A ramp target detection and ACC control method

Abstract

The application provides a ramp target detection and ACC control method, and belongs to the intelligent network connection field. The method firstly determines the scene where the vehicle is located according to the data collected by the camera, and then performs ACC control according to the scene. The application adopts a driving assistance camera to replace the original camera and radar scheme, reduces the cost of the whole vehicle, and solves the problems of false deceleration caused by the misselection of the target in front of the vehicle in the ramp scene and frequent acceleration and deceleration on the uphill and downhill, improves the adaptive cruise comfort, reduces the user complaints, and improves the driving safety.

Description

Technical Field

[0001] This invention belongs to the field of intelligent connected vehicles, specifically relating to a slope target detection and ACC control method. Background Technology

[0002] Models equipped with driver assistance cameras can achieve ACC adaptive cruise control, which can maintain a steady speed and follow the vehicle in front, adapting to the speed of the vehicle in front and achieving stop-start and start-go functions across the entire speed range and all road types.

[0003] While driver assistance cameras can enable ACC (Adaptive Cruise Control) to follow the vehicle in front across the entire speed range, in scenarios where the vehicle is going uphill (e.g., the vehicle in front is at the crest of the hill) or downhill (e.g., the vehicle in front is going uphill), the vehicle may select the vehicle in front and trigger unintended deceleration, which does not match the actual driving scenario. When both the vehicle and the vehicle in front are going uphill or downhill, inaccurate speed and distance measurements during ACC control lead to frequent DCEC (Deceleration Control and Elevation Control) switching, resulting in a poor driving experience. When following the vehicle in front uphill, if the vehicle in front reaches the crest, it may lose sight of the target, causing the vehicle to accelerate suddenly and then decelerate significantly upon recognizing the target again. Summary of the Invention

[0004] The purpose of this invention is to solve the problems existing in the prior art and provide a slope target detection and ACC control method to solve the problems of "ACC with driver assistance camera misselecting the target directly in front in slope scene, resulting in false deceleration, frequent switching of DCEC when following the vehicle uphill and downhill, and easy loss of target at the top of the slope and then recognition of target, resulting in large acceleration switching and deceleration actions of the vehicle".

[0005] This invention is achieved through the following technical solution:

[0006] This invention provides a slope target detection and ACC control method. The method first determines the scene in which the vehicle is located based on data collected by a camera, and then performs ACC control based on the scene.

[0007] A further improvement of the present invention is that:

[0008] The operation of determining the scene in which the vehicle is located based on data collected by the camera includes:

[0009] If the following conditions are met, it is determined to be Scenario 1:

[0010] The vehicle is currently on a flat road.

[0011] There is a slope ahead;

[0012] The relative distance between the vehicle and the vehicle in front is greater than the set threshold;

[0013] The aspect ratio of the vehicle in front remains stable.

[0014] No changes were found in the image of the rear of the vehicle in front.

[0015] A further improvement of the present invention is that:

[0016] The operation of determining the scene in which the vehicle is located based on data collected by the camera includes:

[0017] If the following conditions are met, then it is determined to be Scenario 2:

[0018] The vehicle is located on a downhill slope;

[0019] The aspect ratio of the vehicle in front first gradually decreases and then gradually increases;

[0020] The sequence of changes in the image of the rear of the vehicle in front is as follows: the lower part disappears first, leaving only the upper part, and then the upper, middle, and lower parts appear in sequence.

[0021] A further improvement of the present invention is that:

[0022] The operation of determining the scene in which the vehicle is located based on data collected by the camera includes:

[0023] If the following conditions are met, it is determined to be Scenario 3:

[0024] The vehicle is located uphill;

[0025] The sequence of changes in the image of the rear of the car in front is: the upper part disappears first, then the middle part disappears, and finally the lower part disappears.

[0026] A further improvement of the present invention is that:

[0027] The operation of determining the scene in which the vehicle is located based on data collected by the camera includes:

[0028] If the following conditions are met, then it is determined to be Scenario 4:

[0029] The vehicle is located uphill;

[0030] The image of the rear of the vehicle in front remains unchanged;

[0031] The aspect ratio of the vehicle in front remains stable;

[0032] The camera detected no breaks in the lane lines ahead, and the changes in the 3D depth of the road were consistent with the changes caused by the slope values ​​given by the sensors or maps.

