Decision-making method and device, electronic equipment and storage medium

Abstract

The application discloses a decision method and device, electronic equipment and a storage medium. The method comprises the following steps: traversing obstacles and predicted trajectories of the obstacles to obtain obstacles appearing on a path planning of a vehicle; projecting the obstacles appearing on the path planning of the vehicle to an S-T graph, and calculating a lateral distance of the predicted trajectory of the obstacle close to the vehicle on the path planning of the vehicle; based on a pre-configured lateral risk estimation model, calculating a maximum passing speed of the vehicle safely passing at a target moment t according to a speed of the obstacle at the target moment t and the lateral distance of the obstacle and the vehicle; when calculating a penalty value of the vehicle at the target moment t, increasing the penalty value of the vehicle at the target moment t when the speed of the vehicle exceeds the maximum passing speed of the vehicle safely passing, and outputting a decision trajectory. According to the application, the maximum speed of the vehicle safely passing is calculated on one hand, and the lateral safety is ensured on the other hand.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a decision-making method, apparatus, electronic device, and storage medium. Background Technology

[0002] In autonomous driving, decision-making and planning algorithms typically employ a lateral and longitudinal separation architecture. Specifically, lateral decision-making and planning first plans a smooth path curve without speed information, while longitudinal decision-making and planning then plans a safe, comfortable, and efficient speed curve, i.e., a trajectory, based on this path.

[0003] Mainstream lateral decision-planning algorithms can guarantee spatial safety and collision-free operation between the output path and static obstacles. Furthermore, they also guarantee spatiotemporal safety and collision-free operation between the final output trajectory and all obstacles. Mainstream longitudinal decision-planning algorithms guarantee safety only if obstacles are within a certain lateral distance threshold of the path (typically set as half the vehicle's width plus a buffer value). If the lateral distance of an obstacle from the path exceeds this threshold, the longitudinal decision-planning algorithm will completely ignore that obstacle. While this theoretically guarantees spatiotemporal safety and collision-free operation, due to a series of lateral uncertainties between the vehicle and obstacles, such as intent, perception, and control errors, simply setting a fixed threshold cannot guarantee absolute lateral safety. Summary of the Invention

[0004] This application provides a decision-making method, apparatus, electronic device, and storage medium to ensure sufficient lateral security.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] In a first aspect, embodiments of this application provide a decision-making method, wherein the method includes:

[0007] Traverse the obstacles and their predicted trajectories to obtain the obstacles appearing on the vehicle's path planning;

[0008] The obstacles appearing in the vehicle path planning are projected onto the ST map, and the lateral distance from the predicted trajectory of the obstacle near the vehicle to the planned path of the vehicle is calculated. The ST map is used to represent the relationship between the planned path length and time. The obstacles near the vehicle are obtained after screening.

[0009] Based on a pre-configured lateral risk estimation model, the maximum speed at which the vehicle can safely pass through at the target time t is calculated according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0010] When calculating the penalty value of the vehicle at the target time t, the penalty value is increased if the vehicle's speed at the target time t exceeds the maximum passing speed for safe passage, and the decision trajectory is output.

[0011] In some embodiments, the method further includes:

[0012] Determine whether there is a situation where the speed of a vehicle exceeds the maximum safe passage speed of the vehicle at the target time t. If so, penalize the decision-making process of the vehicle that exceeds the maximum passage speed.

[0013] Otherwise, no penalty will be imposed on the decision-making process of the vehicle itself.

[0014] In some embodiments, the method further includes:

[0015] Based on the predicted trajectories of obstacles laterally adjacent to the vehicle, the lateral distance between the vehicle and the predicted trajectories of the obstacles laterally adjacent to the vehicle is obtained, and the lateral distance is added to the decision-making process. In some embodiments, the pre-configured lateral risk estimation model includes:

[0016] Lateral collision risk modeling is performed based on the TTC model or RSS model;

[0017] Based on the lateral collision risk modeling results, the collision time TTC = L / (V1+V1') is obtained;

[0018] Wherein, L is the lateral distance between the vehicle and the vehicle in front, V1 is the lateral speed of the vehicle in front, V1' is the lateral speed of the vehicle, and TTC is used to determine whether it is safe at the current lateral distance and lateral speed.

[0019] In some embodiments, the pre-configured horizontal risk estimation model further includes:

[0020] There is a fixed steering wheel vibration error during vehicle operation, and this error is transmitted to the vehicle chassis and manifests as a deflection angle of the front wheels.

[0021] θ = f * δ;

[0022] Wherein, f is the transmission ratio from the vehicle steering wheel to the front wheel steering, and δ is the vehicle steering wheel angle;

[0023] The additional lateral velocity is obtained based on the deflection angle of the vehicle's front wheels.

