Vehicle lane changing control method and related device

Abstract

The invention provides a vehicle lane changing control method and a related device, and relates to the technical field of automatic driving. According to the invention, during vehicle lane changing, the current lane changing scene is identified based on the road information of the current road where the vehicle to be subjected to lane changing is located, and the target lane changing scene is obtained, so that subsequent lane changing can be matched with an actual road scene, and the lane changing accuracy and reliability are improved. In addition, according to the method, the target lane changing space of the vehicle to be subjected to lane changing is determined, longitudinal lane changing motion data in the target lane changing scene are calculated, and the vehicle to be subjected to lane changing is controlled to conduct lane changing operation according to the target lane changing space and based on longitudinal lane changing reference data and transverse lane changing motion data in the target lane changing scene. According to the method, the lane changing reference data are determined from the transverse angle and the longitudinal angle respectively, the longitudinal lane changing reference data are matched with the target lane changing scene, the accuracy of determining the longitudinal lane changing reference data is improved, and therefore the accuracy of follow-up lane changing control is improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and more specifically, to a vehicle lane change control method and related devices. Background Art

[0002] With the continuous development of autonomous driving technology, intelligent navigation assistance functions have received more and more attention and use from consumers, and their performance in improving driving safety and reducing driving fatigue has been continuously improved.

[0003] Lane changing is not only a common operation in daily driving, but also a very important vehicle behavior and action in the intelligent navigation assistance function. In the intelligent navigation assistance function, how to realize the automatic vehicle lane changing operation is a technical problem that technicians in this field need to solve urgently. Summary of the invention

[0004] In view of this, the present application provides a vehicle lane change control method and related devices to solve the problem of the urgent need to realize automated vehicle lane change operations.

[0005] In order to solve the above technical problems, this application adopts the following technical solutions:

[0006] A vehicle lane change control method, comprising:

[0007] Based on the road information of the current road where the vehicle to be changed lanes is located, the current lane change scene is identified to obtain the target lane change scene;

[0008] Determining a target lane-changing space for the vehicle to change lanes;

[0009] Calculating longitudinal lane-changing motion data in the target lane-changing scenario;

[0010] Referring to the target lane change space, based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario, the vehicle to be lane changed is controlled to perform a lane change operation.

[0011] Optionally, based on the road information of the current road where the vehicle to be lane-changed is located, the current lane-changing scene is identified to obtain a target lane-changing scene, including:

[0012] Acquire road information of a current road where the vehicle to be changed lanes is located; the road information includes road environment information and target vehicle motion information; the target vehicle motion information includes motion information of the target vehicle; the target vehicles include a first type of target vehicle located in front of the vehicle to be changed lanes and a second type of target vehicle located in the lane to be changed lanes;

[0013] Determining whether there is room for lane change at the current moment according to the road information;

[0014] If it exists, the direct lane change scenario is used as the target lane change scenario.

[0015] Optionally, if there is no lane change space at the current moment, the method further includes:

[0016] Calculating an average vehicle density of the lane to be changed;

[0017] Counting the number of vehicles in the lane to be changed that are passed by the vehicle to be changed within a preset time period;

[0018] determining a lane changing scenario based on the average vehicle density and the number of vehicles, and using the determined lane changing scenario as a target lane changing scenario;

[0019] The lane changing scenarios include: a first dense traffic lane changing scenario, a second dense traffic lane changing scenario and a third dense traffic lane changing scenario; the traffic density in the first dense traffic lane changing scenario, the second dense traffic lane changing scenario and the third dense traffic lane changing scenario increases in sequence.

[0020] Optionally, if the target lane change scenario is a direct lane change scenario;

[0021] Calculating longitudinal lane-changing motion data in the target lane-changing scenario includes:

[0022] Calculating the risk factor of the vehicle to be changed lanes and the target vehicle;

[0023] Determining fusion data based on the risk coefficients of the vehicle to be changed and the target vehicle; the fusion data includes an initial value, a change rate, and a fusion correction coefficient;

[0024] According to the fusion data, motion information of a virtual target is calculated; the virtual target is obtained by fusion of at least one of the target vehicles;

[0025] The motion information of the virtual target is used to generate an acceleration control amount, and the acceleration control amount is used as the longitudinal lane change motion data in the direct lane change scenario.

[0026] Optionally, referring to the target lane change space and based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario, controlling the vehicle to be lane changed to perform a lane change operation includes:

[0027] Based on the acceleration control amount and the lateral lane change motion data in the direct lane change scenario, a virtual target following driving operation is performed to perform a lane change operation in the target lane change space.

[0028] Optionally, if the target lane change scenario is a first dense traffic lane change scenario, calculating the longitudinal lane change motion data in the target lane change scenario includes:

[0029] Performing trajectory planning operations on the vehicle to be lane-changed to obtain an acceleration and deceleration control curve;

[0030] Calculating an evaluation coefficient of the acceleration / deceleration control curve;

[0031] Calculating an overall evaluation coefficient of the acceleration / deceleration control curve according to the evaluation coefficient and a weighted value of the evaluation coefficient;

[0032] The acceleration / deceleration control curve corresponding to the minimum overall evaluation coefficient is used as the longitudinal lane changing motion data in the first dense traffic lane changing scenario.

[0033] Optionally, if the target lane change scenario is a second dense traffic lane change scenario, calculating the longitudinal lane change motion data in the target lane change scenario includes:

[0034] Using the tracking arrival space mode, the vehicle to be changed lane is accelerated and decelerated so that the speed of the vehicle to be changed lane is the same as the speed of the designated target vehicle in the lane to be changed lane;

[0035] performing a correction operation on the target lane change space to obtain a lane change space to be analyzed;

[0036] Performing a filtering operation on the lane change space to be analyzed to obtain lane change parameters; the lane change parameters include a reference lane change space and a speed of a target vehicle that is closest to the reference lane change space;

[0037] Calculating the acceleration of the vehicle to be changed lanes according to the lane change parameter;

[0038] The acceleration of the vehicle to be lane-changed is used as the longitudinal lane-changing motion data in the second dense traffic lane-changing scenario.

[0039] Optionally, if the target lane change scenario is a third dense traffic lane change scenario, calculating the longitudinal lane change motion data in the target lane change scenario includes:

[0040] If a new lane-changing space is detected during the process in which the vehicle to be changed lanes and the vehicle in the lane to be changed lanes are accompanied, the speed of the vehicle to be changed lanes is dynamically compensated using the new lane-changing space to obtain a target speed;

[0041] Calculating the acceleration of the vehicle to be changed lanes according to the target speed;

[0042] The acceleration of the vehicle to be lane-changed is used as the longitudinal lane-changing motion data in the third dense traffic lane-changing scenario.

[0043] A vehicle lane change control device, comprising:

[0044] A scene recognition module is used to identify the current lane change scene based on the road information of the current road where the vehicle to be changed lanes is located, and obtain a target lane change scene;

[0045] A space determination module, used to determine a target lane-changing space for the vehicle to change lanes;

[0046] A data calculation module, used to calculate longitudinal lane change motion data in the target lane change scenario;

[0047] The lane change control module is used to control the vehicle to be changed to perform a lane change operation with reference to the target lane change space and based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario.

[0048] An electronic device comprises at least one processor and a memory connected to the processor, wherein:

[0049] The memory is used to store computer programs;

[0050] The processor is used to execute the computer program so that the electronic device can implement the above-mentioned vehicle lane change control method.

