Vehicle control method, program product, and vehicle

Abstract

The present disclosure relates to a vehicle control method, a program product and a vehicle. The vehicle control method is applied to a first vehicle and includes: in a case where the first vehicle is to change lanes to a target lane, determining a perceived second vehicle; generating N pieces of candidate trajectory information of the first vehicle; generating predicted trajectory information corresponding to M predicted driving conditions of the second vehicle, the M predicted driving conditions including predicted driving conditions with different occurrence probabilities; determining a target generation value of each trajectory combination in a plurality of trajectory combinations; and controlling the first vehicle to travel according to candidate trajectory information in a trajectory combination corresponding to a minimum target generation value. By the above scheme, the predicted driving conditions of the second vehicle with different occurrence probabilities are considered, the frequent jumping phenomenon of the first vehicle driving trajectory decision is reduced, the collision risk is reduced, and the safety and efficient passing of the first vehicle in the lane changing process are ensured.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent driving technology, specifically to a vehicle control method, a program product, and a vehicle. Background Technology

[0002] With the development of intelligent driving technology, the level of vehicle intelligence has been improved. Based on data sensed by cameras, sensors, and other devices installed on the vehicle, decisions can be made to plan the vehicle's next driving trajectory and control the vehicle's movement. Changing lanes is a common driving scenario, and how to plan the vehicle's driving trajectory to ensure safe and efficient passage during lane changes is a problem that needs to be solved. Summary of the Invention

[0003] The purpose of this disclosure is to provide a vehicle control method, program product, and vehicle to plan the driving trajectory of a vehicle during lane changing, thereby ensuring driving safety and efficient passage when changing lanes.

[0004] To achieve the above objectives, in a first aspect, this disclosure provides a vehicle control method applied to a first vehicle, the method comprising: When the first vehicle is about to change lanes to the target lane, a second vehicle is identified, including vehicles that affect the first vehicle's lane change. N candidate trajectory information of the first vehicle are generated, and the candidate trajectory information is used to characterize the driving trajectory of the first vehicle in a future time period, where N is a positive integer greater than 1; Generate predicted trajectory information corresponding to M predicted driving conditions for the second vehicle. The predicted trajectory information is used to characterize the driving trajectory of the second vehicle in the future time period if the vehicle is driven under the predicted driving conditions. The M predicted driving conditions include driving conditions with different predicted probabilities of occurrence, where M is a positive integer greater than 1. For each of the multiple trajectory combinations, a target cost value for the trajectory combination is determined, wherein the trajectory combination includes one candidate trajectory and one predicted trajectory of the second vehicle; The first vehicle is controlled to move based on the candidate trajectory information in the trajectory combination corresponding to the minimum target cost.

[0005] Optionally, the target cost of the trajectory combination is determined based on the safety cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first speed and first position of the first vehicle at multiple future times, and the predicted trajectory information in the trajectory combination includes the second speed and second position of the second vehicle at multiple future times. The safety value of the trajectory combination is determined in the following manner: For each future moment, a safe distance is determined based on the first speed of the first vehicle at the future moment, the preset maximum deceleration of the first vehicle, the second speed of the second vehicle at the future moment, and the preset maximum deceleration of the second vehicle; a first predicted distance is determined based on the first position of the first vehicle at the future moment and the second position of the second vehicle at the future moment; and a safety cost is determined based on the safe distance and the first predicted distance. The safety value of the trajectory combination is determined based on the safety value of each future moment.

[0006] Optionally, determining the safety cost at the future moment based on the safe distance and the first predicted distance includes: If the first predicted distance is less than the safe distance but greater than 0, then the safety cost at the future time is determined based on the ratio of the first predicted distance to the safe distance, wherein the larger the ratio, the smaller the safety cost at the future time.

[0007] Optionally, the target cost of the trajectory combination is determined based on the first vehicle cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first acceleration and first turning radius of the first vehicle at multiple future times, and the speed of the first vehicle at the latest future time. The first vehicle cost of the trajectory combination is determined in the following manner: For each future moment, the dynamic constraint cost of the future moment is determined based on the first acceleration and first turning radius of the first vehicle at the future moment, as well as the preset maximum acceleration and preset minimum turning radius of the first vehicle. The merging efficiency cost is determined based on the speed of the first vehicle at the latest future time and the speed limit of the target lane. The first vehicle cost of the trajectory combination is determined based on the inflow efficiency cost and the dynamic constraint cost at each of the future moments.

[0008] Optionally, determining the dynamic constraint cost at the future time based on the first vehicle's first acceleration and first turning radius at the future time, and the first vehicle's preset maximum acceleration and preset minimum turning radius, includes: If the first acceleration is greater than the preset maximum acceleration, and the first turning radius is greater than or equal to the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference between the first acceleration and the preset maximum acceleration. If the first turning radius is less than the preset minimum turning radius, and the first acceleration is less than or equal to the preset maximum acceleration, then the dynamic constraint cost is determined based on the second difference between the preset minimum turning radius and the first turning radius. If the first acceleration is greater than the preset maximum acceleration and the first turning radius is less than the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference and the second difference.

[0009] Optionally, the target cost of the trajectory combination is determined based on the second vehicle cost of the trajectory combination; the predicted trajectory information in the trajectory combination includes the second position of the second vehicle at the plurality of future times; The second vehicle value of the trajectory combination is determined in the following manner: For each future time, a second predicted distance is determined based on the current position of the second vehicle and the second position of the second vehicle at the future time; The second vehicle cost of the trajectory combination is determined based on the probability of occurrence of the predicted driving condition corresponding to the predicted trajectory information in the trajectory combination and the second predicted distance corresponding to each future time.

[0010] Optionally, determining the target cost of the trajectory combination includes: The first vehicle cost of the trajectory combination is determined based on the dynamic constraints of the first vehicle, the speed limit of the target lane, and the candidate trajectory information in the trajectory combination. The second vehicle's cost value is determined based on the current location of the second vehicle, the predicted trajectory information in the trajectory combination, and the probability of occurrence of the predicted driving conditions corresponding to the predicted trajectory information in the trajectory combination. The safety cost of the trajectory combination is determined based on the candidate trajectory information and the predicted trajectory information in the trajectory combination; The target value is determined based on the first vehicle value, the second vehicle value, and the safety value of the trajectory combination.

[0011] Optionally, the probability of occurrence of each of the M predicted driving conditions is obtained through a probability prediction model. The input of the probability prediction model includes: the remaining distance of the first vehicle changing lanes to the target lane, the trajectory information of the first vehicle at the current time and in the historical time period, the trajectory information of the second vehicle at the current time and in the historical time period, and the lane curvature of the original lane in which the first vehicle was located before changing lanes to the target lane.

