

[0121] The ultrasonic sensor 101 is used to collect obstacle information within the field of view, such as identifying whether there are obstacles within the scannable range around the vehicle and determining the distance of the obstacles.

[0122] The controller 102 is connected to the ultrasonic sensor 101 and is used to construct a grid map according to obstacle information and update the grid map according to the grid map updating method of the above embodiment.

[0123] According to the automatic driving assistance system 100 of the embodiment of the present invention, a grid map is constructed based on ultrasound, and the controller 102 performs grid probability update according to the grid map update method of the above embodiment. Different probability update parameters are used for the grid points at each position, which can effectively limit the enlargement of the obstacle edge and improve the obstacle detection accuracy.

[0124] In some embodiments, the automated driving assistance system 100 further includes an onboard visual device, which is connected to the controller 102 and is used to obtain visual information of the vehicle environment; the controller 102 is also used to determine the direction of the ultrasonic sensor and the position of the obstacle based on the visual information. For example, the onboard visual device may include a camera, etc.

[0125] A fifth aspect of the present invention further provides a vehicle, such as Fig.12 As shown, the vehicle 1 includes the electronic device 10 of the above embodiment, or, as Fig.13 As shown, the vehicle 1 includes the automatic driving assistance system 100 of the above embodiment.

[0126] According to the vehicle 1 of the embodiment of the present invention, the grid map is updated through the electronic device 10 or the automatic driving assistance system 100 of the above embodiment. The grid points at each position use different probability update parameters, which can effectively limit the enlargement of the obstacle edge and improve the obstacle detection accuracy.

[0127] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example.

[0128] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A grid map updating method, characterized in that: include: Obtaining location information of a target point and obstacle attributes of the target point, wherein the target point is a point in a grid map corresponding to a field of view angle range of the ultrasonic sensor; Obtaining a probability update parameter of the target point according to the position information and the obstacle attribute; Update the obstacle judgment probability of the target point according to the probability update parameter; When there is an obstacle within the field of view of the ultrasonic sensor, the updating efficiency of the probability updating parameter is the highest in the direction where the obstacle is located, and the updating efficiency of the probability updating parameter decreases successively on both sides away from the direction where the obstacle is located.

2. The grid map updating method according to claim 1, characterized in that: When the obstacle does not exist within the field of view of the ultrasonic sensor, the updating efficiency of the probability updating parameters in the direction of the ultrasonic sensor is the highest, and the updating efficiency of the probability updating parameters decreases successively on both sides away from the direction of the ultrasonic sensor.

3. The grid map updating method according to claim 1, characterized in that: Obtaining a probability update parameter of the target point according to the position information and the obstacle attribute information includes: The probability update parameter of the target point is obtained by a normal distribution probability model, the type of the probability update parameter is determined according to the obstacle attribute of the target point, the normal distribution probability model takes the target point angle as input, the target point angle is the angle between the target point line and the upper radial line of the ultrasonic sensor, and the target point line is the line between the target point and the ultrasonic sensor; Wherein, the target point angle is determined according to the position information.

4. The grid map updating method according to claim 3, characterized in that: The normal distribution probability model includes a normal distribution equation and a first mapping relationship, wherein the first mapping relationship is a mapping relationship between the target point angle and the abscissa value of the normal distribution equation; Obtaining the probability update parameter of the target point corresponding to the obstacle attribute through the normal distribution probability model includes: obtaining the horizontal coordinate value of the target point angle corresponding to the normal distribution equation according to the first mapping relationship.

5. The grid map updating method according to claim 4, characterized in that: When the target point angle is greater than the reference angle, the first mapping relationship is expressed as: Wherein, X is the horizontal coordinate value of the normal distribution equation, L α is the target point angle, and the value range of X is from 0 to X β Between, L α The value of is between α and β. When X = 0, L α =α, X = X β When Lα=β, α is the reference angle, β is the half angle of the field of view of the ultrasonic sensor, and X β is the abscissa value of the normal distribution equation corresponding to the half angle of the field of view.