[0051] The present application provides a vehicle lane change control method and related devices. When the vehicle changes lanes, the current lane change scene is identified based on the road information of the current road where the vehicle to be changed is located, and the target lane change scene is obtained, so that the subsequent lane change can match the actual road scene, thereby improving the accuracy and reliability of the lane change. In addition, in the present invention, the target lane change space of the vehicle to be changed is determined, the longitudinal lane change motion data under the target lane change scene is calculated, and the target lane change space is referred to, and based on the longitudinal lane change reference data and lateral lane change motion data under the target lane change scene, the vehicle to be changed is controlled to perform a lane change operation. The present invention determines the lane change reference data from two perspectives, lateral and longitudinal, respectively, and the longitudinal lane change reference data matches the target lane change scene, thereby improving the accuracy of the determination of the longitudinal lane change reference data, thereby improving the accuracy of the subsequent lane change control. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0053] Figure 1 A flow chart of a vehicle lane change control method provided by an embodiment of the present invention;

[0054] Figure 2A flowchart of a method for determining a scene provided by an embodiment of the present invention;

[0055] Figure 3 A flowchart of a method for determining motion data provided by an embodiment of the present invention;

[0056] Figure 4 A flowchart of another method for determining motion data provided by an embodiment of the present invention;

[0057] Figure 5 A flowchart of another method for determining motion data provided by an embodiment of the present invention;

[0058] Figure 6 A schematic diagram of the module structure of a multi-algorithm fusion spatial tracking anthropomorphic lane changing system provided by an embodiment of the present invention;

[0059] Figure 7 A flowchart of a multi-algorithm fusion spatial tracking anthropomorphic lane changing method provided by an embodiment of the present invention;

[0060] Figure 8 A schematic structural diagram of a vehicle lane change control device provided by an embodiment of the present invention;

[0061] Fig. 9 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] With the continuous development of autonomous driving technology, intelligent navigation assistance functions have received more and more attention and use from consumers, and their performance in improving driving safety and reducing driving fatigue has been continuously improved.

[0064] The typical feature of the intelligent navigation assistance function, which is different from L2 intelligent driving products, is that it can complete point-to-point navigation between toll stations on highways by changing lanes through autonomous decision-making. Therefore, lane changing is not only a common operation in daily driving, but also a very important vehicle behavior and action in the intelligent navigation assistance function. As a highly complex and highly comprehensive vehicle behavior process, its performance directly affects the safety, comfort and traffic efficiency of the intelligent navigation assistance function product.

[0065] The lane-changing process can be decomposed into two actions: one is to control the steering wheel to make the vehicle move laterally, and the other is to control the accelerator and brake to make the vehicle move longitudinally. The lane-changing performance, including collision safety, longitudinal comfort, lane-changing success rate, traffic efficiency, user-friendliness, etc., are all strongly related to longitudinal movement. Therefore, the longitudinal movement state is an important factor affecting the lane-changing performance.

[0066] At present, during the lane change process, the longitudinal movement process can be to maintain a constant speed during the lane change process of the ego vehicle, so the ego vehicle can only change lanes when it is in the changeable lane space of the target lane. In other scenarios, the ego vehicle drives normally in the lane and waits, and the lane change is not timely. In addition, the ego vehicle can accelerate during the lane change process, but the triggering strategy and working method are relatively simple. After the intention to change lanes, the ego vehicle accelerates to shorten the distance with the front vehicle to assist in lane change. The safety of this method and the necessity of barrier-free vehicles on the side and rear are not high.

[0067] To this end, the present invention can perform acceleration and deceleration actions based on the surrounding environment conditions including road conditions, vehicle conditions, and the vehicle's own motion state, actively reach the lane change space, achieve acceleration and deceleration lane changes, and significantly improve the success rate of lane changes.

[0068] Based on the above content, an embodiment of the present invention provides a vehicle lane change control method, which can be applied to vehicle equipment such as a controller.

[0069] Reference Figure 1 , a vehicle lane change control method may include:

[0070] S11. Based on the road information of the current road where the vehicle to change lanes is located, identify the current lane change scene to obtain a target lane change scene.

[0071] In this embodiment, the vehicle to be lane-changed refers to a vehicle that is about to change lanes. Before executing step S11 in the embodiment of the present invention, it is necessary to determine whether there is an intention to change lanes. The intention to change lanes can be realized by a lane-changing intention acquisition module, which includes a vehicle body state sensor and a software signal acquisition module, and is used to obtain automatic lane-changing request information including driver lighting information and an intelligent navigation assistance system. The lane-changing intention is a necessary condition for starting the vehicle acceleration and deceleration control system of the spatial tracking anthropomorphic lane change.

[0072] After determining the intention to change lanes, it is also necessary to determine whether the current road supports lane changes. For example, if the center line of the current road is a solid line, the lane line of the current road is not clearly identified, or the turning radius of the current road is too small, it is not suitable for lane changes. Even if there is an intention to change lanes, the lane change operation will not be performed. The lane change operation will only be performed if the current road supports lane changes.

[0073] When determining whether the current road supports lane change, it can be determined based on road environment information in the road information of the current road where the vehicle to change lanes is located.

[0074] The road information includes road environment information and target vehicle motion information.

[0075] Among them, the road environment information includes:

[0076] Lane line confidence signal, lane line length, lane line angle, lane line curvature, lane line curvature derivative, lane radius sequence, etc.

[0077] Subsequently, it can be determined based on the road environment information whether the static conditions of the current road environment meet the requirements for activating the vehicle acceleration and deceleration control system for agile automatic lane change.

[0078] In addition, the target vehicle motion information includes the motion information of the target vehicle; the target vehicle includes a first type of target vehicle located in front of the vehicle to be changed and a second type of target vehicle located in the lane to be changed. The motion information includes the distance, motion speed, and motion acceleration of the target vehicle.

[0079] Among them, the first type of target vehicle refers to the first target vehicle (the vehicle in front of the vehicle to be changed lanes), and the second type of target vehicles includes: the second target vehicle (the vehicle in front of the vehicle in the lane to be changed lanes, wherein the vehicle is the vehicle to be changed lanes), the third target vehicle (the vehicle in front of the vehicle in the lane to be changed lanes), the fourth target vehicle (the vehicle behind the vehicle in the lane to be changed lanes), and the fifth target vehicle (the vehicle behind the vehicle in the lane to be changed lanes).

[0080] In addition to determining whether the current road supports lane changing based on the road information of the current road, the current lane changing scene can be identified based on the road information of the current road to obtain a target lane changing scene.

[0081] The target lane change scenario in the present invention can be divided into a direct lane change scenario, a first dense traffic lane change scenario, a second dense traffic lane change scenario, and a third dense traffic lane change scenario according to the current traffic volume and other conditions. The traffic density in the first dense traffic lane change scenario, the second dense traffic lane change scenario, and the third dense traffic lane change scenario increases in sequence. In actual scenarios, the first dense traffic lane change scenario can be a scenario where sparse traffic cannot change lanes directly, the second dense traffic lane change scenario is a dense traffic cannot change lanes directly, and the third dense traffic lane change scenario is a target lane congestion scenario.

[0082] In the direct lane change scenario, the vehicle to be changed can change lanes directly. In the indirect lane change scenario, that is, in the first dense traffic lane change scenario, the second dense traffic lane change scenario, or the third dense traffic lane change scenario, direct lane change is not possible and lane change planning needs to be carried out according to the actual situation.

[0083] S12: Determine a target lane-changing space for the vehicle to change lanes.

[0084] Specifically, when it is determined based on the road environment information that the static conditions of the current road environment meet the requirements for starting the vehicle acceleration and deceleration control system for agile automatic lane change, the target lane change space can be determined in real time or periodically.

[0085] In the embodiment of the present invention, the optimal lane-changing space can be determined by searching the lane-changing space, which is the target position that the vehicle to be changed lanes needs to reach to complete the lane change.

[0086] The variable lane space search divides the road into five spatial areas according to the four target vehicles in the lane to be changed (the second target vehicle, the third target vehicle, the fourth target vehicle, and the fifth target vehicle mentioned above), calculates the longitudinal length of each spatial area and the predicted length after 5 seconds, and if the current length and the predicted length are both greater than 2 times the length of the vehicle to be changed, it is considered that the spatial area initially meets the usability. Then, according to the longitudinal distance between the usable spatial area and the vehicle to be changed, select one variable lane space in front of and behind the vehicle to be changed.

[0087] In the above two lane-changing spaces, one of the spatial regions is selected as the lane-changing space for the proposed lane change through scene decision and motion prediction calculation, specifically:

[0088] If the current lane is changing down a ramp and the distance between the vehicle to be changed and the ramp entrance is less than a certain time distance, the distance is related to the number of lane changes made by the vehicle to be changed. The more lane changes, the greater the distance. The rear space area is selected as the lane changing space for the proposed lane change; if there is a ramp entrance on the adjacent side of the current lane, the rear space area is selected as the lane changing space for the proposed lane change; if the current lane is about to end, the rear space area is selected as the lane changing space for the proposed lane change.