[0012] In a second aspect, this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the vehicle control method provided in the first aspect of this disclosure.

[0013] Thirdly, this disclosure provides a vehicle, including: A memory on which computer programs are stored; A processor is configured to execute the computer program in the memory to implement the steps of the vehicle control method provided in the first aspect of this disclosure.

[0014] Through the above technical solution, during the process of the first vehicle changing lanes to the target lane, N candidate trajectory information for the first vehicle and M predicted driving conditions for the second vehicle can be generated. The target cost of multiple trajectory combinations is determined. Based on the candidate trajectory information in the trajectory combination corresponding to the minimum target cost, the first vehicle's driving is controlled, generating a better trajectory control decision for the first vehicle. The M predicted driving conditions include driving conditions with different probabilities of occurrence, taking into account the uncertainty of the second vehicle's next driving intention, covering multiple key driving conditions for the second vehicle, avoiding only considering the driving condition with the highest probability while ignoring possible low-probability but high-risk driving conditions. By taking into account the predicted driving conditions of the second vehicle with different probabilities of occurrence, the frequent jumps in the first vehicle's driving trajectory decision are reduced, and the risk of collision is lowered, ensuring the safe and efficient passage of the first vehicle during lane changes.

[0015] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings: Figure 1 This is an illustrative diagram illustrating a scenario where vehicles merge from a ramp onto the main road.

[0017] Figure 2 This is a flowchart illustrating an exemplary vehicle control method.

[0018] Figure 3 This is a flowchart illustrating an exemplary method for determining the safety value of a combination of trajectories.

[0019] Figure 4 This is a flowchart illustrating an exemplary method for determining the first vehicle cost value of a trajectory combination.

[0020] Figure 5This is a flowchart illustrating an exemplary method for determining the second vehicle cost value of a trajectory combination.

[0021] Figure 6 This is a flowchart illustrating an exemplary method for determining the target value of a combination of trajectories.

[0022] Figure 7 This is a block diagram of an exemplary vehicle control device.

[0023] Figure 8 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0024] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0025] First, let's introduce the application scenarios of this disclosure. The safe and efficient passage of heavy trucks merging onto ramps has become a challenge for intelligent driving systems for commercial vehicles. Heavy trucks have inherent characteristics such as large size, high inertia, long braking distance, and large turning radius. In the ramp merging scenario, the traffic flow connection between the main road and the ramp is complex. When merging, it is necessary to seize the safe traffic flow gap on the main road, which poses a challenge to the decision-making and planning capabilities of heavy trucks.

[0026] Figure 1 This is an exemplary schematic diagram illustrating a scenario where vehicles merge from a ramp onto the main road, such as... Figure 1 As shown, vehicle 11, which is equivalent to a rectangle in ramp 101, is the vehicle itself, i.e., the first vehicle mentioned below. Vehicle 11 is, for example, a commercial vehicle such as a heavy truck. Traffic flows orderly in the one-way lane of the main road. Vehicle 11 travels along ramp 101 to the end, about to merge into the target lane 102 of the main road. (Refer to...) Figure 1 As an example, in one embodiment, the vehicle control method of this disclosure can be applied to the process of a vehicle merging from a ramp into a main road.

[0027] It should be noted that the following text will be in the format of "The 'The 'The 'The 'The 'The 'The 'The 'The '' The " The " The " The " The " The " The " The " The " The " The " The " The " The " The " The " The The "The'" The '"""The 'The '" ... Figure 1 The example shown is of a ramp merging into the main road. In other embodiments, the original lane where the vehicle was before changing lanes and the target lane to be entered can be two adjacent lanes on the same road. That is, the vehicle control method disclosed herein can be applied to the process of a vehicle changing lanes from one lane to another adjacent lane.

[0028] Furthermore, the fact that vehicle 11 is a heavy-duty truck is merely exemplary. The vehicle disclosed herein can be any vehicle equipped with an intelligent driving system, which is capable of planning the vehicle's driving trajectory and controlling the vehicle's movement.

[0029] Figure 2is a flowchart illustrating an exemplary vehicle control method, as Figure 2 shown, the method can be applied to a controller of a first vehicle, where the first vehicle may be a vehicle equipped with an intelligent driving system, and the host vehicle mentioned hereinafter refers to the first vehicle. As Figure 2 shown, the vehicle control method includes step 21 to step 25.

[0030] Step 21: determining a perceived second vehicle when the first vehicle is to change lanes to a target lane.

[0031] A lane change of the first vehicle may refer to the first vehicle changing from an original lane to the target lane. For example, as Figure 1 shown, the original lane is a ramp or an auxiliary road, and the target lane is a lane on the main road that the first vehicle intends to enter. As another example, the original lane and the target lane may be two adjacent lanes on the same road.

[0032] A perception system is installed on the first vehicle. The perception system includes, for example, perception devices such as a camera, a millimeter-wave sensor and a radar sensor, and the perception system can perceive data such as the position, speed, acceleration, orientation, and size of objects within a certain range. Referring to Figure 1 shown, when the vehicle 11 is about to merge into the main road, the vehicle 11 can perceive vehicles traveling in the target lane 102.

[0033] The second vehicle includes vehicles that have an impact on the lane change of the first vehicle. For example, taking merging from a ramp into the main road as an example, the second vehicle may include a preceding vehicle and a following vehicle on the target lane. The preceding vehicle is the vehicle that is located ahead of the first vehicle on the target lane and is closest to the first vehicle. The preceding vehicle is, for example, Figure 1 the vehicle 12 shown. The following vehicle is the vehicle that is located behind the first vehicle on the target lane and is closest to the first vehicle. The following vehicle is, for example, Figure 1 the vehicle 13 shown. If there is no vehicle ahead of the first vehicle on the target lane but there is a vehicle behind, that vehicle behind can be used as the second vehicle; or if there is no vehicle behind the first vehicle on the target lane but there is a vehicle ahead, that vehicle ahead can be used as the second vehicle. The number of second vehicles can be one or more, and is not limited.

[0034] Taking the first vehicle changing lanes to an adjacent lane on the same road as an example, the second vehicle may include a preceding vehicle and a following vehicle on the original lane, and a preceding vehicle and a following vehicle on the target lane. The preceding vehicle on the original lane is the vehicle that is located ahead of the first vehicle on the original lane and is closest to the first vehicle, and the following vehicle on the original lane is the vehicle that is located behind the first vehicle on the original lane and is closest to the first vehicle.

