Intersection parking method, device and equipment and computer readable storage medium

Abstract

The invention discloses an intersection parking method, device and equipment and a computer readable storage medium, and is applied to the field of intelligent driving, and the method comprises the steps: determining a target traffic light according to a traffic light recognition result and a traffic light decision rule; when the target traffic light is red and the stop line is not recognized, obtaining a predicted distance between the vehicle and the stop line according to a stop line prediction rule, and performing speed planning based on the predicted distance; and when the target traffic light is red and the stop line is identified, acquiring a real distance between the vehicle and the stop line, and carrying out braking processing based on the real distance. Therefore, according to the method, when the target traffic light is the red light, the stop line is predicted in advance, speed planning is carried out, the safety and reliability of parking at the intersection are guaranteed, dependence on a map is avoided, dependence on visual perception is low, and safe braking at the intersection can be achieved in a special environment or under the working condition that the stop line is fuzzy.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving, and in particular to a method, device, equipment and computer-readable storage medium for parking at an intersection. Background Art

[0002] When the existing technology recognizes that the traffic light is red, it needs to rely on maps or visual perception to recognize the stop line and then take measures to stop the car. However, maps or visual perception have certain limitations and are easily affected by the external environment. Therefore, it is difficult to recognize the stop line based on visual perception, and when the stop line is recognized and the car is slowed down and stopped, it may be too late, which poses certain safety risks.

[0003] Therefore, how to provide a safe and reliable intersection parking method is a technical problem that urgently needs to be solved. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a method, device, equipment and computer-readable storage medium for parking at an intersection, which solve the problem of unsafe and unreliable parking at an intersection in the prior art.

[0005] In order to solve the above technical problems, the present invention provides a method for parking at an intersection, comprising:

[0006] Determine the target traffic light according to the traffic light recognition result and the traffic light decision rule;

[0007] When the target traffic light is red and the stop line is not identified, a predicted distance between the vehicle and the stop line is obtained according to a stop line prediction rule, and speed planning is performed based on the predicted distance;

[0008] When the target traffic light is red and the stop line is identified, the real distance between the vehicle and the stop line is obtained, and a brake stop process is performed based on the real distance.

[0009] Optionally, determining the target traffic light according to the traffic light recognition result and the traffic light decision rule includes:

[0010] When a traffic light is identified and works normally, the traffic light is the target traffic light;

[0011] When multiple traffic lights are identified, the target traffic light is determined according to the location of the traffic light, the compliance of other vehicles with the traffic light, and whether the traffic light is working properly.

[0012] Optionally, determining the target traffic light according to the location of the traffic light, the condition of other vehicles following the traffic light, and whether the traffic light is working normally includes:

[0013] Obtaining weights of the traffic light position, other vehicles' compliance with the traffic light, and whether the traffic light is working properly; the weight of whether the traffic light is working properly is greater than the weight of the traffic light position, and the weight of the traffic light position is greater than the weight of other vehicles' compliance with the traffic light;

[0014] The scores of the traffic lights are obtained according to the weights, and the traffic light corresponding to the highest score is used as the target traffic light.

[0015] Optionally, obtaining the predicted distance between the vehicle and the stop line according to the stop line prediction rule includes:

[0016] The predicted distance between the vehicle and the stop line is predicted based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection and the movement trajectory of the preceding vehicle.

[0017] Optionally, the predicted distance between the vehicle and the stop line is predicted based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle, including:

[0018] Based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection and the movement trajectory of the preceding vehicle, each initial prediction distance is obtained;

[0019] Obtaining the respective probabilities of the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle, wherein the sum of the respective probabilities is 1;

[0020] The predicted distance is obtained according to the initial predicted distances and the probabilities.

[0021] Optionally, the calculating the predicted distance according to the respective initial predicted distances and the respective probabilities includes:

[0022] If the difference between the initial prediction distances is greater than or equal to a preset threshold, the initial prediction distance corresponding to the maximum probability is used as the prediction distance;

[0023] If there is a difference value less than the preset threshold, the target probability and the target initial prediction distance corresponding to the difference value less than the preset threshold are obtained, the new probability is recalculated according to the target probability, and the initial prediction distance corresponding to the new probability is calculated according to the target initial prediction distance; the various probabilities are compared with the new probability, and the initial prediction distance corresponding to the maximum probability is used as the prediction distance.

[0024] Optionally, performing speed planning based on the predicted distance includes:

[0025] determining a driving distance constraint based on the predicted distance;

[0026] Each preliminary speed plan is obtained based on different single time steps and search rules;

[0027] Calculate the loss value of each initial speed plan based on the driving distance constraint;

[0028] The preliminary speed plan corresponding to the minimum loss value is used as the speed plan.

[0029] The present invention also provides a parking device at an intersection, comprising:

[0030] A target traffic light determination module, used for determining a target traffic light according to a traffic light recognition result and a traffic light decision rule;

[0031] A first speed planning module, configured to obtain a predicted distance between the vehicle and the stop line according to a stop line prediction rule when the target traffic light is red and the stop line is not identified, and perform speed planning based on the predicted distance;

[0032] The second speed planning module is used to obtain the real distance between the vehicle and the stop line when the target traffic light is red and the stop line is identified, and perform a braking process based on the real distance.

