Front following target vehicle release algorithm based on self-vehicle motion working condition

Abstract

The invention discloses a front following target vehicle release algorithm based on a self-vehicle motion condition. The algorithm comprises the following steps: acquiring self-vehicle motion parameters through a self-vehicle control system, acquiring a parameter equation of a center line of a self-lane through camera shooting, the width of the self-lane and motion parameters of a front following target vehicle; judging the movement trend of the vehicle; obtaining a virtual optimization self-lane; judging whether a following target vehicle in front of the own vehicle has a trend of switching out of a self-camera to acquire a self-lane or not; judging whether a front following target vehicle meets the condition of switching out of a camera shooting collection self-lane or not; if the front vehicle-following target vehicle meets the switching-out condition, releasing the front vehicle-following target vehicle; and if the front vehicle-following target vehicle does not meet the switching-out condition, determining that the front vehicle-following target vehicle is in a switching-out state, and controlling the vehicle. The release algorithm has the advantages that the driver can release the car following target more reliably and effectively in the car following driving process such as turning or lane changing, ACC control is smoother, and the like.

Description

Technical Field

[0001] The present invention relates to a vehicle control technology for intelligent driving of vehicles, in particular to a following vehicle target vehicle release algorithm based on the driving conditions of the host vehicle for longitudinal control of vehicles. Background Art

[0002] The longitudinal control in an intelligent driving system mainly functions as an Adaptive Cruise Control (ACC). An intelligent vehicle equipped with an adaptive cruise control system uses radar and a computer to identify whether the object approaching the host vehicle is a bicycle, a car, or a pedestrian, and controls the driving state of the vehicle according to the road conditions, completely or partially replacing the driver's operation. When the driver activates the ACC function, the auxiliary driving controller controls the brakes and accelerator of the host vehicle based on the cruise speed set by the driver and the information of the front target detected by the camera or millimeter-wave radar, and controls the longitudinal speed of the host vehicle. When there is no following vehicle target vehicle in front, the ACC controls the vehicle to travel at the cruise speed set by the driver or to accelerate appropriately. When a following vehicle target vehicle is detected in front and the cruise speed set by the driver is less than that of the following vehicle target vehicle, the ACC controls the vehicle to travel at the cruise speed set by the driver and accelerates cautiously. When the cruise speed set by the driver is greater than the speed of the following vehicle target vehicle, the ACC controls the host vehicle to follow the following vehicle target vehicle and decelerates.

[0003] For the ACC longitudinal control, in order to make the host vehicle more comfortable and safe during longitudinal driving, the ACC generally uses a state machine for control, and the state machine of the ACC in its algorithm is as Figure 3 shown.

[0004] As Figure 10, the main states of ACC control are introduced as follows: Off means the system is turned off; Error means the system has a fault; Passive means the system is in a monitoring state; Standby means the system is ready; Active means the system has been activated; Override means that when the ACC function is turned on, the driver operates the accelerator pedal, and the depth of the accelerator pedal exceeds the depth value of the accelerator pedal requested by ACC, and the vehicle responds to the driver's operation of the accelerator pedal; The ON state means that when there is a leading vehicle target vehicle in front of the vehicle and the speed set by the driver is greater than the speed of the leading vehicle target vehicle in front, ACC controls the own vehicle to travel at the speed in front; The SpeedControl state means that when there is no leading vehicle target vehicle in front, or there is a leading vehicle target vehicle but the speed of the leading vehicle target vehicle is greater than the cruising speed set by the driver, ACC controls the own vehicle to travel at the set cruising speed; The tempStop state means that ACC controls the own vehicle to follow the leading vehicle target vehicle to a stop state. When the leading vehicle target vehicle starts again, ACC controls the own vehicle to start following again. The duration of tempStop is generally about 3s. If it exceeds this time, ACC will enter the Stop state; The Stop state means that ACC controls the own vehicle to follow the leading vehicle target vehicle to stop. After the leading vehicle target vehicle drives away, the driver can step on the accelerator or operate the ACC system switch, and the ACC system controls the own vehicle to follow the leading vehicle target vehicle again.

[0005] After the ACC function is turned on, the target screening algorithm will screen out the leading vehicle target vehicle according to the information of the leading vehicle target vehicle detected by the sensor / camera, the lane line information, and the movement of the own vehicle. The accuracy of target screening has a great impact on the control of ACC: misselecting the leading vehicle target vehicle will often cause unnecessary deceleration of ACC, and selecting the leading vehicle target vehicle too late will result in a relatively large deceleration of the own vehicle after selecting the leading vehicle target vehicle. When the leading vehicle target vehicle cuts out of the lane of the own vehicle, the timing of the own vehicle releasing the leading vehicle target vehicle (that is, when to release the leading vehicle target vehicle after the leading vehicle target vehicle cuts out of the lane of the own vehicle) often has a great impact on the control of ACC. When the leading vehicle target vehicle that is being followed cuts out of the lane of the own vehicle, releasing it too early often leads to a risk of collision, and releasing it too late leads to too slow acceleration of the own vehicle, and there will be a long section of the road without vehicles in the front section. The driver may accelerate through, disrupting the smooth rhythm of driving. If this kind of acceleration situation occurs frequently, it will affect the driver's driving experience.

[0006] In the prior art, there are mainly two ways for the leading target to cut out.

[0007] In the first target cut-out scheme, after the leading following target vehicle has completely cut out of the host vehicle lane by a certain distance, it is determined that the leading following target vehicle has cut out, and then the cut-out leading following target vehicle is released. The ACC determines whether to accelerate or decelerate based on the situation of the new leading following target vehicle in the host vehicle lane and performs longitudinal control. In the first scheme, the leading following target vehicle has to cut out a certain distance outside the host vehicle lane before it is considered a target cut-out. This often leads to a relatively late release of the cut-out leading following target vehicle, and the ACC controller cannot respond in advance. For a leading following target vehicle at a long distance, the release is late, which often causes the ACC to start accelerating relatively late. After the leading following target vehicle has cut out and there is no leading following target vehicle in the host vehicle lane, the slow release of the cut-out leading following target vehicle often results in relatively slow ACC driving, affecting the driver's experience. After the leading following target vehicle has cut out, there may still be other leading following target vehicles in the host vehicle lane; if the release of the cut-out leading following target vehicle is slow, it often leads to a response delay for the newly recognized leading following target vehicle. When deceleration is required for the newly recognized leading following target vehicle, it often results in a relatively large deceleration, and this particularly urgent braking deceleration will affect the driver's driving comfort and driving experience.