[0035] The following embodiments are described by taking an example where the second vehicle includes Figure 1 the vehicle 12 and the vehicle 13 shown.

[0036] Step 22: Generate N candidate trajectory information for the first vehicle.

[0037] Candidate trajectory information is used to characterize the driving trajectory of the first vehicle in the future time period, where N is a positive integer greater than 1.

[0038] The N candidate trajectory information corresponds one-to-one with the N candidate driving conditions of the first vehicle. The N candidate driving conditions include, for example, accelerating towards the target lane, decelerating towards the target lane, and driving at a constant speed towards the target lane. Among them, accelerating towards the target lane is the condition in which the first vehicle accelerates at a preset first acceleration, decelerating towards the target lane is the condition in which the first vehicle decelerates at a preset first deceleration, and driving at a constant speed towards the target lane is the condition in which the first vehicle drives at a constant speed at its current speed.

[0039] In one example, the preset first acceleration is, for instance, 2.5 m / s². 2 The preset first deceleration is 2.5 m / s². 2 The N candidate driving conditions may include accelerations of 2.5 m / s². 2 Accelerate, then decelerate at 2.5 m / s 2 There are three operating conditions: deceleration, constant speed at the current vehicle speed, and driving at a constant speed.

[0040] In one example, the preset first acceleration includes 2.5 m / s². 2 and 3m / s 2 The preset first deceleration includes 2.5 m / s². 2 and 3m / s 2 The N candidate driving conditions may include accelerations of 2.5 m / s². 2 Accelerate at an acceleration of 3m / s 2 Accelerate, then decelerate at 2.5 m / s 2 Decelerate to a speed of 3 m / s 2 The five operating conditions are: decelerating, maintaining a constant current speed, and driving at a constant speed.

[0041] The above example is for illustrative purposes only. This disclosure does not limit the value of N. The preset first acceleration and first deceleration can be one or more.

[0042] The duration of a future time period can be preset. A future time period includes multiple future moments, and the earliest of these future moments can be the next moment after the current moment. In other words, trajectory planning is about planning how the vehicle should move next.

[0043] The candidate trajectory information of the first vehicle may include the first speed, first position, first acceleration, and first turning radius of the first vehicle at multiple future moments. It can be understood as being used to reflect where the first vehicle is and how fast it is at each future moment.

[0044] Step 23: Generate the predicted trajectory information corresponding to the M predicted driving conditions of the second vehicle.

[0045] The predicted trajectory information is used to characterize the trajectory of a second vehicle in a future time period if it travels under the predicted driving conditions. The M predicted driving conditions include driving conditions with different probabilities of occurrence. M is a positive integer greater than 1.

[0046] M types of predicted driving conditions include, for example, acceleration driving conditions, deceleration driving conditions, and constant speed driving conditions. Among them, acceleration driving conditions can be conditions in which the second vehicle accelerates in the current direction with a preset second acceleration, deceleration driving conditions can be conditions in which the second vehicle decelerates in the current direction with a preset second deceleration, and constant speed driving conditions can be conditions in which the second vehicle travels at a constant speed in the current direction at the current vehicle speed.

[0047] For specific examples of predicted driving conditions, please refer to the explanation of candidate driving conditions above. There are no restrictions on the value of M. The first acceleration may be the same as or different from the second acceleration, and the first deceleration may be the same as or different from the second deceleration.

[0048] The number of second vehicles can be multiple. In step 23, predicted trajectory information for each second vehicle can be generated, for example, generating... Figure 1 The M predicted trajectory information of vehicle 12 and vehicle 13 are shown.

[0049] The predicted trajectory information of the second vehicle may include the second speed and second position of the second vehicle at multiple future moments. It can be understood as where the second vehicle will be and how fast it will be at each future moment if the second vehicle travels according to the predicted driving conditions.

[0050] Candidate trajectory information is generated based on candidate driving conditions, and predicted trajectory information is generated based on predicted driving conditions. For example, it can be generated through a vehicle dynamics model. The method of generating vehicle driving trajectory information through a vehicle dynamics model can be referred to in relevant technologies.

[0051] In this disclosure, the M predicted driving conditions include driving conditions with different probabilities of occurrence. For example, the probability that the vehicle in front is traveling at a constant speed is P11, the probability that it accelerates with a certain acceleration (which can increase the distance between it and the vehicle behind) is P12, and the probability that it decelerates with a certain deceleration (which can decrease the distance between it and the vehicle behind) is P13. The probability that the vehicle behind is traveling at a constant speed is P21, the probability that it accelerates with a certain acceleration (which can decrease the distance between it and the vehicle in front) is P22, and the probability that it decelerates with a certain deceleration (which can increase the distance between it and the vehicle in front) is P23.

[0052] The next driving intention of the second vehicle (such as the following vehicle accelerating to overtake, decelerating to yield, or maintaining a constant speed) is uncertain. If only the most probable situation of the second vehicle is considered, while ignoring the possible low-probability but high-risk situations, such as the following vehicle suddenly accelerating or the preceding vehicle suddenly decelerating, the actual trajectory of the second vehicle deviates from the predicted highest probability driving trajectory. Due to the characteristics of heavy trucks, such as large size, strong inertia, and long braking distance, if the actual trajectory of the second vehicle deviates from the predicted highest probability driving trajectory, it will lead to the risk of sudden braking and collision when the first vehicle needs to suddenly adjust its driving trajectory.

[0053] For example, if only the predicted scenario with the highest probability of the second vehicle is considered (such as the vehicle in front accelerating and the vehicle behind maintaining a constant speed), the trajectory planning module will generate a decision for the vehicle to accelerate and merge. However, this approach may ignore low-probability but high-risk scenarios (such as the vehicle behind suddenly accelerating), which could lead to a collision risk when the heavy truck accelerates and merges.

[0054] Moreover, as the second vehicle suddenly accelerates or decelerates, the perception data of this vehicle is updated, and the predicted driving trajectory of the second vehicle with the highest probability will change accordingly. This leads to frequent jumps and updates in the driving trajectory decision of this vehicle. Heavy trucks have strict dynamic constraints, such as a large turning radius. Frequent jumps and updates in trajectory decision will reduce driving stability and also affect the continuity of traffic flow in the target lane.

[0055] Therefore, in this disclosure, the predicted driving conditions of the second vehicle with different probabilities of occurrence are considered, covering the key behaviors of the second vehicle such as acceleration, deceleration and constant speed, avoiding ignoring the low probability but high risk behaviors that the second vehicle may have, and taking into account the predicted driving conditions with different probabilities of occurrence, the frequent jumps in the vehicle's driving trajectory decision and the risk of collision are reduced.