6. The grid map updating method according to claim 4, characterized in that: When the target point angle is smaller than the reference angle, the first mapping relationship is expressed as: Wherein, X is the horizontal coordinate value of the normal distribution equation, L α is the target point angle, and the value range of X is from 0 to (-X β ), L α The value of is between α and (-β), when X = 0, L α =α, X = -X β Time α =-β, α is the reference angle, β is the half angle of the field of view of the ultrasonic sensor, X β is the abscissa value of the normal distribution equation corresponding to the half angle of the field of view.

7. The grid map updating method according to claim 5 or 6, characterized in that: The reference angle is the angle between the radial line toward which the ultrasonic sensor is directed and the obstacle line, and the obstacle line is the line from the ultrasonic sensor to the obstacle.

8. The grid map updating method according to claim 7, characterized in that: The orientation of the ultrasonic sensor and the position of the obstacle are determined based on the vehicle environment visual information collected by the vehicle-mounted visual device.

9. The grid map updating method according to claim 4, characterized in that: Obtaining a probability update parameter of the target point corresponding to the obstacle attribute through a normal distribution probability model, further comprising: The ordinate value of the normal distribution equation is obtained through the normal distribution equation according to the abscissa value of the normal distribution equation.

10. The grid map updating method according to claim 9, characterized in that: The normal distribution probability model further includes a second mapping relationship, wherein the second mapping relationship is a corresponding relationship between the ordinate value of the normal distribution equation and the probability update parameter; Obtaining the probability update parameter of the target point corresponding to the obstacle attribute through the normal distribution probability model also includes: searching the second mapping relationship according to the ordinate value of the normal distribution equation to obtain the probability update parameter of the target point corresponding to the obstacle attribute.

11. The grid map updating method according to claim 10, characterized in that: The second mapping relationship is expressed as: Wherein, P is the probability update parameter, y is the ordinate value of the normal distribution equation, P B Update the parameter y for the probability of setting the known point B. max P B The ordinate value of the corresponding normal distribution equation, y α is the ordinate value of the normal distribution equation corresponding to the half angle of the field of view.

12. The grid map updating method according to claim 11, characterized in that: The obstacle attributes include obstacle points and non-obstacle points; When the point B is the obstacle point, the P B greater than 0.5, or when point B is the non-obstacle point, the P B Less than 0.

5.

13. The grid map updating method according to claim 1, characterized in that: The target point is any point in the region of interest within the field of view of the corresponding ultrasonic sensor.

14. The grid map updating method according to claim 13, characterized in that: The region of interest includes a first grid region and a second grid region; The first grid area is a plurality of arcs with the ultrasonic sensor as the center and the distance at which the ultrasonic sensor can detect obstacles as the radius within the field of view of the ultrasonic sensor; The second grid area is an area between the ultrasonic sensor and each of the circular arcs.

15. The grid map updating method according to claim 13, characterized in that: The obstacle attributes include obstacle points and non-obstacle points; The obstacle points correspond to points in the first grid area, and the non-obstacle points correspond to points in the second grid area.

16. The grid map updating method according to claim 14, characterized in that: The plurality of circular arcs in the first grid area are arranged at intervals along the radial direction of the field of view of the ultrasonic sensor, and the arc length of the circular arc at the middle position of the plurality of circular arcs in the radial direction is longer than the arc length of the circular arcs at the two end positions.

17. An electronic device, characterized in that: include: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and when the at least one processor executes the computer program, the grid map updating method according to any one of claims 1 to 16 is implemented.

18. A non-volatile readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the grid map updating method according to any one of claims 1 to 16 is implemented.

19. An automatic driving assistance system, characterized in that: include: Ultrasonic sensor, used to collect obstacle information within the field of view; A controller connected to the ultrasonic sensor, configured to construct a grid map according to the obstacle information, and to update the grid map according to the grid map updating method according to any one of claims 1 to 16.

20. The automatic driving assistance system according to claim 19, characterized in that: The automatic driving assistance system further includes: An on-vehicle visual device, connected to the controller, for obtaining visual information of the vehicle environment; The controller is also used to determine the direction of the ultrasonic sensor and the position of the obstacle according to the visual information.

21. A vehicle, characterized in that: The vehicle comprises the electronic device of claim 17; Alternatively, the vehicle includes the autonomous driving assistance system of claim 19 or 20.

Images