[0008] In the second target cut-out scheme, based on the lateral collision time between the host vehicle and the leading following target vehicle, the lateral distance between the two vehicles at this moment of the longitudinal collision time point, when a certain threshold is reached, it is determined that the target has cut out, the cut-out leading following target vehicle is released, and then the closest target is selected within the host vehicle lane for new longitudinal control. In the second scheme, the cut-out of the leading following target vehicle is determined based on the collision moment and the lateral distance between the two vehicles. The threshold of the collision time is not easy to adjust and requires a large amount of test calibration. And without considering the parameters of the lane line, it often leads to inaccurate judgment on curves. Also, since the motion trend of the leading vehicle is not considered and there is no judgment on whether the leading following target vehicle is cutting out, it often leads to misjudgment of the cut-out of the leading following target vehicle, thus posing a risk of vehicle collision.

[0009] In summary, the two existing cut-out schemes have problems of affecting driving comfort or being prone to collision risks. Summary of the Invention

[0010] The present invention aims to avoid the deficiencies in the above-mentioned existing technologies and provides an algorithm for releasing the leading following target vehicle based on the motion conditions of the host vehicle, so as to make the release of the following target more reliable and effective for the driver during driving following such as turning or lane changing, and make the ACC control smoother.

[0011] The present invention adopts the following technical solutions to solve the technical problems.

[0012] A following vehicle target vehicle release algorithm based on the motion condition of the host vehicle of the present invention includes the following steps: Step 1: Activate the ACC longitudinal control function of the host vehicle, and obtain the motion parameters of the host vehicle through the host vehicle control system; at the same time, obtain the parametric equation of the center line of the lane collected by the camera of the host vehicle, the width W of the host lane lane and the motion parameters of the following vehicle target vehicle in front; Step 2: According to the above-mentioned motion parameters of the host vehicle, the parametric equation of the center line of the lane collected by the camera, and the width W of the host lane lane , judge the motion trend of the host vehicle; Step 3: According to different situations of the motion trend of the host vehicle, optimize the lane lines of the lane collected by the camera to obtain a virtual optimized host lane; Step 4: According to the virtual optimized host lane, judge whether the following vehicle target vehicle in front of the host vehicle has a tendency to cut out of the lane collected by the camera; Step 5: After determining that the following vehicle target vehicle has a tendency to cut out of the lane collected by the camera, judge whether the following vehicle target vehicle meets the conditions for cutting out of the lane collected by the camera; Step 6: If the following vehicle target vehicle meets the conditions for cutting out of the lane collected by the camera, release the following vehicle target vehicle; Step 7: If the following vehicle target vehicle does not meet the conditions for cutting out of the lane collected by the camera, determine that the following vehicle target vehicle is in a cut-out state, and calculate the acceleration value required by the host vehicle to control the host vehicle.

[0013] The structural feature of a following vehicle target vehicle release algorithm based on the motion condition of the host vehicle of the present invention also lies in: Further, in the step 1, the motion parameters of the host vehicle include the longitudinal speed of the host vehicle v x_ego , longitudinal acceleration a x_ego , lateral speed v y_ego , lateral acceleration a y_ego , the width of the host vehicle W ego and the steering wheel angle of the host vehicle A ego .

[0014] Further, in the step 2, in the lane collected by the camera, according to the current value of the motion parameters of the host vehicle in front of the host vehicle, set the predicted motion position of the host vehicle after a predetermined time T as the preview position of the host vehicle; according to the initial position parameters of the host vehicle and the position parameters of the preview position of the host vehicle after the predetermined time period T, judge the motion trend of the host vehicle.

[0015] Further, the distance △y from the center position of the host vehicle at the starting position to the center line of the lane collected by the camera 0 and the distance △y from the center position of the host vehicle at the preview point to the center line of the lane collected by the camera preview and the steering wheel angle of the host vehicle A ego are used to judge the motion trend of the host vehicle.

[0016] Further, in the steps 2 and 3, the motion trend of the host vehicle includes stable left offset, stable right offset, centered driving, and lane change or turning.

[0017] Further, in the step 5, during the process of judging whether the leading following target vehicle meets the condition of cutting out of the lane collected by the camera, the overlap degree between the host vehicle and the leading following target vehicle and the lateral overlap degree between the leading following target vehicle and the host vehicle are judged.

[0018] The present invention also discloses an electronic device, including at least one processor and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the front following target vehicle release algorithm based on the motion conditions of the host vehicle.

[0019] The present invention also discloses a non-transitory computer-readable storage medium storing computer instructions, and the computer instructions are used to cause the computer to execute the front following target vehicle release algorithm based on the motion conditions of the host vehicle.

[0020] The present invention also discloses a computer program product, including a computer program; the computer program realizes the front following target vehicle release algorithm based on the motion conditions of the host vehicle when executed by a processor.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention discloses a front following target vehicle release algorithm based on the motion conditions of the host vehicle. First, the host vehicle motion parameters, the parametric equation of the center line of the lane collected by the camera, and the width W of the host lane are obtained through the host vehicle control system laneThe motion parameters of the leading vehicle ahead and the target vehicle; determine the motion trend of the host vehicle; obtain the virtual optimized self-lane of the host vehicle; determine whether the leading vehicle ahead of the host vehicle has a tendency to cut out of the camera-captured self-lane; determine whether the leading vehicle ahead meets the conditions for cutting out of the camera-captured self-lane; if the leading vehicle ahead meets the conditions for cutting out of the camera-captured self-lane, release the leading vehicle ahead; if the leading vehicle ahead does not meet the conditions for cutting out of the camera-captured self-lane, determine that the leading vehicle ahead is in a cut-out state, and calculate the acceleration value required by the host vehicle to control the host vehicle.

[0022] A leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention has the advantages that it makes the release of the leading vehicle ahead by the driver more reliable and effective during the process of following a vehicle while turning or changing lanes, and makes the ACC control smoother, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flowchart of the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0024] Figure 2 It is a schematic diagram in which the center line of the lane coincides with the center line of the host vehicle in the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0025] Figure 3 It is a schematic diagram of the current position of the host vehicle and the position of the host vehicle at the preview point in the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0026] Figure 4 It is a schematic diagram of several different situations such as the stable left offset, stable right offset, centered driving, and lane change or turning of the host vehicle in the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0027] Figure 5 It is a schematic diagram of the leading vehicle ahead cutting out to the right in the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0028] Figure 6 It is a schematic diagram of the leading vehicle ahead cutting out to the left in the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0029] Figure 7 It is a schematic diagram of the turning process of the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention.