[0056] Step 24: For each of the multiple trajectory combinations, determine the target cost of the trajectory combination.

[0057] The trajectory combination includes a candidate trajectory information and a predicted trajectory information of the second vehicle. The target cost value can represent the cost value if the first vehicle travels with the candidate trajectory information and the second vehicle travels with the predicted trajectory information.

[0058] Step 25: Control the first vehicle to move based on the candidate trajectory information in the trajectory combination corresponding to the minimum target cost.

[0059] The candidate trajectory information in the trajectory combination corresponding to the minimum target cost is optimal for the first vehicle, and therefore can be used as the target trajectory information. The first vehicle's driving can be controlled based on this target trajectory information in the future time period. For example, this target trajectory information can be converted into longitudinal control information such as throttle and lateral control information such as steering angle to control the first vehicle's driving.

[0060] Through the above technical solution, during the process of the first vehicle changing lanes to the target lane, N candidate trajectory information for the first vehicle and M predicted driving conditions for the second vehicle can be generated. The target cost of multiple trajectory combinations is determined. Based on the candidate trajectory information in the trajectory combination corresponding to the minimum target cost, the first vehicle's driving is controlled, generating a better trajectory control decision for the first vehicle. The M predicted driving conditions include driving conditions with different probabilities of occurrence, taking into account the uncertainty of the second vehicle's next driving intention, covering multiple key driving conditions for the second vehicle, avoiding only considering the driving condition with the highest probability while ignoring possible low-probability but high-risk driving conditions. By taking into account the predicted driving conditions of the second vehicle with different probabilities of occurrence, the frequent jumps in the first vehicle's driving trajectory decision are reduced, and the risk of collision is lowered, ensuring the safe and efficient passage of the first vehicle during lane changes.

[0061] Optionally, the probability of occurrence of each of the M predicted driving conditions can be obtained through a probability prediction model. The inputs of the probability prediction model include: the remaining distance of the first vehicle changing lanes to the target lane, the trajectory information of the first vehicle at the current time and in the historical time period, the trajectory information of the second vehicle at the current time and in the historical time period, and the lane curvature of the original lane in which the first vehicle was located before changing lanes to the target lane.

[0062] The remaining distance can be the length of the first vehicle's current position from the target lane. Since the first vehicle influences the second vehicle's movement—for example, if the first vehicle accelerates, the following vehicle might decelerate—the input to the probabilistic prediction model can include the trajectory information of the first vehicle at the current moment and in historical time periods. The trajectory information of the second vehicle at the current moment and in historical time periods can be obtained through the first vehicle's perception data. Vehicle trajectory information includes, for example, data such as position, speed, and acceleration. The lane curvature of the original lane is, for example, the curvature of ramp 101. Inputting this data into the probabilistic prediction model, the model can output the probability of occurrence for each of M predicted driving conditions.

[0063] For example, the probabilistic prediction model can employ a graph neural network (GNN). The probabilistic prediction model can be a pre-trained model or a trained model. The sample data in the training data can refer to the data described above used as input to the probabilistic prediction model, and the labeled data can include the probabilities of occurrence of various driving conditions.

[0064] By employing a probabilistic prediction model, the probability of occurrence of M predicted driving conditions for the second vehicle can be accurately predicted. Driving conditions with different probabilities of occurrence can cover a variety of key conditions for the second vehicle, thereby taking into account the uncertainty of the second vehicle's next driving intention when making driving decisions for the vehicle.

[0065] In one embodiment, the vehicle can construct different passable areas based on perceived obstacles. Static obstacles such as ramp guardrails and main road medians can be marked as impassable areas, providing obstacle avoidance constraints for generating candidate trajectory information. Points within impassable areas are denoted as (x0, y0), and the vehicle's position in the candidate trajectory information is denoted as (x, y). The distance D between (x, y) and the obstacle... obs (x,y) must satisfy D obs (x,y)≥D safe D obs (x,y) can be obtained through D safe For example, add a preset safety margin (such as 0.5m) to the width of the vehicle.

[0066] If D obs (x,y)<D safe This indicates that there is a risk of collision between the vehicle and the obstacle, and such points (x, y) can be excluded when generating candidate trajectory information.

[0067] In addition, with the goal of successfully merging into the target lane, the generated candidate trajectory information must satisfy the speed constraint v. ramp,max ≤v(t)≤v main,max v(t) is the speed of the vehicle at a future time t. ramp,max The speed limit for the original lane, for example, a speed limit of 40 km / h for ramp 101, v main,max The maximum speed limit is set to the maximum speed limit of the main road, for example, the maximum speed limit of target lane 102. In this way, as the vehicle merges from the ramp, its speed gradually increases from the ramp speed limit to the maximum speed limit of the main road, and remains below the maximum speed limit of the main road.

[0068] In one embodiment, the target cost of the trajectory combination can be determined based on the safety cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first speed and first position of the first vehicle at multiple future times, and the predicted trajectory information in the trajectory combination includes the second speed and second position of the second vehicle at the multiple future times; such as Figure 3 As shown, the safety value of the trajectory combination can be determined through steps 31 and 32.

[0069] Step 31: For each future moment, determine the corresponding safe distance based on the first speed of the first vehicle at the future moment, the preset maximum deceleration of the first vehicle, the second speed of the second vehicle at the future moment, and the preset maximum deceleration of the second vehicle; determine the corresponding first predicted distance based on the first position of the first vehicle at the future moment and the second position of the second vehicle at the future moment; determine the safety cost value for the future moment based on the safe distance and the first predicted distance.

[0070] Step 32: Determine the safety value of the trajectory combination based on the safety value at each future moment.

[0071] For example, let time t0 represent the current time, and the number of future times be R, including times t1, t2, and up to tR. The current position of the second vehicle perceived by the first vehicle, as well as the first position of the first vehicle and the second position of the second vehicle at future times, can be represented by two-dimensional coordinates in the vehicle coordinate system of the first vehicle. In the vehicle coordinate system, for example, the positive direction of the x-axis is from the rear to the front of the vehicle, and the negative direction is from the front to the rear of the vehicle; the positive direction of the y-axis is from the right side to the left side of the vehicle, and the negative direction is from the left side to the right side of the vehicle.