[0033] A further improvement of the present invention is that:

[0034] The operation of determining the scene in which the vehicle is located based on data collected by the camera includes:

[0035] If the following conditions are met, then it is determined to be Scenario 5:

[0036] The vehicle is currently on a flat road.

[0037] The aspect ratio of the vehicle in front initially remained stable, then decreased.

[0038] The sequence of changes in the image of the rear of the car in front is as follows: the lower part disappears first, then the middle part disappears, and finally only the upper part remains.

[0039] A further improvement of the present invention is that:

[0040] The operation of determining the scene in which the vehicle is located based on data collected by the camera includes:

[0041] If the following conditions are met, then it is determined to be Scenario Six:

[0042] The vehicle is located on a downhill slope;

[0043] The aspect ratio of the vehicle in front initially remained stable, then decreased.

[0044] The sequence of changes in the image of the rear of the car in front is as follows: the lower part disappears first, then the middle part disappears, and finally only the upper part remains.

[0045] A further improvement of the present invention is that:

[0046] The ACC control operation based on the scenario includes:

[0047] In one scenario, if the relative distance between the vehicle and the vehicle in front is greater than a set threshold, the vehicle will not select the vehicle in front as the target vehicle; if the relative speed calculated using the relative distance between the vehicle and the vehicle in front and the relative speed output by the camera are greater than a set threshold, the vehicle will not select the vehicle in front as the target vehicle.

[0048] In scenarios two, five, and six, if the relative speed calculated using the relative distance between the vehicle and the vehicle in front, and the relative speed output by the camera, are greater than a set threshold, then the acceleration is calculated using the following formula:

[0049]

[0050] In scenario three, if the relative speed calculated using the relative distance between the vehicle and the vehicle in front, and the relative speed output by the camera, are both greater than a set threshold, then the acceleration is calculated using the following formula:

[0051] When the aspect ratio is not 0

[0052] When the aspect ratio is 0, a upperlimitVel =0

[0053] In scenario four, the acceleration is calculated using the following formula:

[0054]

[0055] Where V1 is the speed of the vehicle, V2 is the speed of the vehicle in front, s is the relative distance between the vehicle and the vehicle in front, and C is the speed of the vehicle in front. radslope C1 is the slope coefficient, and C2 is the height-to-width ratio coefficient.

[0056] A further improvement of the present invention is that:

[0057] In scenarios two, three, four, five, and six, the acceleration on the ramp is calculated using the following formula:

[0058] a upperlimitroadslope =a roadslope过渡Vel +a roadslope过渡

[0059] The acceleration during the transition from a flat road to a ramp can be calculated using the following formula:

[0060] a roadslope过渡 =a roadslope *C Vel自车

[0061] Among them, a roadslope过渡Vel =a roadslope *(1-C Vel自车 )

[0062] a roadslope =a upperlimitVel *C roadslope

[0063] C radslope It is the gradient coefficient, C Vel自车 It is the transition coefficient for flat roads and slopes.

[0064] Compared with the prior art, the beneficial effects of the present invention are:

[0065] This invention uses a driver assistance camera to replace the original camera plus radar solution, which reduces the cost of the whole vehicle. It also solves the problems of accidental deceleration caused by misselecting the target directly in front in slope scenarios and frequent acceleration and deceleration when going uphill and downhill, improves the comfort of adaptive cruise control, reduces user complaints, and enhances driving safety. Attached Figure Description

[0066] Figure 1 A schematic diagram of Scene 1;

[0067] Figure 2 A schematic diagram of scenario two;

[0068] Figure 3 A schematic diagram of scenario three;

[0069] Figure 4A diagram of scenario four;

[0070] Figure 5 A diagram of Scene 5;

[0071] Figure 6 A diagram of Scene Six;

[0072] Figure 7 A diagram illustrating how a camera perceives changes in the height-to-width ratio of the vehicle in front.

[0073] Figure 8 A flowchart illustrating the steps of the method of this invention;

[0074] Figure 9 Scenario 3: A diagram illustrating the change in acceleration of the vehicle going uphill, after the front vehicle disappears from the top of the hill;

[0075] Figure 10 Scenario 3: A diagram illustrating the changes in the acceleration of a vehicle going uphill, after the vehicle in front disappears from the top of the hill and is no longer recognized.