[0030] Figure 8 It is the orientation angle of the leading vehicle ahead in the leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention Aobstacle Schematic diagram.

[0031] Figure 9 Lane centerline angle of the leading vehicle release algorithm for the leading vehicle following the host vehicle based on the driving conditions of the host vehicle according to the present invention A lane_obstacle Schematic diagram.

[0032] Figure 10 Structural block diagram of the state machine for ACC longitudinal control in the prior art.

[0033] The present invention will be further described below through specific embodiments in conjunction with the accompanying drawings. Specific embodiments

[0034] See Figures 1 - 9 A leading vehicle release algorithm for the leading vehicle following the host vehicle based on the driving conditions of the host vehicle according to the present invention includes the following steps: Step 1: Activate the ACC longitudinal control function of the host vehicle, and obtain the motion parameters of the host vehicle through the host vehicle control system; at the same time, obtain the parametric equation of the center line of the lane collected by the camera from the lane, the width W of the host lane lane and the motion parameters of the leading vehicle following the host vehicle; The host vehicle control system obtains the video and images of the real host lane (the real road lane where the host vehicle is driving) through the camera; the camera sends the captured road images to the host vehicle control system, and the host vehicle control system processes the road images. Through road image processing, the parametric equation of the left lane line of the lane collected by the camera (the electronic virtual lane corresponding to the real host lane after processing the image of the host lane collected by the camera and stored in the host vehicle control system) and the parametric equation of the right lane line of the lane collected by the camera are obtained, and then the parametric equation of the center line of the lane collected by the camera and the width W of the host lane collected by the camera are obtained according to the parametric equation of the left lane line of the lane collected by the camera and the parametric equation of the right lane line of the lane collected by the camera; at the same time, obtain the motion parameters of the leading vehicle following the host vehicle through the camera. lane ; at the same time, obtain the motion parameters of the leading vehicle following the host vehicle through the camera.

[0035] In order to make the cut-out of the leading vehicle in front more timely and accurate, first, based on the parameters of the lane captured by the sensor (camera) and the relevant motion parameters of the host vehicle, it is necessary to determine whether the host vehicle is driving stably in the central area of the lane captured by the camera or driving stably at a certain distance offset from the center line of the lane captured by the camera. When the host vehicle is stably offset within the lane captured by the camera, it is necessary to optimize the lane captured by the camera to obtain a virtual optimized lane: add 0.4 m to the vehicle width as the width of the virtual optimized lane, and use the center line of the host vehicle as the center line of the virtual optimized lane; use the virtual optimized lane as the basis for the leading vehicle in front to cut out of the lane captured by the camera; first judge that the leading vehicle in front has a tendency to cut out, and then judge that the longitudinal collision point between the host vehicle and the leading vehicle in front and the lateral distance of the leading vehicle in front meet a certain threshold, and then judge that the leading vehicle in front is the cut-out target, that is, judge that the leading vehicle in front is about to cut out of the lane captured by the camera.

[0036] The parametric equations of the left lane line of the lane captured by the camera, the parametric equations of the right lane line of the lane captured by the camera, and the parametric equations of the lane center line of the lane captured by the camera, and several other parametric equations are all expressed by the cubic equation of the following formula (1).

[0037] (1); In formula (1), y is the dependent variable of the center line equation of the lane captured by the camera, x is the independent variable, a 0 is the constant term of the center line equation of the lane captured by the camera, a 1 ~a 3 are the coefficients of the first term to the third term of the center line equation of the lane captured by the camera respectively. In the present invention, the center line of the virtual optimized lane is also expressed by formula (1).

[0038] In the present invention, in the process of calculating the coordinates and equations of the host vehicle, the following host vehicle coordinate system is established at the initial moment: the forward direction of the host vehicle in the real host lane is the X direction, the transverse direction of the host vehicle body perpendicular to the X direction from left to right is the Y direction, and the projection point of the center point of the rear axle of the host vehicle on the ground of the real host lane is the origin O, as Figure 2 shown; the origin O is also the center position of the host vehicle and moves with the movement of the host vehicle. Similarly, the host vehicle coordinate system in the present invention also moves with the movement of the host vehicle and is a moving coordinate system. The center position of the host vehicle is the projection point of the center point of the rear axle of the host vehicle on the ground of the real host lane.

[0039] Step 2: According to the above-mentioned host vehicle motion parameters, the parametric equation of the center line of the lane captured by the camera, and the host lane width W lane , judge the motion trend of the host vehicle; Based on data such as the current values of the ego vehicle's motion parameters obtained in step 1, a constant acceleration model is used to further estimate the future motion position of the ego vehicle and determine the motion trend of the ego vehicle. According to the above parameters, determine whether the motion trend of the ego vehicle is stable left offset, stable right offset, centered driving, lane change or turning.

[0040] Step 3: According to different situations of the motion trend of the ego vehicle, optimize the lane lines of the lane collected by the camera to obtain a virtual optimized ego lane; According to different working conditions such as stable left offset, stable right offset, centered driving, lane change or turning, different rules are used to determine the virtual optimized ego lane under different working conditions; the lane lines of stable offset are adopted and the estimated driving trajectory of the ego vehicle is used. Finally, the parameters of the lane center line equation are updated, and the width of the lane lines of the ego lane is updated according to different working conditions. After the update, the virtual optimized ego lane is obtained.

[0041] Step 4: According to the virtual optimized ego lane, determine whether the leading vehicle in front of the ego vehicle has a tendency to cut out of the lane collected by the camera; For the leading vehicle in front, the method of estimating time is used to calculate whether the leading vehicle in front has a tendency to cut out of the lane collected by the camera, that is, to determine whether the leading vehicle in front has a tendency to change lanes and leave the lane collected by the camera; Step 5: After determining that the leading vehicle in front has a tendency to cut out of the lane collected by the camera, determine whether the leading vehicle in front meets the conditions for cutting out of the lane collected by the camera; After determining that the leading vehicle in front has a tendency to change lanes and leave the lane collected by the camera, it is also necessary to determine that the leading vehicle in front meets the conditions for changing lanes and leaving the lane collected by the camera before it can be determined that the leading vehicle in front has driven out of the lane collected by the camera; Step 6: If the leading vehicle in front meets the conditions for cutting out of the lane collected by the camera, release the leading vehicle in front; According to the motion parameters of the ego vehicle and the leading vehicle in front, when it is determined that the leading vehicle in front meets the cut-out state conditions, it is determined that the leading vehicle in front enters the state of cutting out of the lane collected by the camera. When the leading vehicle in front meets the conditions for completely cutting out of the lane collected by the camera, release the leading vehicle in front; that is, it is determined that the leading vehicle in front has driven out of the lane collected by the camera, and there is no need to use the leading vehicle in front as a following target anymore. If there are no other vehicles on the road ahead for a certain distance, acceleration can be considered. In this way, it can be immediately judged after the leading vehicle in front has left the real ego lane, and then accelerate immediately, avoiding the situation of accelerating a long time after the leading vehicle in front has changed lanes and left from the front, and improving the driving experience of the driver.