[0072] For the j-th second vehicle, the safe distance at the i-th future time ti is represented as: , i∈[1,R]. d safe (ti) can be determined by the following formula (1): (1) in, This represents the first speed of the first vehicle at a future time ti. This indicates the preset system response time, for example, a value of 1 second. This indicates the preset maximum deceleration of the first vehicle. Let represent the second velocity of the j-th second vehicle at a future time ti. This represents the preset maximum deceleration of the j-th second vehicle. It represents the product.

[0073] The preset maximum deceleration of the first vehicle can be stored in the memory of the first vehicle. The memory of the first vehicle can also store the preset maximum deceleration of various types of vehicles. Based on the type of the second vehicle identified, the preset maximum deceleration of the second vehicle can be determined.

[0074] This can be represented as the safe following distance at which the first vehicle and the j-th second vehicle will not collide in the future at time ti. In this disclosure, for each future time, a corresponding safe distance is determined to measure whether there is a risk of collision between the first vehicle and the second vehicle at that future time, so that the safe distance changes dynamically with time and improves the accuracy of collision risk prediction.

[0075] The distance between the first position of the first vehicle at a future time and the second position of the second vehicle at a future time is taken as the first predicted distance corresponding to that future time. For the j-th second vehicle, the first predicted distance corresponding to the i-th future time ti is expressed, for example, as: .

[0076] Optionally, in step 31, determining the safety cost value at a future time based on the safe distance and the first predicted distance includes: if the first predicted distance is less than the safe distance but greater than 0, then determining the safety cost value at a future time based on the ratio of the first predicted distance to the safe distance, wherein the larger the ratio, the smaller the safety cost value at a future time; If the first predicted distance is greater than or equal to the safe distance, the safety cost at future time can be 0; if the first predicted distance is less than or equal to 0, the safety cost at future time can be a preset value, which can be a large value.

[0077] In one embodiment, the following formula (2) is used to determine the safe distance. and first predicted distance Determine the safety value of the first vehicle and the j-th second vehicle at a future time ti. .

[0078] (2) Here, k is a preset safety penalty coefficient, for example, 100. Thus, when the first predicted distance is less than the safe distance but greater than 0, the larger the ratio of the first predicted distance to the safe distance, the lower the collision risk, and the lower the corresponding safety cost at future moments. A first predicted distance less than or equal to 0 indicates that the trajectories of the first and second vehicles overlap, in which case the safety cost can be infinitely large, indicating a collision risk.

[0079] For example, the sum of the safety values ​​at each future moment can be used as the safety value of the first vehicle and the j-th second vehicle, and the sum of the safety values ​​of the first vehicle and each of the second vehicles can be used as the safety value of the trajectory combination.

[0080] Using the above technical solution, the safe distance at a future time can be used to measure whether there is a collision risk between the first and second vehicles at that future time. The safe distance changes dynamically with time, which can improve the accuracy of collision risk prediction. By comparing the first predicted distance with the safe distance, the safety value at the future time can be determined. Based on the safety value at each future time, the safety value of the trajectory combination is obtained. The safety value of the trajectory combination can represent the magnitude of the collision risk when driving according to the trajectory combination.

[0081] In one embodiment, the target cost of the trajectory combination is determined based on the first vehicle cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first vehicle's first acceleration and first turning radius at multiple future times, and the first vehicle's speed at the latest future time; such as Figure 4 As shown, the first vehicle value of the trajectory combination can be determined through steps 41 to 43.

[0082] Step 41: For each future moment, determine the dynamic constraint cost of the future moment based on the first acceleration and first turning radius of the first vehicle at the future moment, as well as the preset maximum acceleration and preset minimum turning radius of the first vehicle.

[0083] Considering the characteristics of heavy trucks, such as large turning radius and strong inertia, the dynamic constraints of the first vehicle may include the first vehicle's first acceleration being less than the preset maximum acceleration and the first turning radius being greater than the preset minimum turning radius. In other words, the driving of the first vehicle needs to meet certain dynamic constraints.

[0084] Alternatively, step 41 can be implemented as follows: If the first acceleration is greater than the preset maximum acceleration and the first turning radius is greater than or equal to the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference between the first acceleration and the preset maximum acceleration.

[0085] For example, the first acceleration of the first vehicle at a future time ti is denoted as a(ti), the first turning radius is denoted as R(ti), and the velocity of the first vehicle at the latest future time, i.e., time tR, is denoted as v(T). The preset maximum acceleration of the first vehicle is denoted as... The preset minimum turning radius of the first vehicle is denoted as R. min For example, set it to 25m.

[0086] like ,and The dynamic constraint cost of future time ti Among them, k acc This represents the preset acceleration penalty coefficient.

[0087] In this embodiment, if at a future time *ti*, the acceleration of the first vehicle does not meet the dynamic constraints, but the turning radius does meet the dynamic constraints, the first difference can be used to determine the vehicle's acceleration. Determine the dynamic constraint cost of future time ti The larger this first difference is, The larger.

[0088] If the first turning radius is less than the preset minimum turning radius and the first acceleration is less than or equal to the preset maximum acceleration, then the dynamic constraint cost is determined based on the second difference between the preset minimum turning radius and the first turning radius.

[0089] like ,and The dynamic constraint cost of future time ti k rad This represents the preset acceleration penalty coefficient.

[0090] In this embodiment, if at a future time *ti*, the acceleration of the first vehicle meets the dynamic constraints, but the turning radius does not meet the dynamic constraints, the second difference can be used as a basis. Determine the dynamic constraint cost of future time ti The larger this second difference is, The larger.

[0091] If the first acceleration is greater than the preset maximum acceleration and the first turning radius is less than the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference and the second difference.

[0092] like ,and The dynamic constraint cost of future time ti .

[0093] In this embodiment, if the acceleration of the first vehicle does not meet the dynamic constraints and the turning radius does not meet the dynamic constraints at a future time *ti*, the dynamic constraint cost at the future time *ti* can be determined based on the first difference and the second difference. The larger the first difference and / or the second difference, The larger.

[0094] Step 42: Determine the merging efficiency cost based on the speed of the first vehicle at the latest future time and the speed limit of the target lane.

[0095] For example, the speed limit for the target lane is represented as v. main,max The cost of merging trajectory combinations Through formula Calculate. k vel This represents a coefficient indicating the preset import efficiency.

[0096] Thus, the closer the speed of the first vehicle at the latest future moment is to the speed limit of the target lane, the faster the first vehicle can merge into the target lane, which means that the merging efficiency is higher and the corresponding cost of merging efficiency is lower.

[0097] Step 43: Determine the first vehicle cost of the trajectory combination based on the inflow efficiency cost and the dynamic constraint cost at each future moment.