[0024] V1=v*θ

[0025] Wherein, V1 is directly proportional to the vehicle's forward speed v, and the proportionality coefficient is the deflection angle θ of the vehicle's front wheels.

[0026] In some embodiments, the pre-configured horizontal risk estimation model further includes:

[0027] The lateral distance is determined by the distance between the obstacle's outline and the vehicle's body.

[0028] A time threshold is set based on the obstacle type. The time threshold is the shortest time calculated based on the current lateral speed and lateral distance using a TTC or RSS model.

[0029] In some embodiments, the pre-configured horizontal risk estimation model further includes:

[0030] The maximum safe longitudinal speed of the vehicle is calculated based on the longitudinal velocity v of the obstacle, the lateral distance L, and the time threshold TTC.

[0031]

[0032] The TTC time threshold and the front wheel angle of the other vehicle are determined according to the type of obstacle, and the deflection angle of the front wheel of the vehicle is an adjustable parameter.

[0033] Secondly, embodiments of this application also provide a decision-making device, wherein the device includes:

[0034] The traversal processing module is used to traverse obstacles and their predicted trajectories to obtain the obstacles appearing in the vehicle's path planning.

[0035] The projection calculation module is used to project the path planning image of the vehicle onto the ST map and calculate the lateral distance from the predicted trajectory of the obstacle near the vehicle to the planned path of the vehicle. The ST map is used to represent the relationship between the length of the planned path and time. The obstacle near the vehicle is obtained after screening.

[0036] The speed calculation module is used to calculate the maximum speed at which the vehicle can safely pass through at the target time t, based on a pre-configured lateral risk estimation model, according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0037] The decision module is used to increase the penalty value of the vehicle at the target time t when calculating the penalty value of the vehicle, if the vehicle's speed exceeds the maximum safe passage speed of the vehicle at the target time t, and output the decision trajectory.

[0038] Thirdly, embodiments of this application also provide an electronic device, including: a processor; and a memory arranged to store computer-executable instructions, which, when executed, cause the processor to perform the above-described method.

[0039] Fourthly, embodiments of this application also provide a computer-readable storage medium that stores one or more programs, which, when executed by an electronic device including multiple applications, cause the electronic device to perform the above-described method.

[0040] The at least one technical solution adopted in this application embodiment can achieve the following beneficial effects: by traversing obstacles and their predicted trajectories, obstacles appearing on the vehicle's path planning are obtained. Then, the obstacles appearing on the vehicle's path planning are projected onto the ST graph, and the lateral distance from the predicted trajectories of obstacles adjacent to the vehicle to the vehicle's planned path is calculated. Based on a pre-configured lateral risk estimation model, according to the speed of the obstacle at target time t and the lateral distance between the obstacle and the vehicle, the maximum safe passage speed of the vehicle at target time t is calculated. Finally, when calculating the penalty value of the vehicle at target time t, a penalty value is added for the vehicle's speed exceeding the maximum safe passage speed at target time t, and the decision trajectory is output. Through the above method, a decision-making process with spatiotemporal joint characteristics is realized. Simultaneously, the adopted lateral risk estimation model includes uniformly modeling lateral uncertainty as lateral speed, and the TTC time threshold and front wheel steering angle can be flexibly set according to the obstacle type. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0042] Figure 1 This is a schematic diagram of the lateral risk estimation model in the embodiments of this application;

[0043] Figure 2 This is a flowchart illustrating the decision-making method in an embodiment of this application;

[0044] Figure 3 This is a schematic diagram of the decision-making device in the embodiments of this application;

[0045] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0047] The processing flow of mainstream vertical decision programming algorithms can be divided into:

[0048] (1) Generate ST projection diagram.

[0049] (2) Vertical decision-making: Generate coarse trajectory based on ST projection map decision-making.

[0050] (3) Vertical planning: The final smoothed trajectory is generated based on the ST projection map and the coarse trajectory.

[0051] First, the process of generating the ST projection map can be simplified to: mapping the obstacle outlines at the points on the predicted obstacle trajectory (for static obstacles, the trajectory can be considered as a fixed value at the current position).

[0052] Secondly, decision-making based on the ST projection graph: using sampling or search algorithms to obtain a coarse trajectory on the ST graph that is safe and has the minimum penalty value COST.

[0053] Finally, the final trajectory is at least one safe, comfortable, and efficient trajectory generated by optimizing the coarse trajectory and the convex space it resides in.