[0089] The relative motion relationship between the above-mentioned target vehicles and the vehicle to be changed lanes is predicted. If the vehicle to be changed lanes drives to the variable lane space in front and is too close to the first target vehicle, the space behind is selected as the lane changing space for the proposed lane change; if the vehicle to be changed lanes drives to the variable lane space in front or behind and is at a reasonable distance from the first target vehicle, a selection is made based on the absolute distances between the two candidate lane changing spaces, and the space with the smaller absolute distance is selected as the lane changing space for the proposed lane change.

[0090] In actual scenarios, if the traffic volume is light, there may be target lane change space. If the traffic volume is heavy, there may not be target lane change space. In this case, the lane change space search operation continues.

[0091] S13. Calculate longitudinal lane-changing motion data in the target lane-changing scenario.

[0092] In the embodiment of the present invention, different methods are used to calculate the longitudinal lane change motion data in different target lane change scenarios. The longitudinal lane change motion data generally refers to the longitudinal acceleration and deceleration.

[0093] S14. Referring to the target lane change space, based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario, control the vehicle to be lane changed to perform a lane change operation.

[0094] In this embodiment, in addition to determining the longitudinal lane change reference data in the target lane change scenario, it is also necessary to determine the lateral lane change motion data. The lateral lane change motion data is determined in the same manner in different scenarios.

[0095] The lateral lane change motion data is obtained by the vehicle automatic lane change planning path. Specifically, the vehicle automatic lane change planning path is performed in the geodetic coordinate system to obtain the vehicle automatic lane change path information, which is a set of coefficients C0-C5 representing a 5th-order polynomial. The specific calculation method is:

[0096] Y = C0 + C1*S + C2*S 2 + C3*S 3 + C4*S 4 + C5*S 5

[0097] Y' = C1 + 2*C2*S + 3*C3*S 2 + 4*C4*S 3 + 5*C5*S 4

[0098] Y'' = 2*C2 + 6*C3*S + 12*C4*S 2 + 20*C5*S 3

[0099] In the formula, Y is the lateral movement distance of the vehicle to be changed, S is the longitudinal movement distance of the vehicle to be changed, Y' is the lateral movement speed of the vehicle to be changed, and Y'' is the lateral movement acceleration of the vehicle to be changed. Through the vehicle automatic lane change path information, the lateral position of the vehicle to be changed at any longitudinal position can be obtained, and then the vehicle crossing time during the lane change process and the movement relationship with the target vehicle in the lane to be changed can be determined. Among them, the target vehicles in the lane to be changed are the second target vehicle, the third target vehicle, the fourth target vehicle, and the fifth target vehicle mentioned above.

[0100] Y, Y' and Y'' in the above content are the lateral lane change motion data in the embodiment of the present invention.

[0101] After the longitudinal lane change reference data and the lateral lane change motion data are known, in the direct lane change scenario and the first dense traffic lane change scenario, the vehicle to be changed can be controlled to change lanes in the target lane change space according to the two motion data. In the second dense traffic lane change scenario, the target lane change space may also be corrected to find a better lane change space.

[0102] In the third dense traffic lane change scenario, due to the heavy traffic, there may not be a target lane change space at the beginning. In this case, you can accompany the vehicle, wait for the right lane change opportunity, and then perform the lane change operation.

[0103] In this embodiment, when a vehicle changes lanes, the current lane change scene is identified based on the road information of the current road where the vehicle to be changed is located, and a target lane change scene is obtained, so that the subsequent lane change can match the actual road scene, thereby improving the accuracy and reliability of the lane change. In addition, in the present invention, the target lane change space of the vehicle to be changed is determined, the longitudinal lane change motion data under the target lane change scene is calculated, and the target lane change space is referred to, and based on the longitudinal lane change reference data and lateral lane change motion data under the target lane change scene, the vehicle to be changed is controlled to perform a lane change operation. The present invention determines the lane change reference data from both lateral and longitudinal perspectives, and the longitudinal lane change reference data matches the target lane change scene, thereby improving the accuracy of the longitudinal lane change reference data determination, thereby improving the accuracy of the subsequent lane change control.

[0104] The above embodiment mentions that it is necessary to identify the current lane change scene based on the road information of the current road where the vehicle to be changed is located, and obtain the target lane change scene. Figure 2 , which is specifically implemented as follows:

[0105] S21. Obtain road information of the current road where the vehicle to change lanes is located.

[0106] Among them, the road information includes road environment information and target vehicle motion information; the target vehicle motion information includes target vehicle motion information; the target vehicles include a first type of target vehicle located in front of the vehicle to be changed lanes and a second type of target vehicle located in the lane to be changed lanes.

[0107] When collecting road information, it can be obtained through roadside terminals, interaction with other vehicles, etc. The specific content of the road information refers to the corresponding description above.

[0108] S22. Determine whether there is lane change space at the current moment based on the road information; if so, execute step S23; if not, execute step S24.

[0109] Specifically, whether there is lane change space at the current moment can be determined by searching the lane change space as described above.

[0110] S23: Taking the direct lane change scenario as the target lane change scenario.

[0111] If the lane changing space is obtained through the variable lane space search, it means that the current vehicle is in the longitudinal space of the variable lane and the vehicle can change lanes directly, that is, the current road allows direct lane changing. The current scene is a direct lane changing scene. In this scene, the searched lane changing space is the target lane changing space.

[0112] S24, calculating the average vehicle density of the lane to be changed.

[0113] After determining that there is no lane change space at the current moment, it means that the vehicle to be changed is not in the longitudinal space of the lane to be changed, and the current traffic flow of the lane to be changed is not suitable for direct lane change, and the lane change space search operation is continued. At the same time, the average vehicle density of the lane to be changed is calculated.

[0114] Specifically, for the scenario where direct lane change is not possible, the average longitudinal distance Dist_Avg occupied by the target vehicle in the lane to be changed is analyzed and calculated using the following formula:

[0115] Dist_Avg = (Dist_x1 – Dist_x2) / (x2 – x1)

[0116] In the formula, x1 is the number of the front target in the lane to be changed, that is, the second target vehicle, x2 is the number of the rear target in the lane to be changed, that is, the fifth target vehicle, Dist_x1 and Dist_x2 are the longitudinal distances between the two vehicles and the lane to be changed vehicle. The average distance calculated above is processed for stability to obtain the average vehicle density of the lane to be changed.

[0117] S25. Count the number of vehicles in the lane to be changed that are passed by the vehicle to be changed within a preset time period.

[0118] Specifically, the number of vehicles passing the lane to be changed within a time period (which can be a preset time period, such as 10 seconds in this embodiment) is calculated, and the number is counted by time window counting and smoothness processing is performed to obtain the number of vehicles passed by the lane to be changed.

[0119] For example, within 10 seconds, the number of vehicles passing by the lane to be changed is 3.

[0120] S26. Determine a lane changing scenario based on the average vehicle density and the number of vehicles, and use the determined lane changing scenario as a target lane changing scenario.

[0121] Among them, the lane changing scenarios include: a first dense traffic lane changing scenario, a second dense traffic lane changing scenario and a third dense traffic lane changing scenario; the traffic density in the first dense traffic lane changing scenario, the second dense traffic lane changing scenario and the third dense traffic lane changing scenario increases successively.

[0122] In detail, a reasonable threshold can be set to divide the non-lane-changeable lanes into the first dense traffic flow (sparse traffic flow) lane change scenario, the second dense traffic flow (dense traffic flow) lane change scenario and the third dense traffic flow (congestion) lane change scenario according to the average vehicle density and the number of vehicles, thereby completing the identification and classification of the lane change scenarios.

[0123] For example, if the average vehicle density is less than A1 and the number of vehicles is less than B1, it is the first dense traffic flow (sparse traffic flow) lane changing scenario; if A1≤average vehicle density<A2 and B1≤number of vehicles<B2, it is the second dense traffic flow (dense traffic flow) lane changing scenario; if A2≤average vehicle density<A3 and B2≤number of vehicles<B3, it is the third dense traffic flow (congestion) lane changing scenario. A1-A3 and B1-B3 are numerical values, which can be configured according to the actual scenario.

[0124] In this embodiment, when performing scene recognition, two dimensions are taken into consideration: the average vehicle density of the lane to be changed and the number of vehicles passed by the vehicle to be changed, so that the traffic flow of the lane to be changed can be correctly analyzed, and the road scene can be accurately divided.