[0098] The first vehicle cost of the trajectory combination can be determined by the following formula (3). .

[0099] (3) Using the above technical solution, the cost of the first vehicle is obtained based on the merging efficiency cost and the cost of dynamic constraints at each future time. The merging efficiency cost can be used to characterize the merging efficiency of the first vehicle merging into the target lane if it travels according to the candidate trajectory information in the trajectory combination. The cost of dynamic constraints at future time can be used to characterize the situation where the first vehicle's travel at that future time meets the dynamic constraints.

[0100] In one embodiment, the target cost of the trajectory combination is determined based on the second vehicle cost of the trajectory combination; the predicted trajectory information in the trajectory combination includes the second position of the second vehicle at multiple future times; like Figure 5 As shown, the second vehicle value of the trajectory combination can be determined through steps 51 and 52.

[0101] Step 51: For each future time, determine the second predicted distance based on the current position of the second vehicle and the second position of the second vehicle at the future time.

[0102] Step 52: Determine the second vehicle cost of the trajectory combination based on the probability of occurrence of the predicted driving conditions corresponding to the predicted trajectory information in the trajectory combination and the second predicted distance corresponding to each future time.

[0103] For example, the current position of the j-th second vehicle is represented as The second position of the j-th second vehicle at future time ti is represented as... P j This represents the probability of the predicted driving condition occurring for the j-th second vehicle in the trajectory combination. The second vehicle cost of the j-th second vehicle... It can be determined by formula (4).

[0104] (4) Through the above technical solution, the second predicted distance can represent the distance between the predicted position of the second vehicle at a future time and the current position of the second vehicle. The cost value of the second vehicle is determined based on the occurrence probability and the second predicted distance. The cost value of the second vehicle can reflect the degree of matching between the predicted driving trajectory of the second vehicle and the current trajectory of the second vehicle.

[0105] In one embodiment, the implementation method for determining the target cost value of trajectory combinations may be as follows: Figure 6 As shown, steps 61 to 64 are included.

[0106] Step 61: Determine the first vehicle cost of the trajectory combination based on the dynamic constraints of the first vehicle, the speed limit of the target lane, and the candidate trajectory information in the trajectory combination.

[0107] Step 62: Determine the value of the second vehicle in the trajectory combination based on the current position of the second vehicle, the predicted trajectory information in the trajectory combination, and the probability of occurrence of the predicted driving conditions corresponding to the predicted trajectory information in the trajectory combination.

[0108] Step 63: Determine the safety value of the trajectory combination based on the candidate trajectory information and the predicted trajectory information in the trajectory combination.

[0109] Step 64: Determine the target value based on the first vehicle value, the second vehicle value, and the safety value of the trajectory combination.

[0110] The implementation method of step 61 can be referred to Figure 4 Step 61 includes steps 41 to 43. The implementation of step 62 can be found in [reference needed]. Figure 5 Step 62 includes steps 51 and 52. The implementation of step 63 can be found in [reference needed]. Figure 3 Step 63 includes steps 31 and 32.

[0111] For example, the weights of the three cost values ​​can be preset, such as the weight of the first vehicle cost value being 0.3, the weight of the second vehicle cost value being 0.2, and the weight of the safety cost value being 0.5. The target cost value can be obtained by weighting the first vehicle cost value, the second vehicle cost value, and the safety cost value.

[0112] In this embodiment, the target cost is determined based on three cost values, which can more comprehensively and accurately measure the cost of trajectory combination, thereby finding candidate trajectory information that is safe and takes into account merging efficiency, and using a better method to control the first vehicle to ensure driving safety and efficient passage when the vehicle changes lanes.

[0113] In one embodiment, determining the target cost of the trajectory combination may include steps 61 and 63, and the target cost of the trajectory combination is determined based on the weighted value of the first vehicle cost and the safety cost, wherein the weights of the first vehicle cost and the safety cost can be preset. In this embodiment, the cost of the trajectory combination may be measured by considering factors such as the vehicle's merging efficiency, whether the vehicle meets the dynamic constraints, and the collision risk between the vehicle and vehicles in the target lane.

[0114] In one embodiment, determining the target cost of the trajectory combination may include steps 62 and 63, and determining the target cost of the trajectory combination based on the cost of the second vehicle and the safety cost. In this embodiment, the cost of the trajectory combination may be measured by considering the degree of matching between the predicted driving trajectory of the second vehicle and the current trajectory of the second vehicle, and the collision risk between the vehicle and vehicles in the target lane.

[0115] In one embodiment, determining the target cost of the trajectory combination may include step 63, and using the safety cost of the trajectory combination as the target cost to accurately predict the collision risk between the vehicle and vehicles in the target lane.

[0116] Optionally, the vehicle control method may also include: If, during the process of controlling the first vehicle's movement, the deviation between the actual trajectory information of the second vehicle and the first predicted trajectory information meets a preset condition, then steps 21 to 25 are executed again. The first predicted trajectory information is the predicted trajectory information in the trajectory combination corresponding to the minimum target cost value.

[0117] The deviation between the actual trajectory information of the second vehicle and the first predicted trajectory information can be determined based on the deviation between the actual speed and the predicted speed of the second vehicle at the same time, or based on the deviation between the actual position and the predicted position of the second vehicle at the same time, or simultaneously based on both speed and position deviations. A preset condition is, for example, that the deviation is greater than a preset deviation threshold. If the deviation meets the preset condition, it indicates that the second vehicle is not actually driving according to the predicted driving conditions of the first vehicle. In this case, the first vehicle's driving trajectory needs to be adaptively adjusted to find a new driving trajectory that minimizes the target cost, i.e., trajectory replanning.

[0118] For example, the vehicle can execute the vehicle control method of this disclosure to safely and efficiently merge into the target lane when the remaining distance is less than a preset distance threshold and the target gap is greater than or equal to a preset gap threshold. The target gap can be the gap between the preceding and following vehicles in the target lane, and the preset gap threshold can be determined based on the length and braking distance of the vehicle. Figure 1 As shown, the target gap refers to the distance between vehicle 12 and vehicle 13, and the target gap can be obtained through the vehicle's perception data.

[0119] During lane changing, the vehicle's status can include: waiting to merge, merging, and merging complete. The initial status is waiting to merge. The status update logic includes: if the vehicle is currently in the waiting to merge state and detects that the target gap is greater than or equal to a preset gap threshold, then update to the merging state and mark it as needing trajectory replanning; if the vehicle is currently in the merging state and has completely merged into the target lane (as confirmed by lane line detection), then update to the merging complete state and clear the relevant cached information of trajectory planning during this lane change; if the vehicle is currently in the merging state and detects a sudden change in the operating condition of the second vehicle, such as the following vehicle suddenly accelerating or the preceding vehicle suddenly decelerating, then maintain the merging state and mark it as needing trajectory replanning.