[0076] Figure 11 Schematic diagrams showing the changes in vehicle acceleration in scenarios two, five, and six of the ramp. Detailed Implementation

[0077] The present invention will now be described in further detail with reference to the accompanying drawings:

[0078] This invention provides a slope target detection and ACC control method. The method first determines the scene in which the vehicle is located based on data collected by a camera, and then performs ACC control based on the scene.

[0079] The principle of target detection and ACC control using the method of this invention is as follows:

[0080] 1) The vehicle's location (flat road, uphill or downhill) is output based on the vehicle's relevant modules (such as the ESP module installed on the vehicle, the driver assistance camera (which can be monocular, binocular or multi-view camera), or the map), and the slope value is output (the existing ESP module or the camera's yaw rate sensor (IMU) can calculate the slope value, or the slope value output by the positioning information provided by the map can be used directly) to the driver assistance camera for acceleration control;

[0081] Existing driver assistance cameras periodically output perceived attributes of targets ahead. These attributes include the vehicle's height, width, speed, and the relative distance between the vehicle in front and the vehicle itself. The camera can also calculate the aspect ratio of the vehicle in front based on these attributes. The aspect ratio of the vehicle in front output by the camera will differ depending on several scenarios: when one vehicle is on a slope and the other is on a straight road; when both vehicles are simultaneously going uphill or downhill; and when both vehicles are on a flat road. When the vehicle in front is going uphill or downhill, the perceived width remains relatively constant, but the perceived height changes significantly. Figure 7 As shown, as the vehicle in front moves and the slope angle changes, the aspect ratio of the rear of the vehicle detected by the camera changes. Therefore, the position of the vehicle in front can be determined by using the changes in the aspect ratio output by the camera. Furthermore, the aspect ratio of the vehicle in front can also be used to assist in determining the relative speed and relative distance to the vehicle in front, and to control the acceleration and deceleration of the vehicle. In addition, the vehicle's controller stores a table of aspect ratio and aspect ratio coefficient C1, as well as a table of slope value and slope coefficient C. radslope The lookup table allows you to find the corresponding height-to-width ratio coefficient C1 and the corresponding slope coefficient C1 based on the real-time obtained height-to-width ratio and slope value. radslope .

[0082] 2) Lane Line Assistance Judgment: The continuity of lane lines identified by cameras can help determine the vehicle's location. Existing cameras can recognize normal white and yellow single lane lines, double yellow lines, dashed lines, and solid lines. They can also identify curbs, flowerbeds, and guardrails as lane lines, which can be used to determine if the vehicle is on a slope. Specifically, if the camera detects a break in the lane line ahead with a length greater than Nm, or if the 3D depth change of the road exceeds a set threshold, or if the map's built-in positioning information indicates a slope ahead, then a slope can be determined.

[0083] 3) The camera identifies the rear information of the target ahead, removes invalid information from the image, retains highly relevant morphemes, and then segments the acquired image. Based on the segmented image information, the rear of the vehicle in front is divided into three parts: upper, middle, and lower. (Specifically, based on the existing camera's neural network and intelligent AI algorithm, the acquired rear image of the vehicle in front is segmented; for example, the glass is considered the upper part, the taillights, brake lights, side marker lights, and connecting parts are considered the middle part, and the tires and rear bumper are considered the lower part.) The changing order of the upper, middle, and lower parts of the rear image can then be used to help determine the relative positions of the vehicle and the vehicle in front. For example, in scenarios two and five below, the rear image appears from top to bottom and then disappears; in scenarios three and six, it disappears from top to bottom and then reappears. Therefore, the relative positions of the two vehicles on the slope can be determined based on the changing order of the rear images.

[0084] 4) such as Figure 8 As shown, the slope value is calculated based on ESP or the yaw rate sensor (IMU) of the camera, or obtained from the location information output by the map. The slope value determines the vehicle's location: flat road, uphill, or downhill. Then, based on the aspect ratio of the vehicle in front, the continuity of lane lines, and the order in which the image of the rear of the vehicle in front disappears, the system determines the current scene (e.g.,...). Figures 1 to 6 As shown in the diagram, for scenario one, no action is taken on the target ahead, i.e., no deceleration or acceleration is performed; for other scenarios, ACC control is performed with the aid of ramp acceleration limits, based on the height-to-width ratio of the vehicle in front and the change in relative speed.