[0125] The above embodiments mention a variety of scenarios. Hereinafter, how to perform data calculation and lane change operations in each scenario will be explained.

[0126] 1. The target lane change scenario is a direct lane change scenario.

[0127] Specifically, refer to Figure 3 , calculating the longitudinal lane-changing motion data in the target lane-changing scenario may include:

[0128] S31, calculating the risk coefficients of the vehicle to change lanes and the target vehicle.

[0129] Specifically, for the direct lane change scenario, the vehicle in front of the own lane, the two vehicles in front of the target lane, and the two vehicles behind the target lane are considered at the same time, and their motion states relative to the own vehicle are converted into risk coefficients.

[0130] Specifically, the risk coefficients of each target vehicle and the vehicle to be changed in the current lane and the lane to be changed are calculated, such as the risk coefficient Dngr_Coe, and the calculation method is:

[0131] Dngr_Coe = K1*(Dist_Tgt – Dist_Dsrd)+ K2*(Spd_Tgt – Spd_Ego)

[0132] Where Dist_Tgt is the relative distance between the target vehicle and the vehicle to be changed, Dist_Dsrd is the appropriate following distance, Spd_Tgt is the speed of the target vehicle, Spd_Ego is the speed of the vehicle to be changed, and K1 and K2 are the weighted average risk coefficient control quantities.

[0133] For the five target vehicles mentioned above, it is necessary to calculate their risk coefficients with the vehicle to be changed lanes respectively to obtain five risk coefficients.

[0134] S32: Determine fusion data based on the risk coefficients of the vehicle to be lane-changed and the target vehicle.

[0135] In practical applications, since the first target vehicle, the third target vehicle, and the fourth target vehicle are relatively close to the vehicle to be changed, they have a greater impact on the vehicle to be changed when changing lanes. If the first target vehicle, the third target vehicle, and the fourth target vehicle are considered separately when changing lanes, it is difficult. Therefore, in this embodiment, the first target vehicle, the third target vehicle, and the fourth target vehicle can be fused into one virtual target through a fusion strategy of the risk coefficient.

[0136] During fusion, it is necessary to determine the fusion data, which includes the initial value, the rate of change, and the fusion correction coefficient. The initial value refers to the first value of the fusion coefficient, and the rate of change refers to the rate of change of the fusion coefficient in each calculation cycle. For example, in the zeroth calculation cycle, the initial value is 0.6 and the rate of change is 0.4. In the first calculation cycle, the fusion coefficient is 0.6+0.4=1, and in the second calculation cycle, the fusion coefficient is 1+0.4=1.4.

[0137] Specifically, by comparing the risk factors of the first target vehicle, the third target vehicle in the lane to be changed, and the fourth target vehicle, the initial value of the fusion coefficient (referred to as the initial value) and the change rate of the fusion coefficient (referred to as the change rate) of the virtual target are determined. Specifically: if the risk factor of the third target vehicle is greater than that of the first target vehicle, the initial value of the fusion coefficient is selected to be 0.6, and the change rate of the fusion coefficient is 0.4; if the risk factor of the third target vehicle is less than that of the first target vehicle, the initial value of the fusion coefficient is selected to be 0, and the change rate of the fusion coefficient is 0.2; if the fourth target vehicle affects the movement of the vehicle to be changed, the fusion coefficient is corrected on the above basis, and the fusion correction coefficient is 0.5.

[0138] S33. Calculate motion information of the virtual target according to the fused data.

[0139] In this embodiment, the virtual target is obtained by fusing at least one of the target vehicles. For example, in this embodiment, the virtual target is obtained by fusing the first target vehicle, the third target vehicle and the fourth target vehicle.

[0140] The calculation formula of the motion information of the virtual target is:

[0141] Virtual_Veh = (Target_Veh3*(FulCoeIni+FulCoeRate*T) + Target_Veh1*(1-(FulCoeIni+FulCoeRate*T)))*FulCorrectCoe + Target_Veh4*(1- FulCorrectCoe)

[0142] In the formula, Virtual_Veh is the motion information of the virtual target, Target_Veh3 is the motion information of the third target vehicle, FulCoeIni is the initial value of the fusion coefficient, FulCoeRate is the change rate of the fusion coefficient, T is the motion information fusion time, Target_Veh1 is the motion information of the first target vehicle, FulCorrectCoe is the fusion correction coefficient considering the fourth target vehicle, and Target_Veh4 is the motion information of the fourth target vehicle. Use the above formula to calculate the various motion information of the lane changing virtual target.

[0143] In each calculation cycle, Virtual_Veh is obtained according to the calculation formula, and the continuous motion information of the virtual target can be obtained by summing up the Virtual_Veh of multiple calculation cycles.

[0144] S34. Generate an acceleration control amount by using the motion information of the virtual target, and use the acceleration control amount as longitudinal lane change motion data in the direct lane change scenario.

[0145] Specifically, for the direct lane change scenario, when the vehicle to be changed lanes is changing lanes, the virtual target is used to call the vehicle following planning module to generate the acceleration control amount in the longitudinal direction. The acceleration control amount is the longitudinal lane change motion data in the direct lane change scenario.

[0146] Since the longitudinal lane change reference data and the lateral lane change motion data are known, the two are integrated to obtain information such as lane change acceleration and speed, and then a virtual target following driving operation is performed to perform a lane change operation in the target lane change space. That is, the present invention performs a virtual target following driving operation based on the acceleration control amount and the lateral lane change motion data in the direct lane change scenario to perform a lane change operation in the target lane change space, thereby realizing lane change anthropomorphism and module reuse.

[0147] 2. The target lane change scenario is the first dense traffic lane change scenario.

[0148] Reference Figure 4 , calculating the longitudinal lane-changing motion data in the target lane-changing scenario may include:

[0149] S41. Performing a trajectory planning operation on the vehicle to change lanes to obtain an acceleration / deceleration control curve.

[0150] Specifically, if the current scenario is the first dense traffic lane change scenario, that is, the sparse traffic cannot change lanes directly, the trajectory of the vehicle to be changed is planned using the trajectory sampling evaluation optimization method and the fifth-order polynomial. The acceleration and deceleration control curve of the vehicle is characterized by a fifth-order polynomial, and the fifth-order polynomial coefficients are A0-A5. The specific calculation method is as follows:

[0151] S = A0 + A1*t + A2*t 2 + A3*t 3 + A4*t 4 + A5*t 5

[0152] S' = A1 + 2*A2*t + 3*A3*t 2 + 4*A4*t 3 + 5*A5*t 4

[0153] S'' = 2*A2 + 6*A3*t + 12*A4*t 2 + 20*A5*t 3

[0154] Where S is the longitudinal movement distance of the vehicle to be changed, t is the movement time of the vehicle to be changed, S' is the movement speed of the vehicle to be changed, and S'' is the movement acceleration of the vehicle to be changed.

[0155] By sampling different arrival space distances, movement times, and arrival speeds, different acceleration and deceleration control curves for the vehicle can be obtained. By obtaining as many acceleration and deceleration control curves as possible, as many vehicle speed trajectory curves as possible can be obtained, thereby increasing the selectable space for the vehicle's motion state. The number of trajectory curves is comprehensively evaluated based on the computing power and performance requirements of the controller. For example, in the embodiment of the present invention, 9 acceleration and deceleration control curves are selected.

[0156] S42: Calculate the evaluation coefficient of the acceleration / deceleration control curve.

[0157] Specifically, when calculating the evaluation coefficient of each acceleration and deceleration curve, the selected evaluation coefficients include the front vehicle collision safety, speed matching consistency, lane target motion relationship consistency, motion feasibility, motion comfort, traffic regulations compliance, etc.

[0158] During the specific calculation, the acceleration / deceleration control curves are evaluated in combination with the vehicle automatic lane change path information. First, each of the acceleration / deceleration control curves is discretized, and then the evaluation coefficients are calculated. After each evaluation coefficient is calculated, it is normalized.

[0159] S43. Calculate the overall evaluation coefficient of the acceleration / deceleration control curve according to the evaluation coefficient and the weighted value of the evaluation coefficient.

[0160] For an acceleration / deceleration control curve, there are multiple evaluation coefficients corresponding to it. All the evaluation coefficients need to be weighted to obtain the final overall evaluation coefficient.