[0033] The present invention also provides a parking device at an intersection, comprising:

[0034] Memory for storing computer programs;

[0035] A processor is used to implement the steps of the above-mentioned intersection parking method when executing the computer program.

[0036] The present invention also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, the steps of the above-mentioned intersection parking method are implemented.

[0037] It can be seen that the present invention determines the target traffic light according to the traffic light recognition result and the traffic light decision rule; when the target traffic light is a red light and the stop line is not recognized, the predicted distance between the vehicle and the stop line is obtained according to the stop line prediction rule, and the speed is planned based on the predicted distance; when the target traffic light is a red light and the stop line is recognized, the real distance between the vehicle and the stop line is obtained, and the brake is stopped based on the real distance. It can be seen that when the target traffic light is a red light, the method predicts the stop line in advance and performs speed planning to ensure the safety and reliability of parking at the intersection, and the target traffic light in the present invention is determined according to the traffic light recognition result and the traffic light decision rule, without relying on the map to determine the target traffic light, and the present invention uses the stop line prediction rule to determine the stop line position, without relying on the map to obtain the stop line position; and compared with relying only on visual perception to identify the stop line position, the present invention also provides a method for obtaining the stop line position based on the stop line prediction rule, which reduces the dependence on visual perception, so that it can be safely braked at the intersection under special circumstances or when the stop line is blurred.

[0038] In addition, the present invention also provides a method, device and computer-readable storage medium for parking at an intersection, which also have the above-mentioned beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0040] Figure 1 A flow chart of a method for parking at an intersection provided by an embodiment of the present invention;

[0041] Figure 2 An example diagram of speed planning provided by an embodiment of the present invention;

[0042] Figure 3 Another example diagram of speed planning provided by an embodiment of the present invention;

[0043] Figure 4 A schematic diagram of the structure of a parking device at an intersection provided by an embodiment of the present invention;

[0044] Figure 5 A schematic structural diagram of an intersection parking device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Please refer to Figure 1 , Figure 1 A flow chart of a method for parking at an intersection provided by an embodiment of the present invention. The method may include:

[0047] S101: Determine a target traffic light according to a traffic light recognition result and a traffic light decision rule.

[0048] The executor of this embodiment is a terminal. This embodiment does not limit the type of terminal, as long as it can complete the operation of the intersection parking method. During the driving process of the vehicle, the camera can collect / identify the traffic light situation and obtain traffic light information such as the location of the traffic light and whether the traffic light is working properly. When the distance between intersections is close, since the camera can stably identify the cone range within about 120 meters in front, it is possible that multiple traffic lights are identified. At this time, it is necessary to determine the target traffic light according to the traffic light decision rule. It can be understood that the target traffic light is the traffic light that is closest to the vehicle and is working normally. This embodiment does not limit the traffic light decision rule. As long as it is a decision rule that can accurately determine the target traffic light.

[0049] Furthermore, in order to improve the accuracy of determining the target traffic light, the above-mentioned determination of the target traffic light according to the traffic light recognition result and the traffic light decision rule may include the following steps:

[0050] Step 11: When a traffic light is identified and works normally, the traffic light is the target traffic light;

[0051] Step 12: When multiple traffic lights are identified, the target traffic light is determined based on the location of the traffic light, the compliance of other vehicles with the traffic light, and whether the traffic light is working properly.

[0052] When a traffic light is identified and the traffic light is working properly, it can be determined that the traffic light is the target traffic light. When multiple traffic lights are identified, the target traffic light can be determined based on the traffic light location, other vehicles following the traffic light, and whether the traffic light is working properly. These traffic light decision conditions are used to determine the target traffic light. The condition based on other vehicles following the traffic light is to ensure that when multiple traffic lights appear at a five-way intersection and meet the requirements at the same time, other vehicles are needed to determine which traffic light rule should be followed (i.e., the target traffic light); the condition based on the traffic light location is because there are many types of traffic lights, which may be regular traffic lights, temporary traffic lights, or roadside auxiliary traffic lights, so multiple situations need to be considered. The normal operation of traffic lights is a basic condition. This embodiment does not limit the specific traffic light decision rules formulated based on the above-mentioned traffic light decision conditions.

[0053] Further, in order to improve the accuracy of determining the target traffic light, the above step 12, when multiple traffic lights are identified, determines the target traffic light according to the location of the traffic light, the condition of other vehicles following the traffic light, and whether the traffic light is working properly, and may include the following steps:

[0054] Step 121: Obtain weights of the traffic light position, other vehicles following the traffic light, and whether the traffic light is working properly; the weight of whether the traffic light is working properly is greater than the weight of the traffic light position, and the weight of the traffic light position is greater than the weight of other vehicles following the traffic light;

[0055] Step 122: Obtain the scores of the various traffic lights according to the weights, and use the traffic light corresponding to the highest score as the target traffic light.