[0042] Step 7: If the leading following target vehicle does not meet the condition of cutting out of the lane captured by the camera, it is determined that the leading following target vehicle is in the cut-out state, and the required acceleration value of the host vehicle is calculated to control the host vehicle.

[0043] When the leading following target vehicle enters the cut-out state, the required acceleration value calculated by screening a new following target according to the target screening algorithm (or there is no following target and enters the cruise control state) and the required acceleration value of the host vehicle calculated from the leading following target vehicle in the cut-out state are weighted and summed as the final requested acceleration to longitudinally control the host vehicle.

[0044] An algorithm for releasing the leading following target vehicle based on the motion conditions of the host vehicle in the present invention enables the host vehicle to make a more accurate judgment on the cut-out of the leading following target vehicle after the ACC function is turned on. Considering the motion of the host vehicle, first optimize the lane captured by the camera, and use the optimized virtual optimized lane as the basis for judging the target cut-out. Then calculate the lateral speed of the leading following target vehicle relative to the left and right lane lines of the lane captured by the camera. After predicting according to the host vehicle speed for a certain period of time, if the leading following target vehicle cuts out of the optimized virtual optimized lane, and then considering that at the longitudinal collision moment of the host vehicle and the leading following target vehicle, the lateral distance between the host vehicle and the leading following target vehicle meets a certain threshold, it is determined that the leading following target vehicle has cut out of the lane captured by the camera. This algorithm optimization can make the judgment of the leading following target vehicle cutting out more in line with the actual intention of the driver and make the control of ACC safer and more reliable.

[0045] During specific implementation, in the said Step 1, the motion parameters of the host vehicle include the longitudinal speed of the host vehicle v x_ego , longitudinal acceleration a x_ego , lateral speed v y_ego , lateral acceleration a y_ego , the width of the host vehicle W ego and the steering wheel angle of the host vehicle A ego .

[0046] The coordinate position of the host vehicle at the current moment (0 moment) is calculated through the said formula (1); in the initial state at the 0 moment, the coordinates of the center position of the host vehicle are ( x 0_ego , y 0_ego ), define x 0_ego = 0, y 0_ego= 0, substituting x = 0 into formula (1) gives the position of the center line of the lane captured by the camera as y 0_lane = a 0 ; as Figure 2 shown in the middle image, at this time the center line of the lane captured by the camera coincides with the center line of the vehicle itself, that is y 0_lane = a 0 = 0, the vehicle itself is driving in the center of the lane captured by the camera without deviation; if the vehicle itself deviates to the left (such as Figure 2 the left image) or to the right (such as Figure 2 the right image), then the X-axis deviates to the left or right along with the vehicle body, and the center line of the lane captured by the camera will not coincide with the center line of the vehicle itself, then y 0_lane = a 0 ≠ 0, the vehicle itself deviates to the left or right in the lane captured by the camera, and the deviation distance is ︱a 0 ︱ = ︱ y 0_lane ︱; y 0_lane > 0 means the vehicle itself deviates to the left in the lane captured by the camera (such as Figure 2 in the left image, △y 0 = - y 0_lane < 0), y 0_lane < 0 means the vehicle itself deviates to the right in the lane captured by the camera (such as Figure 2 in the right image, △y 0 = - y 0_lane > 0).

[0047] Based on this, at the starting position, the distance △y between the center position of the vehicle itself and the center line of the lane captured by the camera 0 is shown in the following formula (2).

[0048] (2).

[0049] △y 0 In the case where the vehicle itself deviates to the left or right in the lane captured by the camera as Figure 2 shown.

[0050] Specifically, in step 2, in the lane captured by the camera, according to the current value of the vehicle movement parameters in front of the vehicle itself, the predicted movement position of the vehicle itself after a predetermined time T is set as the pre-view position of the vehicle; according to the initial position parameters of the vehicle itself and the position parameters of the pre-view position of the vehicle itself after a predetermined time period T, the movement trend of the vehicle itself is judged.

[0051] In the specific implementation of the present invention, the predetermined time period T is set to 1 second, that is, the preview time is set to 1 second. According to the current vehicle speed of the host vehicle and using a constant acceleration model, the position of the host vehicle moving forward in the lane captured by the camera for another 1 s is determined, and this position is used as the preview position of the host vehicle, that is, the position where the host vehicle continues to travel in its current driving state after 1 s, as shown in Figure 3 shown.

[0052] In the present invention, the preview point of the host vehicle refers to the projection point of the center point of the rear axle of the host vehicle on the ground in the lane captured by the camera at the preview position of the host vehicle. According to the position coordinates of the preview point and the distances between the preview point and the left lane line of the lane captured by the camera and the right lane line of the host vehicle lane, etc., the movement trend of the host vehicle is judged. In order to avoid the situation that when the host vehicle is at a low speed, the preview position 1 s ahead is too close to the current position of the host vehicle, resulting in a deviation in the calculation of the movement trend of the host vehicle, the minimum value of the longitudinal distance between the preview point of the host vehicle and the current center position of the host vehicle is set to 10 m. When the host vehicle is moving at a low speed and the longitudinal distance between the preview position of the host vehicle after moving for 1 s and the center position of the host vehicle 1 s ago is less than 10 m, at this time, the front position point where the longitudinal distance from the center position of the host vehicle 1 s ago is 10 m is determined as the preview point.

[0053] As shown in Figure 3 , the longitudinal position of the host vehicle in the X direction after t = 1 s x ego_preview is calculated using the following formula (3).

[0054] (3); Substituting t = 1 into formula (3) gives formula (4), and the calculated value of the longitudinal position x ego_preview is obtained.