[0161] Specifically, the weighted values ​​of different evaluation coefficients may be the same, that is, the above evaluation coefficients are weighted averaged to obtain the overall evaluation coefficient of a single acceleration / deceleration control curve.

[0162] In addition, different weighted values ​​can be set for different evaluation coefficients according to actual scenarios. For example, an adaptive evaluation system for evaluation coefficients can be designed to improve the success rate and robustness of the algorithm. Specifically, the influencing factors of different lane change scenarios are dynamically processed to distinguish the main influencing factors in different scenarios. At the same time, weighted value calculation strategies of evaluation coefficients with different focuses are designed according to different lane change styles. For example, if the current lane change request is a mandatory intention in the off-ramp scenario, the weight of sports comfort is reduced, and the restrictions on acceleration and deceleration are relaxed to facilitate the solution; if the current lane change scenario requires the driver to confirm the lane change, the weight of the collision safety with the preceding vehicle and the preceding vehicle in the lane to be changed is increased to improve the conservatism of the system; if the current lane change is a non-mandatory lane change recommendation scenario, the weight of sports comfort is increased to improve the driving quality.

[0163] After the weighted values ​​of the evaluation coefficients are determined, the weighted sum of the evaluation coefficients is performed to obtain the overall evaluation coefficient of each acceleration / deceleration control curve.

[0164] S44. Using the acceleration / deceleration control curve corresponding to the minimum overall evaluation coefficient as the longitudinal lane-changing motion data in the first dense traffic lane-changing scenario.

[0165] Specifically, by comparing the sizes of the overall evaluation coefficients, the acceleration / deceleration control curve with the smallest overall evaluation coefficient is screened out. This curve is the optimal curve for vehicle acceleration / deceleration, and this curve is used as the longitudinal lane change motion data in the first dense traffic lane change scenario.

[0166] It should be noted that after screening out the acceleration / deceleration control curve with the smallest overall evaluation coefficient, although the overall evaluation coefficient of this curve is the smallest, it may still be greater than the maximum limit of the overall evaluation coefficient. In this case, this curve is still not the optimal curve. At this time, a backup trajectory strategy can be designed as an unsolvable auxiliary. Specifically, when performing trajectory sampling, adjust the constraints of the arrival space, among which the longitudinal distance of the arrival space is adjusted first to make the sampling margin wider; at the same time, ensure that there is enough speed difference between the vehicle to be changed and the related vehicles in the side lane, and keep it at about 5kph to create an opportunity for lane change.

[0167] After re-sampling the trajectory, the acceleration / deceleration control curve is reconstructed, and then the evaluation coefficient is calculated, and the above steps are repeated until an acceleration / deceleration control curve with a minimum overall evaluation coefficient that is less than the maximum limit of the overall evaluation coefficient is selected.

[0168] For scenarios where sparse traffic cannot change lanes directly, an adaptively designed trajectory sampling evaluation optimization method and a planning unsolvable backup trajectory algorithm are used. The final acceleration and deceleration control curve selected is the acceleration-time control curve for a period of time in the future. The value is taken according to the time from the start of planning to the present, and the acceleration control amount at the current moment is obtained. The acceleration is used to track and change lanes in the variable lane space, thereby improving the algorithm's solution success rate and robustness.

[0169] 3. The target lane change scenario is the second dense traffic lane change scenario.

[0170] Reference Figure 5 , calculating the longitudinal lane-changing motion data in the target lane-changing scenario, including:

[0171] S51. Using the tracking arrival space mode, the vehicle to be lane-changed is accelerated or decelerated so that the speed of the vehicle to be lane-changed is the same as the speed of the designated target vehicle in the lane to be lane-changed.

[0172] In this embodiment, if the current scene is dense traffic and direct lane change is not possible, when the speed difference between the vehicle to be changed and the designated target vehicle in the target lane (such as the third target vehicle or the fourth target vehicle that has a greater impact on the lane change of the vehicle) is large, it is difficult to obtain an effective solution using the planning method. In the case of no solution or easy no solution, the tracking arrival space mode based on speed and distance double-layer dynamic proportional control is adopted to control the acceleration and deceleration of the vehicle to be changed, comprehensively consider the position of the reachable point and the speed of the traffic flow to calculate the acceleration, and reasonably constrain the change in the speed and acceleration of the vehicle to ensure acceleration and deceleration in a comfortable experience. The specific steps are:

[0173] In the tracking arrival space mode, the large speed difference between the vehicle to be changed and the designated target vehicle in the target lane is considered. At this time, the speed is tracked first, that is, the weight of the distance item is reduced when the speed difference is large, so that the speed is the main control target. When the speed of the ego vehicle is close to that of the designated target vehicle, the weight of the distance item is gradually increased, and finally the ego vehicle is in the target lane change space and has the same speed as the designated target vehicle.

[0174] S52: performing a correction operation on the target lane change space to obtain a lane change space to be analyzed.

[0175] Since the above target lane change space is only the variable lane space calculated at a certain moment, for the scene where lane change can be made directly and the scene where lane change cannot be made directly in sparse traffic, due to the small traffic volume, the target lane change space can be used directly for lane change operation. However, in the scene where lane change cannot be made directly in dense traffic, due to the large traffic volume, it is necessary to meet the demand in the planning frame. In the tracking arrival space mode, the above target lane change space needs to be corrected to find a better lane change space to be analyzed.

[0176] First, the target lane-changing space must be maintained when the vehicle to be changed lanes changes lanes. At this point, lane-changing preparations have already begun. Regardless of whether there is a theoretically better lane-changing space, the target lane-changing space should be limited to the lane-changing space directly to the side of the vehicle. Considering that the speed of the vehicle to be changed lanes and the third and fourth target vehicles is the same, it is safest to ensure safety. Therefore, the position of the lane-changing space is selected according to the following formula:

[0177] Dist_LaneChgSpce=Dist_Veh3−(Dist_Veh4 * Spd_Ego

[0178] −Dist_Veh4*Spd_Veh3) / ( Spd_Ego − Spd_Veh4)

[0179] Among them, Dist_LaneChgSpce is the distance between the lane change space to be analyzed and the vehicle to be changed, Spd_Ego is the speed of the vehicle to be changed, Dist_Veh3 is the relative distance between the third target vehicle and the vehicle to be changed, Spd_Veh3 is the speed of the third target vehicle, Dist_Veh4 is the relative distance between the fourth target vehicle and the vehicle to be changed, and Spd_Veh4 is the speed of the fourth target vehicle.

[0180] It should be noted that only considering TTC (Time To Collision) cannot guarantee the safety of absolute distance, and when there is only the third or fourth target vehicle, the calculated target point will be very far away. It is also necessary to limit the minimum distance between the variable lane space and the third and fourth target vehicles and its maximum value.

[0181] S53: Perform a filtering operation on the lane change space to be analyzed to obtain a lane change parameter.

[0182] The lane change parameters include a reference lane change space and a speed of a target vehicle that is closest to the reference lane change space.

[0183] Specifically, in actual scenarios, there are situations where the lane change space repeatedly jumps between the previous and next lane change spaces due to environmental changes. To solve this problem, a time-based dynamic first-order low-pass filter is applied to the lane change space position to be analyzed. When the lane change point position changes continuously in a small range within the same lane change space, a smaller filter coefficient is used to make it smoother. When the lane change space position to be analyzed jumps from one lane change space to another, a period of confirmation is performed first. After confirming that the lane change space to be analyzed has indeed changed to another lane change space, the filter coefficient is released for a period of time, so that the lane change space position to be analyzed quickly and smoothly transitions to the new lane change space, and then the smaller filter coefficient is restored again, so that the lane change space to be analyzed changes smoothly in the new lane change space.

[0184] The above strategy can obtain a stable reference lane change space, the speed of the target vehicle closest to the reference lane change space, a dynamic control coefficient of speed (specifically, a speed control coefficient) and a dynamic control coefficient of distance (specifically, a distance control coefficient).

[0185] S54. Calculate the acceleration of the vehicle to change lanes according to the lane change parameter.