[0085] The specific steps in the method of this invention to determine the scene in which the vehicle is located based on the data collected by the camera are as follows:

[0086] If the following conditions are met, it is determined to be Scenario 1:

[0087] The vehicle is currently on a flat road.

[0088] There is a slope ahead;

[0089] The relative distance between the vehicle and the vehicle in front is greater than the set threshold;

[0090] The aspect ratio of the vehicle in front remains stable.

[0091] No changes were found in the image of the rear of the vehicle in front.

[0092] If the following conditions are met, then it is determined to be Scenario 2:

[0093] The vehicle is located on a downhill slope;

[0094] The aspect ratio of the vehicle in front first gradually decreases and then gradually increases;

[0095] The sequence of changes in the image of the rear of the vehicle in front is as follows: the lower part disappears first, leaving only the upper part, and then the upper, middle, and lower parts appear in sequence.

[0096] If the following conditions are met, it is determined to be Scenario 3:

[0097] The vehicle is located uphill;

[0098] The sequence of changes in the image of the rear of the car in front is: the upper part disappears first, then the middle part disappears, and finally the lower part disappears.

[0099] If the following conditions are met, then it is determined to be Scenario 4:

[0100] The vehicle is located uphill;

[0101] The image of the rear of the vehicle in front remains unchanged;

[0102] The aspect ratio of the vehicle in front remains stable;

[0103] The camera detected no breaks in the lane lines ahead, and the changes in the 3D depth of the road were consistent with the changes caused by the slope values ​​given by the sensors or maps.

[0104] If the following conditions are met, then it is determined to be Scenario 5:

[0105] The vehicle is currently on a flat road.

[0106] The aspect ratio of the vehicle in front initially remained stable, then decreased.

[0107] The sequence of changes in the image of the rear of the car in front is as follows: the lower part disappears first, then the middle part disappears, and finally only the upper part remains.

[0108] If the following conditions are met, then it is determined to be Scenario Six:

[0109] The vehicle is located on a downhill slope;

[0110] The aspect ratio of the vehicle in front initially remained stable, then decreased.

[0111] The sequence of changes in the image of the rear of the car in front is as follows: the lower part disappears first, then the middle part disappears, and finally only the upper part remains.

[0112] The specific operations for ACC control based on the scenario in the method of this invention are as follows:

[0113] 1. The control method for Scenario 1 is as follows:

[0114] Scene as Figure 1As shown, the vehicle is on a flat road with a slope in between. This is a simulated slope scenario. In this scenario, the camera determines whether to select and control the target based on the relative distance and speed between the two vehicles. The relative distance and speed are determined using an "OR" relationship, as detailed below:

[0115] If the relative distance between the vehicle and the vehicle in front is greater than the set threshold (e.g., 70m), the vehicle will not select the vehicle in front as the target vehicle, that is, it will not perform acceleration or deceleration control.

[0116] If the relative speed calculated using the relative distance between the vehicle and the vehicle in front is greater than the relative speed output by the camera, the vehicle will not select the vehicle in front as the target vehicle, that is, it will not perform acceleration or deceleration control.

[0117] Specifically, the formula for calculating relative speed using the relative distance between the vehicle and the vehicle in front is as follows:

[0118] Relative velocity calculated using distance = c × |distance N cycles ago - distance in the current cycle| / t

[0119] In the above formula, distance is the relative distance output by the camera, and c is the filtering coefficient, which can make the relative speed smoother. The target distance for N cycles or beyond N cycles is not selected. The main focus is on whether the target ID changes. The relative speed calculated using the above formula can solve the problem of abrupt changes in the relative speed directly output by the camera.