[0056] In this embodiment, the weight can be expressed by setting a score, or by a weight coefficient. For example, a corresponding score is set for the above traffic light decision condition, the total score is 100 points, and the traffic light with the highest score is used as the target traffic light; or a corresponding weight coefficient can be set for the above traffic light decision condition, and the sum of the weight coefficients is 1. In order to better understand step 121 and step S122, the following example can be referred to: score each traffic light according to the location of the traffic light, the compliance of other vehicles with the traffic light, and whether the traffic light is working properly. The full score for the location of the traffic light is 30 points, the full score for the compliance of other vehicles with the traffic light is 20 points, and the full score for the compliance of the traffic light with 50 points, and the total score is 100 points. It should be noted that when there is no car on the current road or the lane situation of other vehicles is unclear, the scoring of the compliance of other vehicles with the traffic light is not triggered, that is, in the above two cases, the score of the compliance of other vehicles with the traffic light is assigned 0 points.

[0057] Specifically, the specific process of scoring based on the traffic light position can refer to the following example:

[0058] When the height z value of the traffic light is greater than the preset height (such as 6m), it means that the traffic light is not a temporary traffic light, but a regular traffic light, and a value of 10 is assigned at this time; when the y value of the traffic light is within the lane, it means that the traffic light is not an auxiliary traffic light, but a regular traffic light, and a value of 10 is assigned at this time; when the traffic light is "always on", a value of 10 is assigned. This condition is to prevent the two intersections from being too close to each other and the vehicle from mistakenly identifying the traffic light at the second intersection when it is too close to the first intersection. Among them, "always on" can be the timer starting from the first time the traffic light is recognized, and when the continuous appearance time is greater than a certain value, it means that it is always on; or according to the first recognized traffic light position x, combined with the displacement of the vehicle in the time t, when the subsequent calculation of the distance difference with the traffic light in each time (displacement-x) is within a certain range (such as 15 meters), it means that it is always on. It should be noted that the x, y, and z of the traffic light are represented in three-dimensional space. According to x, the distance between the vehicle and the traffic light can be obtained; according to y, the distance between the traffic light and the lane can be obtained; according to z, the height of the traffic light can be obtained.

[0059] Specifically, the specific process of scoring based on other vehicles' compliance with traffic lights can refer to the following example:

[0060] When there are other vehicles on the current road, the other vehicles are judged whether they have passed according to the lanes in which they are located and the perceived traffic light status. When the vehicle status is consistent with the traffic light status, 20 points are assigned, otherwise 0 points (or 20 points can also be subtracted). Specifically, the vehicle status judgment and the vehicle trajectory prediction process include: when the z value of the traffic light is greater than or equal to the preset height (such as 6 meters), it means that the traffic light is a regular traffic light, and it is judged whether it will invade the first distance range before the traffic light x (such as the first 40 meters); when the z value of the traffic light is less than 6 meters, it means that the traffic light is a temporary traffic light, and it is judged whether it will invade the second distance range of the traffic light x (the first 10 meters). It can be understood that there is a difference in the position of the temporary traffic light and the regular traffic light, so the second distance range is smaller than the first distance range. If it invades, it means that the vehicle status is green light to pass, otherwise it means that the red light is prohibited from passing. The 20-point maximum score for other vehicles' compliance with traffic lights should be strictly less than the 30-point maximum score for traffic light position, in order to avoid the situation where some vehicles do not start at the green light and thus affect the vehicle itself. At the same time, it should not be less than the full score of z and y in the traffic light position judgment, because at a five-way intersection, two traffic lights may meet the requirements at the same time, so it is necessary to use the help of other vehicles to determine which traffic light rule should be followed.

[0061] Specifically, the specific process of scoring based on whether the traffic light is working properly can refer to the following example:

[0062] The traffic light is assigned 50 points when it is on and 0 points when it is off.

[0063] S102: When the target traffic light is red and the stop line is not recognized, the predicted distance between the vehicle and the stop line is obtained according to the stop line prediction rule, and speed planning is performed based on the predicted distance.

[0064] When the target traffic light is determined in step S101, the traffic light that the vehicle needs to follow is determined. When the target traffic light is a red light, a brake stop process is required. If the stop line has not been identified at this time, the stop line position can be predicted according to the stop line prediction rule, and the predicted distance can be obtained based on the predicted stop line position. Speed ​​planning is performed based on the predicted distance, and driving is carried out according to the speed planning. This avoids the unsafe problem of excessive speed and rapid brake stop when the stop line is identified. This embodiment does not linearly perform the stop line prediction rule, as long as the stop line position is predicted by the prediction method.

[0065] Furthermore, in order to improve the accuracy of stop line prediction, the above-mentioned method of obtaining the predicted distance between the vehicle and the stop line according to the stop line prediction rule may include the following steps:

[0066] Step 21: Predict the predicted distance between the vehicle and the stop line based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle.