[0186] Specifically, for scenarios where direct lane change is not possible in dense traffic, the acceleration of the vehicle to be changed is calculated using the calculated reference lane change space and the speed of the target vehicle closest to the space. The specific formula is:

[0187] Accel_Target = (Spd_Veh – Spd_Ego)* Spd_Coe + Dist_LaneChgSpce *SpceDist_Coe

[0188] Where Accel_Target is the acceleration of the vehicle to be changed lanes, Spd_Veh is the speed of the target vehicle closest to the reference lane changing space, Spd_Ego is the speed of the vehicle to be changed lanes, Spd_Coe is the speed control coefficient, Dist_LaneChgSpce is the distance between the vehicle to be changed lanes and the reference lane changing space, SpceDist_Coe is the distance control coefficient of the reference lane changing space, and Set_Spd is the speed set by the driver.

[0189] S55: Using the acceleration of the vehicle to be lane-changed as longitudinal lane-changing motion data in the second dense traffic lane-changing scenario.

[0190] Specifically, the acceleration calculated in the above steps, i.e., the longitudinal lane-changing motion data in the second dense traffic lane-changing scenario, is used to perform a lane-changing operation in the reference lane-changing space.

[0191] In this embodiment, for scenarios where direct lane change is not possible due to dense traffic, a vehicle motion planning control algorithm with strong stability is designed to accurately determine the timing of lane change. By interacting with the lane change command of the lateral function, the vehicle is ensured to reach the lane change space through acceleration and deceleration and achieve sensitive lane change.

[0192] 4. The target lane change scenario is the third dense traffic lane change scenario.

[0193] Calculating longitudinal lane-changing motion data in the target lane-changing scenario includes:

[0194] 1) If a new lane-changing space is detected during the process in which the vehicle to be changed lanes travels with the vehicle in the lane to be changed lanes, the speed of the vehicle to be changed lanes is dynamically compensated using the new lane-changing space to obtain a target speed.

[0195] Specifically, if the current target lane is congested, it is difficult to find a variable lane space that meets the conditions within the perception range. At this time, the variable lane space given above is often in front of the sensing lead vehicle (before the second target vehicle) or behind the following vehicle (behind the fifth target vehicle), that is, a non-real variable lane space outside the perception range. At this time, blind tracking is not conducive to the lane change of the vehicle to be changed. Therefore, under congested conditions, the vehicle to be changed lanes chooses to change lanes by following the traffic flow, that is, the vehicle to be changed lanes follows the speed of the traffic flow in the lane to be changed. During the accompanying process, the speed of the vehicle to be changed lanes is mainly considered to design the speed of the vehicle to be changed lanes, to assist in completing the lane change. Specifically:

[0196] During the accompanying process, in order to avoid the vehicle to be changed lanes being in the same difficult lane-changing position for a long time and to be more actively looking for lane-changing space, the vehicle to be changed lanes will be accompanied at a speed slightly higher or lower than the speed of the traffic in the lane to be changed. That is, in this embodiment, for the target lane congestion scenario, since there is no lane-changing space at this time, the algorithm of following the speed of the traffic flow is designed to control the vehicle to move in the direction of the speed that increases the probability of successful lane change.

[0197] Furthermore, if the current vehicle to be changed lanes is close to the ramp merging point, the vehicle to be changed lanes is close to the target ramp entrance, and the vehicle to be changed lanes is close to the road end point, it will be accompanied at a speed slightly lower than the traffic speed of the lane to be changed, and the rest will be accompanied at a speed slightly higher than the traffic speed of the lane to be changed.

[0198] During the accompanying process, when a new variable lane space is detected near the vehicle to be changed, and it is predicted that this space can be stably maintained for a period of time based on the motion relationship between the vehicle to be changed and related vehicles, in order to increase the lane change success rate of the vehicle to be changed, the speed of the vehicle to be changed will be dynamically compensated according to the distance of the new variable lane space to obtain the target speed, so that the vehicle to be changed can quickly approach the variable lane space. The specific formula is:

[0199] Spd_Target = MIN(Spd_Flow * Follow_Coe + Dist_LaneChgSpce * SpceDist_Coe, Set_Spd)

[0200] Wherein, Spd_Target is the target speed of the vehicle to be changed lanes, Spd_Flow is the congested traffic speed of the target vehicle in the lane to be changed lanes, Follow_Coe is the correction coefficient for the speed of the following traffic flow, so that the vehicle to be changed lanes can travel at a speed slightly lower or slightly higher than the traffic flow speed, Dist_LaneChgSpce is the distance between the vehicle to be changed lanes and the new lane changing space, and SpceDist_Coe is the correction coefficient of the new lane changing space to the target speed. If it is in front of the vehicle to be changed lanes, it is a positive value, and if it is behind it, it is a negative value, so that the vehicle to be changed lanes can quickly approach the variable lane space.

[0201] 2) calculating the acceleration of the vehicle to be lane-changed according to the target speed, and using the acceleration of the vehicle to be lane-changed as the longitudinal lane-changing motion data in the third dense traffic lane-changing scenario.

[0202] Specifically, for the target lane congestion scenario, the calculated target speed of the vehicle to be changed is used to calculate the acceleration of the vehicle to be changed. The specific formula is:

[0203] Accel_Target = (Spd_Target – Spd_Ego)* Spd_Coe

[0204] Where Accel_Target is the acceleration of the vehicle to change lanes, Spd_Target is the target speed of the vehicle to change lanes in a congested scenario, Spd_Ego is the speed of the vehicle to change lanes, and Spd_Coe is the speed control coefficient.

[0205] After the acceleration is obtained, the acceleration can be used as the longitudinal lane-changing motion data in the third dense traffic lane-changing scenario. The longitudinal lane-changing motion data and the lateral lane-changing motion data are used to perform a lane-changing operation in the new lane-changing space.

[0206] When changing lanes, the acceleration of the vehicle to be changed is controlled by requesting the actuator to perform corresponding actions, so that the vehicle can accurately accelerate and decelerate to track the variable lane space, and assist in completing the anthropomorphic lane change action.

[0207] In this embodiment, by making decisions and dividing the automatic lane change scenarios of the intelligent driving assistance system, designing the variable lane space tracking algorithm and arbitration use strategy in different scenarios, the present invention significantly improves the anthropomorphism, safety and success rate of automatic lane change. The present invention has the following advantages:

[0208] First of all, the present invention organically integrates the advantages of various algorithms through scene recognition decision-making and multi-algorithm fusion application. Compared with the lane changing control method of switching the following vehicle target, it has good adaptability and response effects in different scenarios, and the lane changing process is more humanized.

[0209] Secondly, the present invention comprehensively considers various lane-changing scenarios; the vehicle's movements take into account both agility and comfort; it has full-scenario coverage and agile and comfortable lane-changing performance, achieving the best hardware configuration performance of the same level of products on the market.

[0210] Finally, the present invention designs scenario-specific usage strategies for different algorithms. Compared with strategies that rely too much on numerical optimization algorithms and learning-based strategies, it significantly reduces the software's performance requirements for hardware, reduces product costs, and is suitable for large-scale applications.

[0211] Another embodiment of the present invention provides a multi-algorithm fusion spatial tracking anthropomorphic lane change method and system module structure diagram. Figure 6 A multi-algorithm fusion space tracking anthropomorphic lane change control system of the present invention includes a lane change intention acquisition module 100, a road environment information acquisition module 200, a target vehicle motion information acquisition module 300, a road scene recognition decision and variable lane space search module 400, a direct lane change scene processing module 500, a sparse traffic non-direct lane change scene processing module 600, a dense traffic non-direct lane change scene processing module 700, a target lane congestion scene processing module 800, a vehicle acceleration generation and scene arbitration output module 900, and a vehicle control execution module 1000.

[0212] Among them, the lane change intention acquisition module 100 includes a vehicle body state sensor and a software signal acquisition module, which is used to obtain information including the driver's lighting information and the automatic lane change request information of the intelligent navigation assistance system; the road environment information acquisition module 200 and the target vehicle motion information acquisition module 300 can be a front-view camera and a forward radar as well as a corner radar or a surround camera, which are used to obtain lane line parameter information, including lane line confidence, lane line length, lane line angle, lane line curvature, lane line curvature derivative, and to obtain the motion information of the target vehicles around the vehicle, including the distance, motion speed, and motion acceleration of the target vehicle; road scene recognition decision and variable lane space The search module 400 identifies, decides and classifies the current lane change scenario, and searches for variable lane space; the four types of lane change scenario processing modules include a direct lane change scenario processing module 500, a sparse traffic non-direct lane change scenario processing module 600, a dense traffic non-direct lane change scenario processing module 700, and a target lane congestion scenario processing module 800, which respectively obtain the information required for lane change control acceleration for different lane change scenarios; the vehicle acceleration generation and scenario arbitration output module 900 is used to generate control acceleration and arbitrate and output acceleration for lane change scenarios; the vehicle control execution module 1000 is composed of a vehicle power system and a braking system, and is used to execute acceleration and deceleration control information.