[0055] (4); At the same time, comparing the longitudinal position x ego_preview calculated by formula (4) with 10 m, taking the maximum value between x ego_preview and 10 m as the final value of the longitudinal position of the preview point x ego_preview_final , as shown in the following formula (5).

[0056] (5); The lateral position of the host vehicle after t = 1 s y ego_preview is calculated using the following formula (6).

[0057] (6); Substitute \(t = 1\) into formula (6) to obtain formula (7), and get the calculated value of the lateral position y ego_preview .

[0058] (7); Therefore, at the moment \(t = 1s\), which is 1s after the initial moment \(t = 0\), the position of the center line of the lane collected by the camera at the preview point y lane_preview is shown in the following formula (8).

[0059] (8); Therefore, the distance \(\Delta y\) between the center position of the host vehicle at the preview point and the center line of the lane collected by the camera can be obtained preview as shown in the following formula (9).

[0060] (9); In specific implementation, based on the distance \(\Delta y\) between the center position of the host vehicle at the starting position and the center line of the lane collected by the camera 0 , the distance \(\Delta y\) between the center position of the host vehicle at the preview point and the center line of the lane collected by the camera preview and the steering wheel angle of the host vehicle A ego to judge the movement trend of the host vehicle.

[0061] In specific implementation, use the parameters calculated in the above formulas to judge the movement trend of the host vehicle. These parameters include but are not limited to the distance \(\Delta y\) between the center position of the host vehicle at the starting position and the center line of the lane collected by the camera 0 , the distance \(\Delta y\) between the center position of the host vehicle at the preview point and the center line of the lane collected by the camera preview and the steering wheel angle of the host vehicle A ego and so on.

[0062] In specific implementation, in steps 2 and 3, the movement trend of the host vehicle includes stable left offset, stable right offset, driving in the center, and lane change or turning.

[0063] In specific implementation, in step 5, during the process of judging whether the leading following target vehicle meets the condition of cutting out of the lane collected by the camera, judge the overlap degree between the host vehicle and the leading following target vehicle and the lateral overlap degree between the leading following target vehicle and the host vehicle.

[0064] The present invention also discloses an electronic device, including at least one processor and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the above-mentioned forward following target vehicle release algorithm based on the motion condition of the host vehicle.

[0065] The present invention also discloses a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to cause the computer to execute the above-mentioned forward following target vehicle release algorithm based on the motion condition of the host vehicle.

[0066] The present invention also discloses a computer program product, including a computer program; when the computer program is executed by a processor, the above-mentioned forward following target vehicle release algorithm based on the motion condition of the host vehicle is implemented.

[0067] The determination of the motion trend of the host vehicle is divided into four different situations: stable left offset, stable right offset, driving in the center, and lane change or turning, as Figure 4 shown.

[0068] 1. Stable left offset; when the host vehicle is in the initial state △y 0 <0, and at the preview point, the host vehicle is within a certain range to the left of the center line of the host lane collected by the camera, △y preview <-0.5, and the absolute value of the steering wheel angle A ego <30°, it is considered that the host vehicle is in stable left offset; as Figure 4 shown in the leftmost picture, the entire host vehicle ego car is to the left of the center line of the host lane collected by the camera.

[0069] When the host vehicle is in stable left offset, at this time the host vehicle is to the left of the center line of the host lane collected by the camera. If the center line of the host lane collected by the camera is used as the center line of the host vehicle at this time, there will be a large deviation from the actual position of the host vehicle. Therefore, at this time, it is necessary to optimize and reconfirm the virtual optimized host lane to make the virtual optimized host lane more in line with the actual situation of the host vehicle itself.

[0070] At this time, the optimization strategy for the virtual optimized own lane is as follows: Since the own vehicle is stably offset to the left within the lane captured by the camera, it is necessary to shift both the left and right lane lines of the virtual optimized own lane to the left to ensure that the own vehicle travels in the middle within the left and right lane lines of the virtual optimized own lane; when the leading vehicle following target vehicle cuts out to the right side of the lane captured by the camera, since the own vehicle is traveling on the left side within the lane captured by the camera, there is no collision risk with the leading vehicle following target vehicle that cuts out to the right side area within the lane captured by the camera. Therefore, this leading vehicle following target vehicle can be quickly released. Under the motion condition of such stable left offset of the own vehicle, it is appropriate to shift the virtual optimized own lane as a whole to the left. At the same time, since the own vehicle is traveling on the left side within the lane captured by the camera, when the leading vehicle following target vehicle cuts out to the left side of the lane captured by the camera, the leading vehicle following target vehicle needs to be released relatively slowly to prevent the own vehicle from accelerating and colliding with the leading vehicle following target vehicle because the leading vehicle following target vehicle is released prematurely when it cuts out to the vicinity of the left side within the lane captured by the camera. The left shift of the virtual optimized own lane is mainly achieved by the following method, considering the offset △y of the starting position of the own vehicle at the starting moment 0 and the offset △y at the preview point preview . These two values can reflect the severity of the lateral offset of the own vehicle within the lane captured by the camera. Therefore, the center line equation of the virtual optimized own lane after the left offset can be expressed by the following formula (10).

[0071] (10); The width of the lane lines of the virtual optimized own lane also needs to be optimized. In the present invention, the width of the virtual optimized own lane is optimized by adding 0.2 m to one side of the own vehicle width. Using this method is mainly to prevent giving the virtual optimized own lane too wide a lane, resulting in too slow release when the leading vehicle following target vehicle cuts out; using the own vehicle width W ego as the benchmark for the width of the virtual optimized own lane will make the release of the leading vehicle following target vehicle more accurate. And because the entire virtual optimized own lane is shifted to the left, reducing the width of the virtual optimized own lane can optimize the release timing of the leading vehicle following target vehicle. The optimized width of the virtual optimized own lane W lane is: ( W lane = W ego + 0.4) meters.

[0072] 2. Stable right offset; When the own vehicle is in the initial state △y 0 > 0, and at the preview point, the own vehicle is within a certain range on the left side of the lane line, △y preview > 0.5, and the absolute value A of the steering wheel angle ego <30°, then it is considered that the own vehicle is in stable right offset. For exampleFigure 4 As shown in the middle figure, the entire ego car is on the right side of the center line of the lane captured by the camera.