[0067] It should be noted that this embodiment predicts the stop line position from many aspects. The distance between the vehicle and the stop line predicted based on the target traffic light position definitely exists, while other predicted distances may not be predicted because they have prerequisites, such as the prerequisite for the auxiliary traffic light position is that the auxiliary traffic light is identified in step S101; the prerequisite for the motion trajectory of pedestrians at the intersection is that there are pedestrians at the intersection; the prerequisite for the motion trajectory of non-motor vehicles at the intersection is that there are non-motor vehicles at the intersection; the prerequisite for the position of stationary vehicles at the intersection is that there are stationary vehicles at the intersection; the prerequisite for the motion trajectory of the front vehicle is that there is a moving front vehicle.

[0068] Further, in order to improve the accuracy of stop line prediction, the above step 21, predicting the predicted distance between the vehicle and the stop line based on the target traffic light position, the auxiliary traffic light position, the pedestrian movement trajectory at the intersection, the non-motor vehicle movement trajectory at the intersection, the stationary vehicle position at the intersection and the movement trajectory of the preceding vehicle, may include the following steps:

[0069] Step 211: Predict each initial prediction distance based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle;

[0070] Step 212: Obtain the respective probabilities of the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle, where the sum of the respective probabilities is 1;

[0071] Step 213: Calculate the predicted distance based on each initial predicted distance and each probability.

[0072] It is understandable that if the prerequisite is not met, the predicted initial prediction distance can be determined to be 0. For example, if there are no pedestrians at the intersection, the initial prediction distance obtained based on the prediction of the movement trajectory of pedestrians at the intersection is 0. The probability is based on whether the stop line prediction condition can accurately determine the stop line setting. The lower the probability, the lower the possibility of the stop line position determined by the stop line prediction condition. This embodiment does not limit each probability, and can be limited according to actual conditions. Specifically, each probability can be determined according to the table lookup method. The following examples can be referred to:

[0073] Table 1 Probability example table

[0074]

[0075] Furthermore, the above-mentioned obtaining of the corresponding initial predicted distance based on the target traffic light position may specifically include the following steps:

[0076] If the target traffic light is determined to be a regular traffic light according to its position, the first initial predicted distance is the horizontal coordinate value of the target traffic light minus the first distance;

[0077] If the target traffic light is determined to be a temporary traffic light according to its position, the first initial predicted distance is the horizontal coordinate of the target traffic light minus the second distance; the first distance is greater than the second distance.

[0078] Specifically, you can refer to the following example: if the z value of the target traffic light is greater than or equal to the preset height (such as z ≥ 6m), it is determined to be a regular traffic light, and the first initial predicted distance s1 is equal to the x value of the target traffic light minus 40m; if the z value of the target traffic light is less than the preset height (such as z < 6m), then the first initial predicted distance s1 is equal to the x value of the target traffic light minus 20m.

[0079] Furthermore, obtaining the corresponding initial predicted distance of the auxiliary traffic light position may specifically include the following steps:

[0080] Determine whether the auxiliary traffic light is within the preset range of the roadside;

[0081] If yes, then a second initial predicted distance is calculated based on the auxiliary traffic light position.

[0082] Specifically, the following example can be referred to: the preset range is set to 1.5m. Check whether there is a curb on the left side of the current road. If so, check whether there is an auxiliary traffic light within 1.5m on the left side of the curb (the auxiliary traffic light cannot be the target traffic light). If the traffic light exists within 1.5m on the left side of the curb, the second initial predicted distance s2 is equal to the x value of the auxiliary traffic light plus a preset value (such as 2m). The same is true for the right curb. Generally, only the auxiliary traffic light on one side of the curb can be detected. Among them, the distance calculation method between the auxiliary traffic light and the left curb can include two methods: the first method: use the sensor to obtain a series of scattered points on the curb, and a curve can be obtained by fitting the scattered points. The x value of the auxiliary traffic light is brought into the curve to obtain the y value on the curve. According to this y value and the y value of the auxiliary traffic light, the offset of the auxiliary traffic light relative to the curb can be obtained. Another method is to calculate the difference between the lateral displacement of the left curb and the vehicle at the current position of the vehicle, and then substitute the x value of the auxiliary traffic light into the equation of the left lane line of the vehicle to calculate the y value. Then, the y value minus the lateral displacement difference of the vehicle's starting position can be used to obtain the offset of the auxiliary traffic light relative to the curb. The offset is the distance.

[0083] Further, obtaining the corresponding initial predicted distance based on the movement trajectory of pedestrians at the intersection may specifically include the following steps:

[0084] A motion trajectory of a pedestrian at an intersection perpendicular to the current road is obtained, and a third initial predicted distance is calculated based on the motion trajectory of the pedestrian at the intersection.

[0085] It is understandable that pedestrians can only be detected at the intersection when there are no cars on the road, and this step is performed. Specifically, a straight line is fitted based on the motion trajectory of the pedestrian at the intersection. If the straight line and the motion trajectory of the vehicle satisfy the following relationship, the point with x1=0 on the motion trajectory of the pedestrian at the intersection is taken, and its y1 value is calculated. The third initial predicted distance (such as s3=y1-5m) is calculated based on y1. The relationship is as follows:

[0086] When the vehicle's trajectory is not perpendicular to the pedestrian's trajectory at the intersection, the slope ; The following conditions must be met: Angle Smaller than the preset angle, such as <-2.144 (115°), or the product of the slope of the first point of the vehicle's motion trajectory and the slope of the first point of the pedestrian's motion trajectory meets the preset conditions, such as m1m2=-1, or the slope Greater than 3.732 (85°);

[0087] When the vehicle's motion trajectory is perpendicular to the pedestrian's motion trajectory at the intersection, the angle is 90°.