[0120] 2. The control methods for scenarios two, five, and six are as follows:

[0121] like Figure 2 , Figure 5 , Figure 6 As shown, scenarios two, five, and six are all real slope scenarios. In scenarios two, five, and six, the vehicle in front is always in the driver's field of vision, but the actual distance is greater than the distance output value sensed by the camera. If the driver maintains its original control, there will be no danger, nor will there be frequent switching between acceleration and deceleration. However, if the driver judges based on the distance sensed by the camera, it will first decelerate and then continue to accelerate. Therefore, if there is no relevant limit on acceleration, the driver will first decelerate and then accelerate suddenly, resulting in frequent switching between acceleration and deceleration, which makes the driver's experience very unpleasant.

[0122] In scenarios two, five, and six, if the relative speed calculated using the relative distance between the vehicle and the vehicle in front, and the relative speed output by the camera, exceed a set threshold, the vehicle's acceleration is limited based on the calculated relative speed. Simultaneously, acceleration is also limited based on the aspect ratio of the vehicle in front and the slope, thus avoiding frequent acceleration and deceleration. Figure 11 As shown.

[0123] Specifically, in scenarios two, five, and six, the acceleration is calculated using the following formula:

[0124]

[0125] Where V1 is the vehicle's speed, transmitted directly from its CAN signal; V2 is the speed of the vehicle in front, sensed and output by a camera; s is the relative distance between the vehicle and the vehicle in front; and C... radslope C1 is the slope coefficient, and C2 is the aspect ratio coefficient. Because the aspect ratio coefficient C1 and the slope coefficient C2... radslope It is obtained by looking up the height-to-width ratio and slope value of the vehicle in front in real time based on the camera's calculations, so it is obtained through C1, C radslope The change in aspect ratio can control the change in acceleration 'a'. Specifically, when the aspect ratio is close to a fixed value but not a normal ratio (such as...), Figure 11 As shown in the horizontal segment of the aspect ratio curve, when the vehicle's acceleration is 0, it maintains its current speed for steady driving. As the aspect ratio gradually increases, the vehicle's acceleration is no longer limited, and the normal ACC following control is restored. The acceleration is converted into torque request input to the power unit—engine or motor—to control the vehicle to follow the vehicle in front stably.

[0126] 3. The control method for scenario three is as follows:

[0127] like Figure 3 As shown, Scenario 3 involves the vehicle going uphill with the vehicle in front at the top of the slope; this is a true incline scenario. In Scenario 3, the vehicle in front disappears from the driver's view. If the driver does not take any control measures, the vehicle will first accelerate rapidly, then recognize the target vehicle, and then decelerate. The entire process involves frequent acceleration and deceleration switching, resulting in a very unpleasant driving experience.

[0128] In scenario three, if the relative speed calculated using the relative distance between the vehicle and the vehicle in front, and the relative speed output by the camera, exceed a set threshold, then the vehicle's acceleration and deceleration will be limited based on the calculated relative speed. Simultaneously, acceleration will be limited based on the aspect ratio of the vehicle in front and the slope value, as detailed below:

[0129] like Figure 9 As shown, in scenario three, the car in front disappears at the top of the hill, and the vehicle is detected again after it goes uphill. To avoid frequent acceleration and deceleration, the vehicle's acceleration is limited based on the aspect ratio. Simultaneously, if the vehicle is going uphill (the sensor or map explicitly signals to the camera whether the vehicle is going uphill, downhill, or on flat ground; for example, if the signal is 2, it means...), then... Figures 9 to 11 The uphill setting signal indicates that the vehicle is uphill. If the signal is 1, it means... Figures 9 to 11The downhill setting signal indicates that the vehicle is going downhill (if the signal is 0, the vehicle is on a flat road). When the aspect ratio is close to 0, the vehicle's acceleration is 0, maintaining the current speed for steady driving. When the vehicle goes uphill and the vehicle in front reappears, the acceleration restriction on the vehicle is reduced according to the aspect ratio of the vehicle in front, and following control is restored.