[0054] However, the longitudinal decision-making scheme can only guarantee no collisions with obstacles appearing in the ST projection map, neglecting the potential lateral collision risk of obstacles outside the ST map. This is because, to appear on the ST projection map, the lateral distance between the obstacle's outline and the path curve must be within a threshold (half the vehicle width + a buffer value). This buffer value represents the minimum permissible lateral distance between the obstacle's outline and the vehicle's outline, typically a small value, such as 0.2m. Analysis shows that if the lateral distance is greater than this buffer value, the longitudinal decision-making scheme will ignore the obstacle. Since a lateral distance greater than this buffer value theoretically means no lateral collision, it does not appear in the ST map. However, in practice, it ignores the collision risk caused by the uncertainty of the lateral movement between the vehicle and the obstacle.

[0055] Taking driving common sense as an example, if there are stationary cars on both sides of the road ahead, the minimum distance a car needs to pass these stationary cars is only slightly greater than 0.2m (e.g., 0.21m). According to existing longitudinal decision-making algorithms, the lateral distance of the stationary cars from the path is greater than the buffer value, preventing them from entering the ST graph. Therefore, the longitudinal decision-making process completely ignores the stationary cars, potentially resulting in a high-speed curve. However, human drivers do the opposite; when passing stationary cars, they reduce their speed to a reasonable range and pass at a low speed. If the buffer value is artificially increased in the existing algorithm, forcing the stationary cars into the ST graph, the longitudinal decision-making process will plan a speed curve that comes to a complete stop before the stationary cars, making passage impossible. The reason for this difference is that existing algorithms ignore the risk of lateral collisions with obstacles.

[0056] Thus, due to errors in the autonomous vehicle's control and the stationary vehicle's perception, a certain lateral risk exists. Furthermore, as the autonomous vehicle's speed increases, the lateral wind direction also strengthens, preventing the autonomous vehicle from traveling at excessively high speeds. Only by controlling the autonomous vehicle's speed within a certain range (the value of which is related to the lateral distance) can the lateral risk be reduced to below the psychological expectations of a human driver, allowing for safe passage. In addition, existing longitudinal decision-making and planning algorithms cannot adequately handle the risk of lateral collisions with dynamic obstacles.

[0057] To address the aforementioned shortcomings, this application proposes an improved longitudinal decision-making algorithm. This algorithm fully considers all nearby obstacles that have not yet entered the ST map, calculates the maximum passage speed based on a lateral risk estimation model, and penalizes vehicle decisions that exceed the maximum passage speed in the longitudinal decision-making process, thereby achieving safe passage. Furthermore, the longitudinal decision-making algorithm also provides a mathematical model for lateral risk estimation, which can calculate the maximum safe passage speed of the vehicle using the known speeds of obstacles and their lateral distances.

[0058] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0059] This application provides a decision-making method, such as... Figure 2 As shown, a flowchart of the decision-making method in an embodiment of this application is provided. The method includes at least the following steps S210 to S240:

[0060] Step S210: Traverse the obstacles and their predicted trajectories to obtain the obstacles appearing on the vehicle path planning.

[0061] Obstacle information can be acquired through the vehicle's localization and perception module and transmitted to downstream modules. The predicted trajectory of the obstacle is determined based on its distance from the vehicle and its speed. It can be understood that obstacles include, but are not limited to, vehicles in front, stationary obstacles, and other slow-moving road users.

[0062] By traversing the aforementioned obstacles and their trajectory information, the obstacles appearing in the vehicle's path planning can be obtained. It can be understood that these obstacles include all possible obstacles.

[0063] Step S220: Project obstacles appearing on the vehicle path planning onto the ST map, and calculate the lateral distance from the predicted trajectory of the obstacle near the vehicle to the planned path of the vehicle. The ST map is used to characterize the relationship between the planned path length and time. The obstacles near the vehicle are obtained through screening.

[0064] Following the methods described in related technologies, obstacles appearing in the vehicle's path planning are projected onto the ST map. Based on the projection results from the ST map, the lateral distances from the predicted trajectories of other nearby obstacles to the planned path of the vehicle are calculated. It is important to note that these nearby obstacles are pre-selected using rules, which reduces the computational load compared to not selecting obstacles.

[0065] In related technologies, the longitudinal decision-making process only considers obstacles appearing in the ST diagram. This is equivalent to treating obstacles as two-dimensional objects in the ST space of the vehicle path, where S represents path length and T represents time, thus ignoring the lateral distance dimension of the obstacles.

[0066] The above process not only needs to consider obstacles in the ST diagram, but also obstacles that are laterally close. By directly incorporating lateral distance into the decision-making process, the lateral distance between obstacles and the path is restored, thus possessing the three-dimensional spatiotemporal joint decision-making characteristics under the vehicle path.