[0088] in, represents the slope; is the slope of the first point of the vehicle motion trajectory fitting; It is the slope of the first point of the pedestrian motion trajectory at the intersection.

[0089] It should be noted that the method for obtaining the corresponding initial predicted distance based on the motion trajectory of non-motor vehicles at the intersection is similar to the method for obtaining the corresponding initial predicted distance based on the motion trajectory of pedestrians at the intersection. The motion trajectory of pedestrians at the intersection is changed to the motion trajectory of non-motor vehicles at the intersection, such as two-wheeled vehicles or three-wheeled vehicles. A straight line is fitted based on the motion trajectory of the non-motor vehicles at the intersection. If the straight line and the straight line of the vehicle's motion trajectory satisfy the above relationship, the point with x2=0 on the motion trajectory of the non-motor vehicle at the intersection is taken, and its y2 value is calculated. The fourth initial predicted distance (such as s4=y2-8m) is calculated based on y2.

[0090] Further, the above-mentioned obtaining the corresponding initial predicted distance based on the position of the stationary vehicle at the intersection may include the following steps:

[0091] When there is a stationary vehicle in the lane of the ego vehicle, the target stationary vehicle is determined according to the position of the stationary vehicle;

[0092] The first distance is obtained by projecting the position of the target stationary vehicle onto the center line equation of the vehicle lane;

[0093] A corresponding fifth initial predicted distance is obtained according to the position of the target stationary vehicle and the first distance.

[0094] Specifically, when there is a vehicle at the intersection, determine whether the vehicle is stationary. If it is stationary and in the own vehicle lane, take the vehicle with the largest x value among the stationary vehicles, that is, the vehicle closest to the own vehicle among the stationary vehicles as the target stationary vehicle, and project it onto the centerline equation of the own vehicle lane. The y3 value obtained at this time (that is, the first distance) is added to the preset distance (such as half the vehicle length of the target stationary vehicle) to obtain the fifth initial predicted distance, that is, s5=y3+0.5*vehicle length.

[0095] Further, obtaining the corresponding initial predicted distance based on the motion trajectory of the preceding vehicle may specifically include the following steps:

[0096] Obtain the trajectory of the vehicle in front of you on the lane, and predict the stopping position of the vehicle in front of you based on the trajectory of the vehicle in front of you;

[0097] A sixth initial predicted distance is obtained based on the stop position of the preceding vehicle.

[0098] Specifically, the preceding vehicle in this embodiment is the preceding vehicle closest to the vehicle in the lane of the vehicle. If the preceding vehicle is predicted to be stationary at the future position x based on the motion trajectory of the preceding vehicle, the sixth initial predicted distance s6=x+0.5*vehicle length. The process of determining the future stop position x can be determined according to the number of seconds displayed by the traffic light. For example, the number of seconds t displayed by the target traffic light is obtained. When the number of seconds t is less than a preset time (such as 10s), the stop position x after t seconds is found according to the motion trajectory of the preceding vehicle.

[0099] Furthermore, in order to improve the accuracy of stop line prediction, the above step 213, obtaining the predicted distance according to each initial predicted distance and each probability, may include the following steps:

[0100] Step 2131: If the difference between the initial prediction distances is greater than or equal to a preset threshold, the initial prediction distance corresponding to the maximum probability is used as the prediction distance;

[0101] Step 2132: If there is a difference less than the preset threshold, obtain the target probability and target initial prediction distance corresponding to the difference less than the preset threshold, recalculate the new probability based on the target probability, and calculate the initial prediction distance corresponding to the new probability based on the target initial prediction distance; compare each probability with the new probability, and take the initial prediction distance corresponding to the maximum probability as the prediction distance.

[0102] It is understandable that each of the above initial prediction distances may have a value of 0. For example, when there are no pedestrians at the intersection, the initial prediction distance predicted based on the movement trajectory of pedestrians at the intersection is 0. There may also be situations where multiple conditions are met at the same time, resulting in multiple non-zero initial prediction distances, such as s1=21 (with a probability of 0.1), s2=15 (with a probability of 0.2), and s3=13 (with a probability of 0.3). If the preset threshold is set to 1 meter, then any two initial prediction distances are greater than 1 meter. At this time, the s3 seat prediction distance corresponding to the maximum probability is used. If the preset threshold is set to 5 meters, there is a situation where the difference between s2 and s3 is less than 5 meters. At this time, a new probability needs to be calculated. and the new initial prediction distance , the calculation formulas are as follows:

[0103] ;

[0104] ;

[0105] in, is the order of target probability from large to small, is the initial prediction distance of the target corresponding to each sorted target probability.

[0106] According to the above formula, the new probability in the above example is: , the initial prediction distance corresponding to the new probability is: . Re-comparing the probabilities, , then the distance corresponding to probability 0.4 is is the final predicted distance.