[0073] When the ego car is stably offset to the right, at this time the ego car is on the right side of the center line of the lane captured by the camera. If the center line of the lane captured by the camera is used as the center line of the ego car at this time, there will be a large deviation from the actual position of the ego car. Therefore, at this time, it is necessary to optimize and reconfirm the virtual optimized ego lane to make it more in line with the actual situation of the ego car itself.

[0074] When the ego car is stably offset to the right, the entire virtual optimized ego lane needs to be shifted to the right; when the ego car is driving on the right side within the lane captured by the camera and the leading vehicle target vehicle in front cuts out to a certain area on the left side of the lane captured by the camera, since the ego car is driving on the right side within the lane captured by the camera, there is no collision risk with the leading vehicle target vehicle in the left area within the lane captured by the camera. Therefore, the leading vehicle target vehicle in front can be released. In this motion condition of the ego car, in order to achieve this purpose, it is appropriate to shift the virtual optimized ego lane to the right. At the same time, since the ego car is driving on the right side within the lane captured by the camera, when the leading vehicle target vehicle in front cuts out to the right area of the lane captured by the camera, the release of the leading vehicle target vehicle in front needs to be slower to prevent the ego car and the leading vehicle target vehicle in front from colliding in the right area within the lane captured by the camera. The right shift of the virtual optimized ego lane is mainly achieved by the following method, considering the offset △y of the starting position of the ego car at the starting moment 0 and the offset △y at the preview point preview , these two values can reflect the severity of the lateral offset of the ego car within the lane captured by the camera. Therefore, the center line equation of the virtual optimized ego lane after the right shift can be expressed by the following formula (11) (the same as formula 10 for the left shift above).

[0075] (11); The width of the virtual optimized ego lane also needs to be optimized: The width of the virtual optimized ego lane is optimized by adding 0.2m to one side of the ego car width. Using this method is mainly to prevent giving too wide a lane, which is likely to cause the release of the leading vehicle target vehicle in front to be too slow. Using the width of the ego car W ego as the benchmark for the width of the virtual optimized ego lane will make the release of the leading vehicle target vehicle in front more accurate. Because the entire virtual optimized ego lane is shifted to the right and the width of the virtual optimized ego lane is reduced, the release timing of the leading vehicle target vehicle in front can be optimized. The lane width of the virtual optimized ego lane after optimization W lane is: ( W lane = W ego + 0.4) meters.

[0076] 3. Center driving; when the vehicle has no stable left offset and no stable right offset, and the absolute value of the steering wheel angle A ego < 30°, it is considered that the vehicle is driving in the center of the lane captured by the camera in its own lane. As Figure 2 shown, at this time, the center line of the vehicle coincides with the center line of the lane captured by the camera in its own lane.

[0077] When the vehicle is driving in the center, the lane captured by the camera in its own lane is not optimized, and the center line of the lane captured by the camera is directly used as the center line of the vehicle (the two coincide at this time). That is, using formula (1), the width of the lane captured by the camera in its own lane is the actual road lane width detected by the camera W lane . At this time, W lane is not equal to W ego + 0.4.

[0078] 4. Lane change or turn; when the absolute value of the steering wheel angle of the vehicle A ego ≥ 30°, it is considered that the vehicle is stably changing lanes or turning (as Figure 7 shown).

[0079] When the vehicle is changing lanes or turning, since the direction of the vehicle changes relatively quickly at this time, and the process of changing lanes or turning is often relatively short, a lateral constant acceleration model is adopted, and the center line of the vehicle is used as the center line of the virtual optimized lane in its own lane. See the following formula (12).

[0080] (12); Since the movement trajectory line of the vehicle changes relatively quickly during the process of changing lanes or turning, in order to prevent the misrelease of the following target vehicle in front, the width of the virtual optimized lane in its own lane W lane directly uses the width of the vehicle W ego , that is: W lane = W ego .

[0081] In the present invention, the virtual optimized lane in its own lane is a virtual lane line adjusted according to the movement of the vehicle, which is different from the lane captured by the camera in its own lane obtained by the vehicle control system through the camera (only when driving in the center, the virtual optimized lane in its own lane is the lane captured by the camera). When the vehicle is offset left and right in the lane captured by the camera in its own lane, the center line equation can be expressed by the following formula (10) and formula (11), and the width of the virtual optimized lane in its own lane Wlane The width of the host vehicle W ego Add 0.2 meters to each of the left and right sides, that is: W lane = W ego + 0.4

[0082] As Figure 4 shown, using the optimized virtual self-lane, the algorithm principle for releasing the leading vehicle target vehicle ahead is: obtain the parameters of the leading vehicle target vehicle ahead from the camera, and use the target screening algorithm to obtain the motion parameters of the leading vehicle target vehicle ahead: longitudinal speed v x_obstacle , lateral speed v y_obstacle , longitudinal position x obstacle , lateral position y obstacle , longitudinal acceleration a xobstacle , vehicle width W obstacle , then the orientation angle A obstacle of the leading vehicle target vehicle ahead is calculated by the following formula (13), and the orientation angle A obstacle is as Figure 8 shown

[0083] (13); The method for calculating the angle of the center line of the lane at the position of the leading vehicle target vehicle ahead is: use the longitudinal position of the following vehicle target to find the point on the center line of the lane parallel to the target longitudinal position, and then take the position parameters of the points 1m ahead and 1m behind longitudinally. That is, substitute x obstacle +1 and x obstacle -1 into the above formulas (10), (11), and (12) respectively according to the motion conditions of the host vehicle, and obtain y lane_obstacle+1 and y lane_obstacle-1 respectively. Then, the angle A lane_obstacle of the center line of the lane near the leading vehicle target vehicle is calculated by the following formula (14), and the angle A lane_obstacle is as Figure 9 shown

[0084] (14); Then the included angle between the leading vehicle target vehicle in front and the center line of the lane where the camera collects images A obstacle2lane The calculation formula is shown in the following formula (15).

[0085] (15); Then the lateral speed of the center position of the leading vehicle target vehicle in front (the center position of the leading vehicle target vehicle in front is the projection point of the center point of the rear axle of the host vehicle on the ground of the real host lane) to the center line of the lane where the camera collects images v y_obstacle2lane The calculation formula is shown in the following formula (16).

[0086] (16); After obtaining the corrected motion parameters of the leading vehicle target vehicle in front, according to the above relevant parameters of the leading vehicle target vehicle in front, the release algorithm of the leading vehicle target vehicle in front is judged.