[0120] In this disclosure, the triggering conditions for trajectory replanning of the vehicle may include: entering the merging state for the first time; the deviation between the actual trajectory information of the second vehicle and the first predicted trajectory information meets the preset conditions; there is a collision risk between the vehicle and the obstacle; the vehicle travels to the end of the planned trajectory. If any of these conditions are triggered, the vehicle can re-execute steps 21 to 25, that is, perform trajectory replanning.

[0121] Therefore, this disclosure takes into account both the dynamic constraints of the vehicle and the uncertainty of the vehicle's driving conditions in the target lane, so as to achieve safe, efficient and stable passage in scenarios such as merging into a ramp or changing adjacent lanes. It ensures that the decision trajectory conforms to the dynamic constraints of the vehicle, ensures the feasibility of the trajectory, reduces trajectory decision jumps, ensures driving stability, and enables the vehicle to pass safely and efficiently when changing lanes.

[0122] Based on the same inventive concept, this disclosure also provides a vehicle control device. Figure 7 This is a block diagram of an exemplary vehicle control device 70, which can be applied to a first vehicle, such as... Figure 7 As shown, the vehicle control device 70 includes: The first determining module 71 is used to determine a perceived second vehicle when the first vehicle is about to change lanes to the target lane. The second vehicle includes vehicles that affect the lane change of the first vehicle. The first generation module 72 is used to generate N candidate trajectory information for the first vehicle, wherein the candidate trajectory information is used to characterize the driving trajectory of the first vehicle in a future time period, and N is a positive integer greater than 1. The second generation module 73 is used to generate prediction trajectory information corresponding to M predicted driving conditions of the second vehicle. The prediction trajectory information is used to characterize the driving trajectory of the second vehicle in the future time period if the predicted driving conditions are used. The M predicted driving conditions include driving conditions with different predicted probabilities of occurrence, and M is a positive integer greater than 1. The second determining module 74 is used to determine the target cost of each trajectory combination among multiple trajectory combinations, wherein the trajectory combination includes a candidate trajectory information and a predicted trajectory information of the second vehicle. The control module 75 is used to control the first vehicle to drive based on the candidate trajectory information in the trajectory combination corresponding to the minimum target cost.

[0123] Optionally, the target cost of the trajectory combination is determined based on the safety cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first speed and first position of the first vehicle at multiple future times, and the predicted trajectory information in the trajectory combination includes the second speed and second position of the second vehicle at multiple future times. The second determining module 74 is used to determine the safety value of the trajectory combination in the following manner: For each future moment, a safe distance is determined based on the first speed of the first vehicle at the future moment, the preset maximum deceleration of the first vehicle, the second speed of the second vehicle at the future moment, and the preset maximum deceleration of the second vehicle; a first predicted distance is determined based on the first position of the first vehicle at the future moment and the second position of the second vehicle at the future moment; and a safety cost is determined based on the safe distance and the first predicted distance. The safety value of the trajectory combination is determined based on the safety value of each future moment.

[0124] Optionally, the second determining module 74 is configured to: if the first predicted distance is less than the safe distance and greater than 0, determine the safety cost value at the future time based on the ratio of the first predicted distance to the safe distance, wherein the larger the ratio, the smaller the safety cost value at the future time.

[0125] Optionally, the target cost of the trajectory combination is determined based on the first vehicle cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first acceleration and first turning radius of the first vehicle at multiple future times, and the speed of the first vehicle at the latest future time. The second determining module 74 is used to determine the first vehicle cost of the trajectory combination in the following manner: For each future moment, the dynamic constraint cost of the future moment is determined based on the first acceleration and first turning radius of the first vehicle at the future moment, as well as the preset maximum acceleration and preset minimum turning radius of the first vehicle. The merging efficiency cost is determined based on the speed of the first vehicle at the latest future time and the speed limit of the target lane. The first vehicle cost of the trajectory combination is determined based on the inflow efficiency cost and the dynamic constraint cost at each of the future moments.

[0126] Optionally, the second determining module 74 is used to: If the first acceleration is greater than the preset maximum acceleration, and the first turning radius is greater than or equal to the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference between the first acceleration and the preset maximum acceleration. If the first turning radius is less than the preset minimum turning radius, and the first acceleration is less than or equal to the preset maximum acceleration, then the dynamic constraint cost is determined based on the second difference between the preset minimum turning radius and the first turning radius. If the first acceleration is greater than the preset maximum acceleration and the first turning radius is less than the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference and the second difference.

[0127] Optionally, the target cost of the trajectory combination is determined based on the second vehicle cost of the trajectory combination; the predicted trajectory information in the trajectory combination includes the second position of the second vehicle at the plurality of future times; The second determining module 74 is used to determine the second vehicle cost of the trajectory combination in the following manner: For each future time, a second predicted distance is determined based on the current position of the second vehicle and the second position of the second vehicle at the future time; The second vehicle cost of the trajectory combination is determined based on the probability of occurrence of the predicted driving condition corresponding to the predicted trajectory information in the trajectory combination and the second predicted distance corresponding to each future time.

[0128] Optionally, the second determining module 74 includes: The first determining submodule is used to determine the first vehicle cost of the trajectory combination based on the dynamic constraints of the first vehicle, the speed limit of the target lane, and the candidate trajectory information in the trajectory combination. The second determining submodule is used to determine the second vehicle cost of the trajectory combination based on the current position of the second vehicle, the predicted trajectory information in the trajectory combination, and the probability of occurrence of the predicted driving condition corresponding to the predicted trajectory information in the trajectory combination. The third determining submodule is used to determine the safety cost of the trajectory combination based on the candidate trajectory information and the predicted trajectory information in the trajectory combination; The fourth determining submodule is used to determine the target value based on the first vehicle value, the second vehicle value, and the safety value of the trajectory combination.

[0129] Optionally, the probability of occurrence of each of the M predicted driving conditions is obtained through a probability prediction model. The input of the probability prediction model includes: the remaining distance of the first vehicle changing lanes to the target lane, the trajectory information of the first vehicle at the current time and in the historical time period, the trajectory information of the second vehicle at the current time and in the historical time period, and the lane curvature of the original lane in which the first vehicle was located before changing lanes to the target lane.