[0107] Furthermore, in order to ensure that the vehicle can stop safely at the intersection, the speed planning based on the predicted distance may include the following steps:

[0108] Step 31: Determine the driving distance constraint according to the predicted distance;

[0109] Step 32: Obtaining various preliminary speed plans based on different single time steps and search rules;

[0110] Step 33: Calculate the loss value of each initial speed plan based on the driving distance constraint;

[0111] Step 34: The preliminary speed plan corresponding to the minimum loss value is used as the speed plan.

[0112] It should be noted that the premise of speed planning based on the predicted distance is that the target traffic light is red and the stop line is not recognized. For specific speed planning, please refer to Figure 2 . Figure 2 The horizontal axis s in the y represents the distance, the horizontal axis s represents the predicted distance, and the vertical axis v represents the speed. dt represents a single time step, and only dt units are allowed to move each time. Figure 2 The circled area on the left shows that when only dt units are allowed each time, the greater the speed, the greater the distance traveled. The driving distance constraint is formulated based on the predicted distance s, which can be understood as Figure 2 The middle rectangular area and the lower triangular area, Figure 2 The p in is the probability value corresponding to the above-mentioned predicted distance. Figure 2 The left circle area can further understand step 32, and several speed plans can be found based on the search rules. It can be understood that the following rules need to be followed during the speed planning process:

[0113] (1) Only dt units are allowed to move each time;

[0114] (2) It must move in the s direction, that is, straight ahead;

[0115] (3) The speed difference between the previous and next two frames must satisfy the maximum acceleration / deceleration of the ego vehicle.

[0116] Construct a loss function f. It is understandable that the smaller the loss f, the better. Calculate the loss value of each initial speed plan based on the driving distance constraint and the loss function; take the initial speed plan corresponding to the minimum loss value as the speed plan. The formula of the loss function is as follows:

[0117] .

[0118] is the judgment function. When the trajectory is on the constraint, Equal to 10000, and 0 when not on a constraint; It is also a judgment function. When the trajectory is after s+50, is equal to 10000, otherwise it is equal to 0 (to prevent the vehicle from accelerating beyond the constraint); a, b, c, d are weight coefficients; i represents the time step experienced; dt is a single time step; v t+1 is the speed planned for the next moment; v t is the speed planned at the current moment; s t is the current position of the vehicle at the current moment.

[0119] S102: When the target traffic light is red and a stop line is identified, the real distance between the vehicle and the stop line is obtained, and a brake stop process is performed based on the real distance.

[0120] When the vehicle is traveling at the speed planned, the target traffic light is still red, but the stop line can be identified because it is close to the stop line. At this time, the real distance between the vehicle and the stop line is obtained, and the brake is stopped based on the real distance. For details, please refer to Figure 3 , the driving distance constraint changes from the predicted distance to the actual distance, and the speed planning is performed based on the actual distance to achieve the intersection braking. It should be noted that if the target traffic light changes from red to green during driving, all speed planning will be invalidated, and the vehicle will drive normally without stopping at the intersection.

[0121] The method for parking at an intersection provided by the embodiment of the present invention is applied, and the target traffic light is determined according to the traffic light recognition result and the traffic light decision rule; when the target traffic light is a red light and the stop line is not recognized, the predicted distance between the vehicle and the stop line is obtained according to the stop line prediction rule, and the speed is planned based on the predicted distance; when the target traffic light is a red light and the stop line is recognized, the real distance between the vehicle and the stop line is obtained, and the brake is stopped based on the real distance. It can be seen that when the target traffic light is a red light, the stop line is predicted in advance and the speed is planned to ensure the safety and reliability of parking at the intersection, and the target traffic light of the method is determined according to the traffic light recognition result and the traffic light decision rule, without relying on the map to determine the target traffic light, and the method uses the stop line prediction rule to determine the stop line position, without relying on the map to obtain the stop line position; and compared with relying only on visual perception to identify the stop line position, the method also provides a method for obtaining the stop line position based on the stop line prediction rule, which reduces the dependence on visual perception, so that it can be safely braked at the intersection under special circumstances or when the stop line is blurred.

[0122] The following is an introduction to the parking device for intersections provided in an embodiment of the present invention. The parking device for intersections described below and the parking method for intersections described above can be referenced to each other.

[0123] Please refer to Figure 4 , Figure 4 A schematic diagram of the structure of a parking device at an intersection provided by an embodiment of the present invention may include:

[0124] A target traffic light determination module 100, configured to determine a target traffic light according to a traffic light recognition result and a traffic light decision rule;

[0125] A first speed planning module 200, for obtaining a predicted distance between the vehicle and the stop line according to a stop line prediction rule when the target traffic light is red and the stop line is not identified, and performing speed planning based on the predicted distance;

[0126] The second speed planning module 300 is used to obtain the real distance between the vehicle and the stop line when the target traffic light is red and the stop line is identified, and perform a braking process based on the real distance.