[0087] The specific calculation process of the release algorithm of the leading vehicle target vehicle in front is as follows: Detect the lateral speed of the leading vehicle target vehicle in front v y_obstacle ; When detecting that the lateral speed of the leading vehicle target vehicle in front v y_obstacle is greater than a certain threshold value (in one embodiment, this threshold value is set to 0.01 m / s), that is v y_obstacle > 0.01 m / s, then it is determined that the leading vehicle target vehicle in front is cutting out to the right side of the lane where the camera collects images. Then, it is estimated whether the leading vehicle target vehicle in front cuts out of the lane where the camera collects images after 1 s (a certain compensation can be made according to the cut-out lateral speed). Since the lateral acceleration of the leading vehicle target vehicle in front is generally small and the error is relatively large, in the present invention, for the leading vehicle target vehicle in front, a lateral uniform speed model is used to calculate the longitudinal position x obstacle_preview and the lateral position y obstacle_preview of the leading vehicle target vehicle in front.

[0088] At the preview point of the leading vehicle target vehicle in front (the calculation method of the preview point of the leading vehicle target vehicle in front is similar to that of the above-mentioned preview point of the host vehicle, but it should be calculated based on the leading vehicle target vehicle in front), the longitudinal position x obstacle_preview of the leading vehicle target vehicle in front is shown in the following formula (17).

[0089] (17); The longitudinal position xobstacle_preview Substitute the equation of formula (1), formula (10) or formula (11) for the center line of the lane collected by the camera after correction (judged according to the movement trend, if the host vehicle has no offset, use the center line of the lane collected by the camera, and when the host vehicle has an offset, use the center line calculated after the offset) to obtain the lateral position of the center line of the lane collected by the camera at the preview point y obstacle_preview_lane .

[0090] At the preview point, the lateral position of the leading vehicle following target vehicle y obstacle_preview The calculation formula is shown in the following formula (18).

[0091] (18); Then at the preview point, the distance △ between the center position of the leading vehicle following target vehicle (the center position of the leading vehicle following target vehicle is the projection point of the center point of the rear axle of the host vehicle on the ground of the true host lane) and the center line of the lane collected by the camera y obstacle_preview_lane The calculation formula is shown in the following formula (19).

[0092] (19).

[0093] If the position of the leading vehicle following target vehicle after 1 s meets the following conditions: , it is considered that the leading vehicle following target vehicle has a tendency to cut out of the lane collected by the camera.

[0094] After determining that the leading vehicle following target vehicle has a tendency to cut out of this lane, then judge whether it meets the cut-out release conditions. The judgment conditions include the following two: the first judgment condition and the second judgment condition.

[0095] 1. As Figure 5 , in the first judgment condition, judge through the overlap degree of the host vehicle and the leading vehicle following target vehicle: judge the overlap degree of the leading vehicle following target vehicle and the host vehicle. The distance from the center position of the leading vehicle following target vehicle (the center position of the leading vehicle following target vehicle is the projection point of the center point of the rear axle of the host vehicle on the ground of the true host lane) to the center line of the lane collected by the camera is greater than half of the width of the host vehicle W ego plus half of the width of the leading vehicle following target vehicle W obstacle , as shown in the following formula (21); substitute the longitudinal position of the leading vehicle following target vehicle x obstacle into the equation of formula (1), formula (10) or formula (11) for the corrected center line of the lane to obtain the lateral position of the center line of the lane at the preview point y obstacle_lane。The lateral position of the leading vehicle being followed is y obstacle 。Then, at the center position of the leading vehicle being followed (the center position of the leading vehicle being followed is the projection point of the center point of the rear axle of the host vehicle on the ground in the true host lane), the distance △ between the center position of the leading vehicle being followed and the center line of the lane where the camera captures the host lane y obstacle_lane The calculation formula is shown in the following formula (20). As Figure 5 shown.

[0096] (20); (21); In the first judgment condition, according to formula (21), judge the overlap degree between the leading vehicle being followed and the host vehicle. When the above formula (21) is satisfied, judge that the overlap degree between the leading vehicle being followed and the host vehicle is 0, that is, the leading vehicle being followed and the host vehicle do not overlap in the Y direction.

[0097] 2. As Figure 6 , in the second judgment condition, since the leading vehicle being followed is cutting out to the right side of the lane, it is only necessary to judge that when the host vehicle moves to the current longitudinal position of the leading vehicle being followed according to the current motion parameters, there is no overlap in the lateral position between the host vehicle and the leading vehicle being followed.

[0098] (22) (23); Obtain the longitudinal distance △s between the host vehicle and the leading vehicle being followed at the current moment through the host vehicle control system. The host vehicle adopts a constant acceleration model, substitute △s into formula (22), and the solution of t is △t. △t is the time required for the host vehicle to move from the current moment to the current position of the leading vehicle being followed; substitute △t as the time t into formula (23), and the lateral position of the host vehicle moving to the current position of the leading vehicle being followed after △t is △ y ego= y ego preview (see formula 6 and formula 23), and then judge the lateral overlap degree between the leading vehicle being followed and the host vehicle.

[0099] (24); (25); In the second judgment condition, according to formula (25), judge the lateral overlap degree from the leading vehicle being followed to the host vehicle (when the second judgment condition is met, it is determined that the lateral overlap degree between the leading vehicle being followed and the host vehicle is 0).

[0100] In summary, the conditions for determining that the leading vehicle ahead is in the state of cutting out on the right side are: satisfying the second judgment condition but not satisfying the first judgment condition. The condition for the leading vehicle ahead to completely cut out to the right is to satisfy the first judgment condition.

[0101] The above is the algorithm for the leading vehicle ahead to cut out on the right side. The algorithm for the leading vehicle ahead to cut out on the left side is similar to the above-mentioned right-side cutting-out algorithm, except that the moving direction of the leading vehicle ahead is different, and the moving direction is adjusted from right to left.

[0102] In step 7, in order to prevent the ACC controller from releasing the target too quickly when judging that the leading vehicle ahead is cutting out, resulting in a relatively large acceleration and a collision risk with the leading vehicle ahead, the control algorithm of the ACC needs to perform a weighted processing on the output acceleration when the state of the leading vehicle ahead is the cutting-out state. The calculation formula is shown in the following formula (26).