[0130] like Figure 10 As shown, in scenario three, the vehicle in front disappears at the top of the hill. If the vehicle does not detect the vehicle in front again after going uphill, to avoid frequent acceleration and deceleration, the vehicle's acceleration is limited based on the aspect ratio. Simultaneously, if the vehicle is going uphill, when the aspect ratio approaches 0, the vehicle's acceleration is 0, maintaining the current speed for steady driving. (A height-to-width ratio close to 0 indicates the vehicle in front is either at the top of the hill or leaving it. To prevent the vehicle from suddenly accelerating towards the top and then being re-identified, triggering deceleration (avoiding frequent acceleration and deceleration), the vehicle's acceleration is also limited to 0 based on the aspect ratio approaching 0, maintaining the current speed.) Once the vehicle is going uphill, if no vehicle in front is detected, and if the vehicle is going downhill, normal ACC following control is restored, and acceleration is no longer limited.

[0131] The formula for calculating acceleration 'a' in scenario three is as follows:

[0132] When the aspect ratio is not 0

[0133] When the aspect ratio is 0, a upperlimitVel =0

[0134] At the beginning, the vehicle normally recognizes the car in front as the target car and follows it. However, as the car in front reaches the top of the hill, the vehicle continues to go uphill, but the tail of the target car gradually disappears. The acceleration and speed of the car in front remain stable or do not change suddenly, but the target disappears. At this point, it is considered to be in scenario three. Then the target disappears, and the vehicle maintains its current speed without accelerating or decelerating, maintaining steady-state driving until the vehicle encounters a new target or changes from uphill to downhill or straight road.

[0135] 3. The control method for scenario four is as follows:

[0136] Since both the driver and the target vehicle are on an uphill slope in Scenario 4, the aspect ratio of the target vehicle remains stable. Therefore, the value of C1 from the table is a fixed coefficient of 1. To avoid frequent acceleration and deceleration of the driver, the driver's acceleration 'a' will be determined by the coefficient C obtained from the table based on the slope value. radslope To impose restrictions, the specific formula for calculating acceleration is as follows:

[0137]

[0138] 4. Gradient acceleration control

[0139] Furthermore, if the ramp scenario is real (i.e., scenarios two through six), additional acceleration control will be implemented. The acceleration required for uphill and downhill driving is further limited because, in this case, the driver feels not only the acceleration from the ground but also the gravitational acceleration caused by the slope. The aim of this reduction is to ensure that the driver feels the same forces on a ramp as on a flat road.

[0140] At the beginning, multiply the previously calculated acceleration value by the gradient factor:

[0141] a roadslope =a upperlimitVel *C roadslope

[0142] a roadslope For the acceleration of the ramp, C radslope It is the slope coefficient, used to calculate the acceleration of a slope section, and is a weighting factor for calculating the upper limit of road slope acceleration.

[0143] To ensure a smooth transition of acceleration limits from flat roads to ramps, a ramp factor is used. This factor is applied to the acceleration caused by the road gradient and the acceleration calculated based on the vehicle's speed, making acceleration and deceleration more comfortable.

[0144] a roadslope过渡 =a roadslope *C Vel自车

[0145] a roadslope过渡 It is the transition acceleration from a flat road to an uphill or downhill slope, C. Vel自车 It is the transition coefficient for flat road slopes (an empirical value, pre-entered).

[0146] Then, based on the transition acceleration, the acceleration used for the transition from the straightaway to the ramp is calculated:

[0147] a roadslope过渡Vel =a roadslope *(1-C Vel自车 )

[0148] Among them, a roadslope过渡 It is the transition acceleration from a flat road to an uphill or downhill slope, a roadslope过渡Vel It refers to the acceleration after the transition, that is, the acceleration after transitioning from a flat road to an uphill or downhill slope.

[0149] Finally, calculate the acceleration values ​​for uphill and downhill sections:

[0150] a upperlimitroadslope =a roadslope过渡Vel +a roadslope过渡 (3)

[0151] The above formula is used for calculations regardless of whether it is uphill or downhill; just distinguish between positive and negative signs.

[0152] This invention proposes a scheme for achieving full-speed adaptive cruise control (ACC) based on a pure vision approach. This solves the problem of misselecting targets ahead on slopes and addresses the issues of misselection of targets triggering erroneous deceleration or inaccurate speed and distance measurement leading to poor acceleration and deceleration control in slope scenarios, which are unique to pure vision-based adaptive cruise control schemes.

[0153] The above technical solution is only one embodiment of the present invention. For those skilled in the art, based on the principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the technical solutions described in the specific embodiments of the present invention. Therefore, the foregoing description is only preferred and not restrictive.