[0127] Based on the above embodiment, the target traffic light determination module 100 may include:

[0128] A first determining unit, configured to, when a traffic light is identified and works normally, determine that the traffic light is the target traffic light;

[0129] The second determination unit is configured to determine the target traffic light according to the location of the traffic light, the compliance of other vehicles with the traffic light, and whether the traffic light is operating normally when multiple traffic lights are identified.

[0130] Based on the above embodiment, the second determining unit may include:

[0131] a weight acquisition subunit, configured to acquire weights of the traffic light position, the condition of other vehicles following the traffic light, and whether the traffic light is working properly; the weight of whether the traffic light is working properly is greater than the weight of the traffic light position, and the weight of the traffic light position is greater than the weight of the condition of other vehicles following the traffic light;

[0132] A calculation unit is used to obtain the scores of each traffic light according to the weight, and take the traffic light corresponding to the highest score as the target traffic light.

[0133] Based on the above embodiment, the first speed planning module 200 may include:

[0134] The speed planning unit is used to predict the predicted distance between the vehicle and the stop line based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection and the movement trajectory of the preceding vehicle.

[0135] Based on the above embodiment, the speed planning unit includes:

[0136] An initial predicted distance acquisition unit is used to predict and obtain each initial predicted distance based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle;

[0137] A probability acquisition unit, used to acquire the respective probabilities of the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle, wherein the sum of the respective probabilities is 1;

[0138] A prediction distance calculation unit is used to calculate the prediction distance according to the initial prediction distances and the probabilities.

[0139] Based on the above embodiment, the prediction distance calculation unit may include:

[0140] A first prediction distance determination subunit, configured to use the initial prediction distance corresponding to the maximum probability as the prediction distance if the difference between the initial prediction distances is greater than or equal to a preset threshold;

[0141] The second prediction distance determination subunit is used to obtain the target probability and target initial prediction distance corresponding to the difference being less than the preset threshold if the difference is less than the preset threshold, recalculate the new probability according to the target probability, and calculate the initial prediction distance corresponding to the new probability according to the target initial prediction distance; compare the various probabilities with the new probability, and take the initial prediction distance corresponding to the maximum probability as the prediction distance.

[0142] Based on any of the above embodiments, the first speed planning module 200 may include:

[0143] A constraint unit, used for determining a driving distance constraint according to the predicted distance;

[0144] A preliminary speed planning unit, used for obtaining various preliminary speed plans based on different single time steps and search rules;

[0145] A loss calculation unit, used for calculating the loss value of each initial speed plan based on the driving distance constraint;

[0146] The speed plan determining unit is used to use the preliminary speed plan corresponding to the minimum loss value as the speed plan.

[0147] It should be noted that the order of the modules and units in the above-mentioned intersection parking device can be changed without affecting the logic.

[0148] The intersection parking device provided by the embodiment of the present invention is applied, and the target traffic light determination module 100 is used to determine the target traffic light according to the traffic light recognition result and the traffic light decision rule; the first speed planning module 200 is used to obtain the predicted distance between the vehicle and the stop line according to the stop line prediction rule when the target traffic light is a red light and the stop line is not recognized, and perform speed planning based on the predicted distance; the second speed planning module 300 is used to obtain the actual distance between the vehicle and the stop line when the target traffic light is a red light and the stop line is recognized, and perform braking processing based on the actual distance. It can be seen that when the target traffic light is red, the device predicts the stop line in advance and performs speed planning to ensure the safety and reliability of parking at the intersection. In the present invention, the target traffic light is determined based on the traffic light recognition result and the traffic light decision rule, and there is no need to rely on the map to determine the target traffic light. The present invention uses the stop line prediction rule to determine the stop line position, and there is no need to rely on the map to obtain the stop line position; and compared to relying only on visual perception to identify the stop line position, the present invention also provides a stop line position based on the stop line prediction rule, which reduces the dependence on visual perception, so that the vehicle can be safely stopped at the intersection under special circumstances or under conditions where the stop line is blurred.

[0149] The following is an introduction to the intersection parking equipment provided by an embodiment of the present invention. The intersection parking equipment described below and the intersection parking method described above can be referenced to each other.

[0150] Please refer to Figure 5 , Figure 5 A schematic diagram of the structure of a parking device at an intersection provided by an embodiment of the present invention may include:

[0151] A memory 10, used for storing computer programs;

[0152] The processor 20 is used to execute a computer program to implement the above-mentioned intersection parking method.

[0153] The memory 10 , the processor 20 , and the communication interface 31 all communicate with each other via the communication bus 32 .

[0154] In the embodiment of the present invention, the memory 10 is used to store one or more programs, and the program may include program code, and the program code includes computer operation instructions. In the embodiment of the present invention, the memory 10 may store programs for implementing the following functions:

[0155] Determine the target traffic light according to the traffic light recognition result and the traffic light decision rule;

[0156] When the target traffic light is red and the stop line is not recognized, the predicted distance between the vehicle and the stop line is obtained according to the stop line prediction rule, and the speed is planned based on the predicted distance;

[0157] When the target traffic light is red and the stop line is recognized, the actual distance between the vehicle and the stop line is obtained, and the brakes are stopped based on the actual distance.

[0158] In a possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function, etc.; the data storage area may store data created during use.