[0130] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0131] Figure 8 This is a block diagram illustrating an electronic device 1900 according to an exemplary embodiment. For example, the electronic device 1900 may be provided as a controller. (Refer to...) Figure 8 The electronic device 1900 includes a processor 1922, which may be one or more, and a memory 1932 for storing computer programs executable by the processor 1922. The computer program stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processor 1922 may be configured to execute the computer program to perform the vehicle control method described above.

[0132] Additionally, the electronic device 1900 may also include a power supply component 1926 and a communication component 1950. The power supply component 1926 can be configured to perform power management of the electronic device 1900, and the communication component 1950 can be configured to enable communication of the electronic device 1900, such as wired or wireless communication. Furthermore, the electronic device 1900 may also include an input / output (I / O) interface 1958. The electronic device 1900 can operate on an operating system stored in memory 1932.

[0133] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the vehicle control method described above. For example, the computer-readable storage medium may be the memory 1932 including program instructions, which may be executed by the processor 1922 of the electronic device 1900 to complete the vehicle control method described above.

[0134] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a processor, which, when executed by the processor, implements the steps of the vehicle control method described above.

[0135] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0136] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0137] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A vehicle control method characterized by, Applied to a first vehicle, the method includes: When the first vehicle is about to change lanes to the target lane, a second vehicle is identified, including vehicles that affect the first vehicle's lane change. N candidate trajectory information of the first vehicle are generated, and the candidate trajectory information is used to characterize the driving trajectory of the first vehicle in a future time period, where N is a positive integer greater than 1; Generate predicted trajectory information corresponding to M predicted driving conditions for the second vehicle. The predicted trajectory information is used to characterize the driving trajectory of the second vehicle in the future time period if the vehicle is driven under the predicted driving conditions. The M predicted driving conditions include driving conditions with different predicted probabilities of occurrence, where M is a positive integer greater than 1. For each of the multiple trajectory combinations, a target cost value for the trajectory combination is determined, wherein the trajectory combination includes one candidate trajectory and one predicted trajectory of the second vehicle; The first vehicle is controlled to move based on the candidate trajectory information in the trajectory combination corresponding to the minimum target cost.

2. The method of claim 1, wherein, The target cost of the trajectory combination is determined based on the safety cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first speed and first position of the first vehicle at multiple future times, and the predicted trajectory information in the trajectory combination includes the second speed and second position of the second vehicle at multiple future times. The safety value of the trajectory combination is determined in the following manner: For each future moment, a safe distance is determined based on the first speed of the first vehicle at the future moment, the preset maximum deceleration of the first vehicle, the second speed of the second vehicle at the future moment, and the preset maximum deceleration of the second vehicle; a first predicted distance is determined based on the first position of the first vehicle at the future moment and the second position of the second vehicle at the future moment; and a safety cost is determined based on the safe distance and the first predicted distance. The safety value of the trajectory combination is determined based on the safety value of each future moment.

3. The method according to claim 2, characterized in that, The step of determining the safety cost value at the future time based on the safe distance and the first predicted distance includes: If the first predicted distance is less than the safe distance but greater than 0, then the safety cost at the future time is determined based on the ratio of the first predicted distance to the safe distance, wherein the larger the ratio, the smaller the safety cost at the future time.

4. The method according to claim 1, characterized in that, The target cost of the trajectory combination is determined based on the first vehicle cost of the trajectory combination; the candidate trajectory information in the trajectory combination includes the first acceleration, the first turning radius of the first vehicle at multiple future times, and the speed of the first vehicle at the latest future time. The first vehicle cost of the trajectory combination is determined in the following manner: For each future moment, the dynamic constraint cost of the future moment is determined based on the first acceleration and first turning radius of the first vehicle at the future moment, as well as the preset maximum acceleration and preset minimum turning radius of the first vehicle. The merging efficiency cost is determined based on the speed of the first vehicle at the latest future time and the speed limit of the target lane. The first vehicle cost of the trajectory combination is determined based on the inflow efficiency cost and the dynamic constraint cost at each of the future moments.

5. The method according to claim 4, characterized in that, The step of determining the dynamic constraint cost at the future moment based on the first vehicle's first acceleration and first turning radius at the future moment, and the first vehicle's preset maximum acceleration and preset minimum turning radius, includes: If the first acceleration is greater than the preset maximum acceleration, and the first turning radius is greater than or equal to the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference between the first acceleration and the preset maximum acceleration. If the first turning radius is less than the preset minimum turning radius, and the first acceleration is less than or equal to the preset maximum acceleration, then the dynamic constraint cost is determined based on the second difference between the preset minimum turning radius and the first turning radius. If the first acceleration is greater than the preset maximum acceleration and the first turning radius is less than the preset minimum turning radius, then the dynamic constraint cost is determined based on the first difference and the second difference.

6. The method according to claim 1, characterized in that, The target value of the trajectory combination is determined based on the second vehicle value of the trajectory combination; the predicted trajectory information in the trajectory combination includes the second position of the second vehicle at the plurality of future times. The second vehicle value of the trajectory combination is determined in the following manner: For each future time, a second predicted distance is determined based on the current position of the second vehicle and the second position of the second vehicle at the future time; The second vehicle cost of the trajectory combination is determined based on the probability of occurrence of the predicted driving condition corresponding to the predicted trajectory information in the trajectory combination and the second predicted distance corresponding to each future time.

7. The method according to claim 1, characterized in that, Determining the target cost of the trajectory combination includes: Based on the dynamic constraints of the first vehicle, the speed limit of the target lane, and the candidate trajectory information in the trajectory combination, the first vehicle cost of the trajectory combination is determined; The second vehicle's cost value is determined based on the current location of the second vehicle, the predicted trajectory information in the trajectory combination, and the probability of occurrence of the predicted driving conditions corresponding to the predicted trajectory information in the trajectory combination. The safety cost of the trajectory combination is determined based on the candidate trajectory information and the predicted trajectory information in the trajectory combination; The target value is determined based on the first vehicle value, the second vehicle value, and the safety value of the trajectory combination.

8. The method according to claim 1, characterized in that, The probability of occurrence of each of the M predicted driving conditions is obtained through a probability prediction model. The inputs of the probability prediction model include: the remaining distance of the first vehicle changing lanes to the target lane, the trajectory information of the first vehicle at the current time and in the historical time period, the trajectory information of the second vehicle at the current time and in the historical time period, and the lane curvature of the original lane in which the first vehicle was located before changing lanes to the target lane.

9. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method according to any one of claims 1-8.

10. A vehicle, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 1-8.

Images