[0103] (26); In formula (26), a requst_final is the final acceleration requested by the ACC controller; a requst_withoutcutout is the requested acceleration calculated by the ACC controller after releasing the leading vehicle ahead that has cut out and screening a new leading vehicle according to the target screening algorithm (or there is no leading vehicle and enters the constant-speed motion); a requst_cutout is the acceleration required by the ACC controller according to the leading vehicle ahead that has cut out; q is the weight coefficient, and 0 ≤ q ≤ 1. The acceleration weight q of the leading vehicle ahead that has cut out can be calibrated and optimized according to the driving experience of the actual vehicle when the leading vehicle ahead cuts out, so that the release of the leading vehicle ahead is smoother and more comfortable.

[0104] When the leading vehicle ahead completely cuts out of the lane, the leading vehicle ahead that has cut out is released, and the ACC controller performs longitudinal control according to the leading vehicle ahead captured by the camera in the lane.

[0105] The leading vehicle ahead release algorithm based on the motion conditions of the host vehicle of the present invention has been proven by experiments that no matter whether the host vehicle is driving in the center of the lane captured by the camera, stably offset to the left, stably offset to the right, or turning or changing lanes, it can accurately release the leading vehicle ahead and can predict in advance the cutting out of the leading vehicle ahead, and give the information of the leading vehicle ahead that has cut out in advance to the ACC controller, making the control of the ACC controller more sensitive and the longitudinal control of the host vehicle more comfortable. After adopting this algorithm, it is possible to avoid the situation that the host vehicle accelerates slowly when the leading vehicle ahead cuts out.

[0106] A following vehicle target vehicle release algorithm based on the driving conditions of the host vehicle according to the present invention has the following technical characteristics through the analysis of the above technical solutions and the verification of the actual vehicle effects.

[0107] 1. It improves the intelligence of the ACC longitudinal control function, can identify the cut-out of the following vehicle target vehicle in front in advance, and makes the system more in line with the driver's operation intention; 2. It improves the comfort of the ACC longitudinal control function. The ACC controller can respond to the cut-out of the following vehicle target vehicle in front in advance, making the acceleration of the host vehicle more comfortable and preventing the ACC from suddenly braking the host vehicle due to the sudden cut-out of the following vehicle target vehicle in front; 3. It improves the accuracy of the ACC longitudinal control target recognition. By using the estimated time method, the judgment and recognition of the cut-out of the following vehicle target vehicle in front are more accurate; 4. It improves the usage efficiency of the ACC longitudinal function. By considering the movement trend of the host vehicle, the lane lines of the host vehicle are optimized according to the movement trend of the host vehicle, making the judgment of the cut-out of the following vehicle target vehicle in front more efficient, avoiding the deceleration of the host vehicle caused by misjudging the cut-out of the following vehicle target vehicle in front, making the entire longitudinal control more comfortable, and increasing the number of times and usage efficiency of the driver turning on the ACC function.

[0108] 5. It improves the safety of the ACC system. Whether on a straight road or a curve, it can accurately identify the cut-out of the following vehicle target vehicle in front, improving the safety of the system.

[0109] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

[0110] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An algorithm for releasing a target vehicle in front of the vehicle based on the vehicle's motion condition, characterized in that: The steps include: Step 1: Activate the ACC longitudinal control function of the vehicle and obtain the vehicle motion parameters through the vehicle control system; at the same time, obtain the parameter equation of the center line of the lane and the lane width W through the camera of the vehicle. lane and the motion parameters of the target vehicle ahead; Step 2: Based on the above vehicle motion parameters, the parametric equation of the center line of the lane captured by the camera and the lane width W lane , determine the movement trend of the vehicle; Step 3: According to different movement trends of the ego vehicle, the lane line of the ego lane captured by the camera is optimized to obtain a virtual optimized ego lane; Step 4: Based on the virtual optimized lane, determine whether the target vehicle in front of the vehicle has a tendency to cut out of the camera-collected lane; Step 5: After determining that the target vehicle ahead has a tendency to cut out of the lane for camera collection, determine whether the target vehicle ahead meets the conditions for cutting out of the lane for camera collection; Step 6: If the target vehicle in front meets the conditions for cutting out the camera acquisition lane, release the target vehicle in front; Step 7: If the target vehicle in front does not meet the conditions for cutting out the camera to collect the own lane, it is determined that the target vehicle in front is in a cut-out state, and the acceleration value required by the ego vehicle is calculated to control the ego vehicle.

2. The method for releasing a target vehicle in front of the vehicle based on the vehicle's motion condition according to claim 1 is characterized in that: In step 1, the vehicle motion parameters include the longitudinal velocity of the vehicle v x_ego , longitudinal acceleration a x_ego , lateral speed v y_ego , lateral acceleration a y_ego , Vehicle width W ego and the steering wheel angle of the vehicle A ego .

3. The release algorithm of the front following target vehicle based on the vehicle motion condition according to claim 2 is characterized in that: In step 2, the camera captures the image in the lane, and in front of the vehicle, according to the current value of the vehicle's motion parameter, sets the expected movement position of the vehicle after a predetermined time T as the preview position of the vehicle; based on the initial position parameters of the vehicle and the position parameters of the preview position of the vehicle after a predetermined time period T, the movement trend of the vehicle is determined.

4. The method for releasing a target vehicle in front of the vehicle based on the vehicle's motion condition according to claim 3 is characterized in that: The distance between the center of the vehicle passing through the starting position and the center line of the lane captured by the camera is △y0, and the distance between the center of the vehicle at the preview point and the center line of the lane captured by the camera is △y preview and the steering wheel angle of the vehicle A ego To determine the movement trend of the vehicle.

5. The front following target vehicle release algorithm based on the vehicle motion condition according to claim 1 is characterized in that: In step 2 and step 3, the movement trend of the vehicle includes stable left deviation, stable right deviation, centering and lane changing or turning.

6. The front following target vehicle release algorithm based on the vehicle motion condition according to claim 1 is characterized in that: In step 5, in the process of determining whether the target vehicle ahead meets the conditions for cutting out the camera to capture the lane ahead, the overlap between the vehicle ahead and the target vehicle ahead and the lateral overlap between the vehicle ahead and the target vehicle ahead are determined.

7. An electronic device comprising at least one processor and a memory connected to the at least one processor in communication; wherein: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the front following target vehicle release algorithm based on the vehicle motion condition as described in any one of claims 1-6.

8. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable the computer to execute the front following target vehicle release algorithm based on the vehicle motion condition according to any one of claims 1-6.

9. A computer program product comprising a computer program; wherein: When the computer program is executed by a processor, the computer program implements a front following target vehicle release algorithm based on the vehicle motion condition according to any one of claims 1 to 6.

Images