[0159] In addition, the memory 10 may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include an NVRAM. The memory stores an operating system and operating instructions, executable modules or data structures, or a subset thereof, or an extended set thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0160] The processor 20 may be a central processing unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, a microprocessor or any conventional processor, etc. The processor 20 may call a program stored in the memory 10 .

[0161] The communication interface 31 may be an interface of a communication module, and is used to connect to other devices or systems.

[0162] Of course, it should be noted that Figure 5 The structure shown does not constitute a limitation on the parking equipment at the intersection in the embodiment of the present invention. In actual applications, the parking equipment at the intersection may include Figure 5 More or fewer components than shown, or combinations of certain components.

[0163] The computer-readable storage medium provided in an embodiment of the present invention is introduced below. The computer-readable storage medium described below and the intersection parking method described above can be referenced to each other.

[0164] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned intersection parking method are implemented.

[0165] The computer-readable storage medium may include: 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.

[0166] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0167] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0168] Finally, it should be noted that, in this article, relationships such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0169] The above is a detailed introduction to a method, device, equipment and computer-readable storage medium for parking at an intersection provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for a person skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for parking at an intersection, characterized in that: include: Determine the target traffic light according to the traffic light recognition result and the traffic light decision rule; When the target traffic light is red and the stop line is not identified, a predicted distance between the vehicle and the stop line is obtained according to a stop line prediction rule, and speed planning is performed based on the predicted distance; When the target traffic light is red and the stop line is identified, the real distance between the vehicle and the stop line is obtained, and a brake stop process is performed based on the real distance.

2. The intersection parking method according to claim 1, characterized in that: The step of determining the target traffic light according to the traffic light recognition result and the traffic light decision rule includes: When a traffic light is identified and works normally, the traffic light is the target traffic light; When multiple traffic lights are identified, the target traffic light is determined according to the location of the traffic light, the compliance of other vehicles with the traffic light, and whether the traffic light is working properly.

3. The intersection parking method according to claim 2, characterized in that: The step of determining the target traffic light according to the location of the traffic light, the condition of other vehicles following the traffic light, and whether the traffic light is working properly includes: Obtaining weights of the traffic light position, other vehicles' compliance with the traffic light, and whether the traffic light is working properly; the weight of whether the traffic light is working properly is greater than the weight of the traffic light position, and the weight of the traffic light position is greater than the weight of other vehicles' compliance with the traffic light; The scores of the traffic lights are obtained according to the weights, and the traffic light corresponding to the highest score is used as the target traffic light.

4. The intersection parking method according to claim 1, characterized in that: The step of obtaining the predicted distance between the vehicle and the stop line according to the stop line prediction rule includes: The predicted distance between the vehicle and the stop line is predicted based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection and the movement trajectory of the preceding vehicle.

5. The intersection parking method according to claim 4, characterized in that: The predicted distance between the vehicle and the stop line is predicted based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle, including: Based on the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection and the movement trajectory of the preceding vehicle, each initial prediction distance is obtained; Obtaining the respective probabilities of the target traffic light position, the auxiliary traffic light position, the movement trajectory of pedestrians at the intersection, the movement trajectory of non-motor vehicles at the intersection, the position of stationary vehicles at the intersection, and the movement trajectory of the preceding vehicle, wherein the sum of the respective probabilities is 1; The predicted distance is obtained according to the initial predicted distances and the probabilities.

6. The intersection parking method according to claim 5, characterized in that: The step of calculating the predicted distance according to the initial predicted distances and the probabilities includes: If the difference between the initial prediction distances is greater than or equal to a preset threshold, the initial prediction distance corresponding to the maximum probability is used as the prediction distance; If there is a difference value less than the preset threshold, the target probability and the target initial prediction distance corresponding to the difference value less than the preset threshold are obtained, the new probability is recalculated according to the target probability, and the initial prediction distance corresponding to the new probability is calculated according to the target initial prediction distance; the various probabilities are compared with the new probability, and the initial prediction distance corresponding to the maximum probability is used as the prediction distance.

7. The intersection parking method according to any one of claims 1 to 6, characterized in that: The performing speed planning based on the predicted distance includes: determining a driving distance constraint based on the predicted distance; Each preliminary speed plan is obtained based on different single time steps and search rules; Calculate the loss value of each initial speed plan based on the driving distance constraint; The preliminary speed plan corresponding to the minimum loss value is used as the speed plan.

8. A parking device at an intersection, characterized in that: include: A target traffic light determination module, used for determining a target traffic light according to a traffic light recognition result and a traffic light decision rule; A first speed planning module, configured to obtain a predicted distance between the vehicle and the stop line according to a stop line prediction rule when the target traffic light is red and the stop line is not identified, and perform speed planning based on the predicted distance; The second speed planning module is used to obtain the real distance between the vehicle and the stop line when the target traffic light is red and the stop line is identified, and perform a brake stop process based on the real distance.

9. A parking device at an intersection, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the intersection parking method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are loaded and executed by the processor, the steps of the intersection parking method according to any one of claims 1 to 7 are implemented.

Images