Claims

1. A ramp object detection and ACC control method, characterized by: The method firstly determines the scene where the ego vehicle is located according to the data collected by the camera, and then performs ACC control according to the scene; The operation of determining the scene where the ego vehicle is located according to the data collected by the camera comprises: If the following conditions are met, it is determined to be scene one: The position where the ego vehicle is located is a flat road; There is a slope in front of the ego vehicle; The relative distance between the ego vehicle and the front vehicle is greater than a set threshold value; The height-width ratio of the front vehicle maintains a stable value; No change is found in the rear image of the front vehicle; The scene comprises at least one of scene one to scene six, and the operation of performing ACC control according to the scene comprises: In scene one, if the relative distance between the ego vehicle and the front vehicle is greater than a set threshold value, the ego vehicle will not select the front vehicle as a target vehicle; if the relative speed calculated by using the relative distance between the ego vehicle and the front vehicle is greater than a set threshold value, the ego vehicle will not select the front vehicle as a target vehicle; In scene two, scene five and scene six, if the relative speed calculated by using the relative distance between the ego vehicle and the front vehicle is greater than a set threshold value, the acceleration is calculated by using the following formula: In scene three, if the relative speed calculated by using the relative distance between the ego vehicle and the front vehicle is greater than a set threshold value, the acceleration is calculated by using the following formula: In scene four, the acceleration is calculated by using the following formula: Wherein, V1 is the speed of the ego vehicle, V2 is the speed of the preceding vehicle, s is the relative distance between the ego vehicle and the preceding vehicle, C radslope is the slope coefficient, and C1 is the height-width ratio coefficient.

2. The ramp object detection and ACC control method according to claim 1, characterized by: The operation of determining the scene where the ego vehicle is located according to the data collected by the camera comprises: If the following conditions are met, it is determined to be scene two: The position where the ego vehicle is located is a downhill; The height-width ratio of the front vehicle gradually becomes smaller and then gradually becomes larger; The change sequence of the rear image of the front vehicle is that the lower part disappears first, only the upper part remains, and then the upper part, the middle part and the lower part appear in sequence.

3. The ramp object detection and ACC control method according to claim 2, characterized in that: The operation of determining the scene where the ego vehicle is located according to the data collected by the camera comprises: If the following conditions are met, it is determined to be scene three: The position where the ego vehicle is located is an uphill; The change sequence of the rear image of the front vehicle is that the upper part disappears first, then the middle part disappears, and finally the lower part disappears.

4. The ramp object detection and ACC control method according to claim 3, characterized by: The operation of determining the scene where the ego vehicle is located according to the data collected by the camera comprises: If the following conditions are met, it is determined to be scene four: The position where the ego vehicle is located is an uphill; No change is found in the rear image of the front vehicle; The height-width ratio of the front vehicle maintains a stable value; The front lane line recognized by the camera does not appear to be broken or the change in the 3D depth of the road is consistent with the change caused by the slope value given by the sensor or the map.

5. The ramp object detection and ACC control method according to claim 4, characterized by: The operation of determining the scene where the ego vehicle is located according to the data collected by the camera comprises: If the following conditions are met, it is determined to be scene five: The position where the ego vehicle is located is a flat road; The height-width ratio of the front vehicle first maintains a stable value and then becomes smaller; The change sequence of the rear image of the front vehicle is that the lower part disappears first, then the middle part disappears, and finally only the upper part remains.

6. The ramp object detection and ACC control method according to claim 5, characterized by: The operation of determining the scene where the ego vehicle is located according to the data collected by the camera comprises: If the following conditions are met, it is determined to be scene six: The position where the ego vehicle is located is a downhill; The height-width ratio of the front vehicle first maintains a stable value and then becomes smaller; The change sequence of the rear image of the front vehicle is that the lower part disappears first, then the middle part disappears, and finally only the upper part remains.

7. The ramp object detection and ACC control method of claim 1, wherein: In scenario two, scenario three, scenario four, scenario five, scenario six, the acceleration on the ramp is calculated by using the following formula: The acceleration when the flat road to the ramp transition is calculated by using the following formula: wherein C radslope is the ramp coefficient, C Vel自车 is the flat-ramp transition coefficient.

Images