[0213] In the above-mentioned multi-algorithm fusion spatial tracking anthropomorphic lane change control system, the lane change intention acquisition module 100, the road environment information acquisition module 200, and the target vehicle motion information acquisition module 300 are connected to the road scene recognition decision and variable lane space search module 400, providing the vehicle's lane change intention, road lane line information, stationary target information, and first to fifth moving target information for identifying and deciding the current lane change scenario, which is divided into four scenarios: direct lane change, sparse traffic flow cannot directly change lanes, dense traffic flow cannot directly change lanes, and target lane congestion.

[0214] The road scene recognition decision and variable lane space search module 400 is connected to four types of lane change scene processing modules, including a direct lane change scene processing module 500, a sparse traffic flow non-direct lane change scene processing module 600, a dense traffic flow non-direct lane change scene processing module 700, and a target lane congestion scene processing module 800. These four modules process information under different lane change scenes and generate information that can be used to calculate acceleration. The direct lane change scene processing module 500 generates a virtual target for following the vehicle when changing lanes, the sparse traffic flow non-direct lane change scene processing module 600 generates the optimal acceleration and deceleration control curve for the vehicle to track the lane change space during the lane change process, the dense traffic flow non-direct lane change scene processing module 700 generates the distance of the lane change space tracking and the speed of the relevant vehicle, and the target lane congestion scene processing module 800 generates the target speed of the traffic that can follow the lane to be changed.

[0215] Four types of lane change scenario processing modules, namely, direct lane change scenario processing module 500, sparse traffic non-direct lane change scenario processing module 600, dense traffic non-direct lane change scenario processing module 700, and target lane congestion scenario processing module 800 are connected to vehicle acceleration generation and scenario arbitration output module 900, and are used to calculate the vehicle control acceleration according to the information generated by the four types of lane change scenario processing modules; vehicle acceleration generation and scenario arbitration output module 900 is connected to vehicle control execution module 1000, which sends a control signal to coordinate the power system and the braking system to work inside the vehicle control execution module 1000, so as to achieve the control goal of tracking the variable lane space to complete the auxiliary lane change.

[0216] Figure 7 This is a flowchart of a multi-algorithm fusion spatial tracking anthropomorphic lane changing method and system provided by the implementation case of the present invention. Figure 7 A multi-algorithm fusion spatial tracking anthropomorphic lane change control system of the present invention comprises the following steps:

[0217] When the vehicle's intention to automatically change lanes is obtained, the vehicle's road environment information and surrounding target vehicle motion information are simultaneously obtained;

[0218] After obtaining the above information, the current lane change scenario is identified and decided according to the road environment information and divided into four categories, and the variable lane space is searched; according to the classification, the scenarios where direct lane change is allowed, the scenarios where direct lane change is not allowed in sparse traffic, the scenarios where direct lane change is not allowed in dense traffic, and the scenarios where the target lane is congested are processed separately, and the corresponding relevant information is generated respectively.

[0219] The vehicle control acceleration is calculated based on the relevant information generated in each scenario, and the acceleration request value is arbitrated and output according to the scenario classification. The acceleration request value is used for vehicle execution control.

[0220] The acceleration / deceleration request value is sent to the execution module for response, thereby realizing the spatial tracking anthropomorphic lane change of the intelligent navigation assistance function.

[0221] In this embodiment, according to the decision and judgment results of the current scene, the planning control information of different algorithm modules is selected to generate the acceleration control amount required for the vehicle to reach the variable lane space, and the control curve is used for vehicle control. The actuator action is used to achieve the acceleration and deceleration of the vehicle to reach the variable lane space. During the lane change process, the lane change space can be tracked through the fusion of multiple algorithms, so that the vehicle can reasonably and effectively accelerate and decelerate to perform anthropomorphic lane change.

[0222] Based on the above embodiment, another embodiment of the present invention provides a vehicle lane change control device, referring to Figure 8 , which may include:

[0223] A scene recognition module 11 is used to recognize the current lane change scene based on the road information of the current road where the vehicle to be changed lanes is located, and obtain a target lane change scene;

[0224] A space determination module 12 is used to determine a target lane-changing space for the vehicle to change lanes;

[0225] A data calculation module 13, used to calculate longitudinal lane change motion data in the target lane change scenario;

[0226] The lane change control module 14 is used to control the vehicle to be changed to perform a lane change operation with reference to the target lane change space and based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario.

[0227] In one implementation, the scene recognition module 11 includes:

[0228] An information acquisition submodule is used to acquire road information of a current road where the vehicle to be changed lanes is located; the road information includes road environment information and target vehicle motion information; the target vehicle motion information includes target vehicle motion information; the target vehicles include a first type of target vehicle located in front of the vehicle to be changed lanes and a second type of target vehicle located in the lane to be changed lanes;

[0229] A space judgment submodule, used to judge whether there is lane change space at the current moment according to the road information;

[0230] The first scene determination submodule is used to use the direct lane change scene as the target lane change scene if it exists.

[0231] In one implementation, the scene recognition module 11 further includes:

[0232] A density calculation submodule, used for calculating the average vehicle density of the lane to be changed if there is no lane change space at the current moment;

[0233] A quantity counting submodule, used for counting the number of vehicles in the lane to be changed that are passed by the vehicle to be changed within a preset time period;

[0234] a second scene determination submodule, configured to determine a lane change scene based on the average vehicle density and the number of vehicles, and use the determined lane change scene as a target lane change scene;

[0235] The lane changing scenarios include: a first dense traffic lane changing scenario, a second dense traffic lane changing scenario and a third dense traffic lane changing scenario; the traffic density in the first dense traffic lane changing scenario, the second dense traffic lane changing scenario and the third dense traffic lane changing scenario increases in sequence.

[0236] In one implementation, if the target lane change scenario is a direct lane change scenario, the data calculation module 13 includes:

[0237] A risk factor calculation submodule, used to calculate the risk factor of the vehicle to be changed lanes and the target vehicle;

[0238] A fusion data determination submodule, used to determine fusion data based on the risk coefficients of the vehicle to be changed and the target vehicle; the fusion data includes an initial value, a change rate, and a fusion correction coefficient;

[0239] A motion information calculation submodule, used to calculate the motion information of a virtual target according to the fusion data; the virtual target is obtained by fusion of at least one of the target vehicles;

[0240] The first motion data determination submodule is used to generate an acceleration control amount by using the motion information of the virtual target, and use the acceleration control amount as the longitudinal lane change motion data in the direct lane change scenario.

[0241] In one implementation, the lane change control module 14 is specifically configured to:

[0242] Based on the acceleration control amount and the lateral lane change motion data in the direct lane change scenario, a virtual target following driving operation is performed to perform a lane change operation in the target lane change space.

[0243] In one implementation, if the target lane change scenario is a first dense traffic lane change scenario, the data calculation module 13 includes:

[0244] A trajectory planning submodule, used to perform trajectory planning operations on the vehicle to be changed lanes to obtain an acceleration and deceleration control curve;

[0245] A first evaluation coefficient calculation submodule, used to calculate the evaluation coefficient of the acceleration / deceleration control curve;

[0246] A second evaluation coefficient calculation submodule, used to calculate the overall evaluation coefficient of the acceleration / deceleration control curve according to the evaluation coefficient and the weighted value of the evaluation coefficient;

[0247] The second motion data determination submodule is used to use the acceleration / deceleration control curve corresponding to the minimum overall evaluation coefficient as the longitudinal lane change motion data in the first dense traffic lane change scenario.

[0248] In one implementation, if the target lane change scenario is the second dense traffic lane change scenario, the data calculation module 13 includes:

[0249] A vehicle speed control submodule, used to use the tracking arrival space mode to perform acceleration and deceleration control on the vehicle to be changed lanes, so that the speed of the vehicle to be changed lanes is the same as the speed of the designated target vehicle in the lane to be changed lanes;

[0250] a space correction submodule, configured to perform a correction operation on the target lane change space to obtain a lane change space to be analyzed;

[0251] A filtering submodule, configured to perform a filtering operation on the lane change space to be analyzed to obtain lane change parameters; the lane change parameters include a reference lane change space and a speed of a target vehicle that is closest to the reference lane change space;

[0252] A first acceleration calculation submodule, used for calculating the acceleration of the vehicle to change lanes according to the lane change parameter;

[0253] The third motion data determination submodule is used to use the acceleration of the vehicle to change lanes as the longitudinal lane change motion data in the second dense traffic lane change scenario.

[0254] In one implementation, if the target lane change scenario is the third dense traffic lane change scenario, the data calculation module 13 includes:

[0255] A dynamic compensation submodule, configured to dynamically compensate the speed of the vehicle to be changed lanes using the new lane change space to obtain a target speed if a new lane change space is detected during the accompanying movement of the vehicle to be changed lanes and the vehicle in the lane to be changed lanes;

[0256] A second acceleration calculation submodule, used to calculate the acceleration of the vehicle to change lanes according to the target speed;

[0257] The fourth motion data determination submodule is used to use the acceleration of the vehicle to change lanes as the longitudinal lane change motion data in the third dense traffic lane change scenario.

[0258] In this embodiment, when a vehicle changes lanes, the current lane change scene is identified based on the road information of the current road where the vehicle to be changed is located, and a target lane change scene is obtained, so that the subsequent lane change can match the actual road scene, thereby improving the accuracy and reliability of the lane change. In addition, in the present invention, the target lane change space of the vehicle to be changed is determined, the longitudinal lane change motion data under the target lane change scene is calculated, and the target lane change space is referred to, and based on the longitudinal lane change reference data and lateral lane change motion data under the target lane change scene, the vehicle to be changed is controlled to perform a lane change operation. The present invention determines the lane change reference data from both lateral and longitudinal perspectives, and the longitudinal lane change reference data matches the target lane change scene, thereby improving the accuracy of the longitudinal lane change reference data determination, thereby improving the accuracy of the subsequent lane change control.

[0259] It should be noted that, for the working process of the modules and sub-modules in the embodiment of the present invention, please refer to the corresponding description in the above embodiment.

[0260] An embodiment of the present application also provides an electronic device, including at least one processor and a memory connected to the processor, wherein:

[0261] The memory is used to store computer programs;

[0262] The processor is used to execute the computer program so that the electronic device can implement the above-mentioned vehicle lane change control method.

[0263] The present application also provides an electronic device in an embodiment. Fig. 9 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiment of the present application. The electronic device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Fig. 9 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0264] like Fig. 9 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0265] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Fig. 9 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0266] An embodiment of the present application also provides a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any vehicle lane change control method provided in the embodiment of the present application.

[0267] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any vehicle lane change control method provided in the embodiment of the present application.

[0268] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle lane change control method, characterized in that: include: Based on the road information of the current road where the vehicle to be changed lanes is located, the current lane change scene is identified to obtain the target lane change scene; Determining a target lane-changing space for the vehicle to change lanes; Calculating longitudinal lane-changing motion data in the target lane-changing scenario; Referring to the target lane change space, based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario, the vehicle to be lane changed is controlled to perform a lane change operation.

2. The vehicle lane change control method according to claim 1, characterized in that: Based on the road information of the current road where the vehicle to be changed lanes is located, the current lane change scene is identified to obtain the target lane change scene, including: Acquire road information of a current road where the vehicle to be changed lanes is located; the road information includes road environment information and target vehicle motion information; the target vehicle motion information includes motion information of the target vehicle; the target vehicles include a first type of target vehicle located in front of the vehicle to be changed lanes and a second type of target vehicle located in the lane to be changed lanes; Determining whether there is room for lane change at the current moment according to the road information; If it exists, the direct lane change scenario is used as the target lane change scenario.

3. The vehicle lane change control method according to claim 2, characterized in that: If there is no lane change space at the current moment, it also includes: Calculating an average vehicle density of the lane to be changed; Counting the number of vehicles in the lane to be changed that are passed by the vehicle to be changed within a preset time period; determining a lane changing scenario based on the average vehicle density and the number of vehicles, and using the determined lane changing scenario as a target lane changing scenario; The lane changing scenarios include: a first dense traffic lane changing scenario, a second dense traffic lane changing scenario and a third dense traffic lane changing scenario; the traffic density in the first dense traffic lane changing scenario, the second dense traffic lane changing scenario and the third dense traffic lane changing scenario increases in sequence.

4. The vehicle lane change control method according to claim 2, characterized in that: If the target lane change scenario is a direct lane change scenario; Calculating longitudinal lane-changing motion data in the target lane-changing scenario includes: Calculating the risk factor of the vehicle to be changed lanes and the target vehicle; Determining fusion data based on the risk coefficients of the vehicle to be changed and the target vehicle; the fusion data includes an initial value, a change rate, and a fusion correction coefficient; According to the fusion data, motion information of a virtual target is calculated; the virtual target is obtained by fusion of at least one of the target vehicles; The motion information of the virtual target is used to generate an acceleration control amount, and the acceleration control amount is used as the longitudinal lane change motion data in the direct lane change scenario.

5. The vehicle lane change control method according to claim 4, characterized in that: Referring to the target lane change space, based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario, controlling the vehicle to be lane changed to perform a lane change operation, including: Based on the acceleration control amount and the lateral lane change motion data in the direct lane change scenario, a virtual target following driving operation is performed to perform a lane change operation in the target lane change space.

6. The vehicle lane change control method according to claim 2, characterized in that: If the target lane change scenario is a first dense traffic lane change scenario, calculating longitudinal lane change motion data under the target lane change scenario includes: Performing trajectory planning operations on the vehicle to be lane-changed to obtain an acceleration and deceleration control curve; Calculating an evaluation coefficient of the acceleration / deceleration control curve; Calculating an overall evaluation coefficient of the acceleration / deceleration control curve according to the evaluation coefficient and a weighted value of the evaluation coefficient; The acceleration / deceleration control curve corresponding to the minimum overall evaluation coefficient is used as the longitudinal lane changing motion data in the first dense traffic lane changing scenario.

7. The vehicle lane change control method according to claim 2, characterized in that: If the target lane change scenario is a second dense traffic lane change scenario, calculating longitudinal lane change motion data under the target lane change scenario includes: Using the tracking arrival space mode, the vehicle to be changed lane is accelerated and decelerated so that the speed of the vehicle to be changed lane is the same as the speed of the designated target vehicle in the lane to be changed lane; performing a correction operation on the target lane change space to obtain a lane change space to be analyzed; Performing a filtering operation on the lane change space to be analyzed to obtain lane change parameters; the lane change parameters include a reference lane change space and a speed of a target vehicle that is closest to the reference lane change space; Calculating the acceleration of the vehicle to be changed lanes according to the lane change parameter; The acceleration of the vehicle to be lane-changed is used as the longitudinal lane-changing motion data in the second dense traffic lane-changing scenario.

8. The vehicle lane change control method according to claim 2, characterized in that: If the target lane change scenario is the third dense traffic lane change scenario, calculating the longitudinal lane change motion data under the target lane change scenario includes: If a new lane-changing space is detected during the process in which the vehicle to be changed lanes and the vehicle in the lane to be changed lanes are accompanied, the speed of the vehicle to be changed lanes is dynamically compensated using the new lane-changing space to obtain a target speed; Calculating the acceleration of the vehicle to be changed lanes according to the target speed; The acceleration of the vehicle to be lane-changed is used as the longitudinal lane-changing motion data in the third dense traffic lane-changing scenario.

9. A vehicle lane change control device, characterized in that: include: A scene recognition module is used to identify the current lane change scene based on the road information of the current road where the vehicle to be changed lanes is located, and obtain a target lane change scene; A space determination module, used to determine a target lane-changing space for the vehicle to change lanes; A data calculation module, used to calculate longitudinal lane change motion data in the target lane change scenario; The lane change control module is used to control the vehicle to be changed to perform a lane change operation with reference to the target lane change space and based on the longitudinal lane change reference data and the lateral lane change motion data in the target lane change scenario.

10. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the vehicle lane change control method as described in any one of claims 1 to 8.

Images