[0067] Step S230: Based on the pre-configured lateral risk estimation model, the maximum speed at which the vehicle can safely pass through at the target time t is calculated according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0068] like Figure 1 As shown, the lateral distance between the vehicle and the vehicle in front is L, and the lateral velocity of the vehicle in front is V. L The lateral speed of the vehicle is V. L Based on a pre-configured lateral risk estimation model, the maximum safe passage speed of the vehicle is calculated according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0069] It is understandable that the target time t can be any time, determined according to actual needs.

[0070] Step S240: When calculating the penalty value of the vehicle at the target time t, increase the penalty value for the vehicle exceeding the maximum safe passage speed at the target time t, and output the decision trajectory.

[0071] When calculating the COST penalty value for the vehicle at the target time t, the vehicle's speed at the target time t should be increased by more than V. t The COST penalty value is used as the lateral risk COST penalty value and added to the vehicle's decision-making process.

[0072] Because obstacle prediction trajectories are taken into account, calculations based on lateral distance and travel speed can be accurate to the actual moment when the obstacle and the vehicle are in contact. This allows for decisions on decelerating at reasonable times and locations to avoid lateral risks, thus exhibiting certain spatiotemporal joint planning characteristics.

[0073] Using the above method, decisions with lateral collision risks will be penalized to a certain extent throughout the entire vertical decision-making process, thereby outputting a safe and risk-free decision trajectory.

[0074] By fully considering all nearby obstacles that have not entered the ST diagram, the maximum safe passage speed of the vehicle can be calculated using the known speed and lateral distance of the obstacles. Then, the maximum passage speed is calculated based on the lateral risk estimation model. In the longitudinal decision-making process, the vehicle's decision to exceed the maximum passage speed is penalized, thereby achieving safe passage.

[0075] Using the above method, lateral uncertainty is uniformly modeled as lateral velocity. Furthermore, the TTC time threshold and front wheel steering angle can be flexibly set according to the type of obstacle.

[0076] By treating obstacles as two-dimensional objects in the vehicle's path ST space (S: path length, T: time), the lateral distance dimension of obstacles is ignored. The optimized longitudinal decision not only needs to consider obstacles in the ST graph, but also obstacles that are laterally adjacent, directly incorporating lateral distance into the decision-making process. Therefore, the three-dimensional attributes of obstacles (distance between the obstacle and the vehicle, lateral velocity of the obstacle, and lateral distance between the obstacle and the vehicle's planned path) are restored, giving it three-dimensional spatiotemporal joint decision-making characteristics under the vehicle's path.

[0077] Traverse all obstacles and their predicted trajectories, projecting obstacles appearing on the vehicle's path onto the ST graph, and calculate the lateral distances from the predicted trajectories of other nearby obstacles to the vehicle's path. Based on the obstacle's velocity at a given time t and its lateral distance from the vehicle, calculate the vehicle's maximum passing speed V. tWhen calculating the vehicle's COST at time t, the vehicle's speed at time t is increased by more than V. t The cost of this cost, as a lateral risk cost, is incorporated into the vehicle's decision-making process.

[0078] It is understandable that, apart from the aforementioned penalty process, other decision-making processes are generally consistent with the mainstream ST-based decision-making processes.

[0079] By incorporating obstacle prediction trajectories, the calculation of lateral distance and speed can be accurate to the precise moment of actual interaction between the obstacle and the vehicle. This allows for decisions to decelerate and avoid lateral risks at appropriate spatiotemporal locations, exhibiting certain spatiotemporal joint planning characteristics. Throughout the longitudinal decision-making process, decisions with lateral collision risks are penalized, resulting in a safe and risk-free decision trajectory.

[0080] In one embodiment of this application, the method further includes: determining whether there is a situation where the speed of the vehicle at the target time t exceeds the maximum passing speed for the vehicle to pass safely; if so, penalizing the decision-making process of the vehicle that exceeds the maximum passing speed; otherwise, not penalizing the decision-making process of the vehicle.

[0081] It's understandable that autonomous driving decision-making penalty costs refer to a mechanism introduced in an autonomous driving system to optimize the decision-making process. This involves imposing additional costs to penalize behaviors or decisions that don't meet expectations. Examples of penalties include obstacle avoidance penalties, traffic rule violations, path smoothness penalties, and penalties for distance from reference lines. The design and adjustment of these penalty costs need to be optimized based on specific driving environments and requirements to ensure the safe and efficient operation of the autonomous driving system.

[0082] When determining the penalty value, it is necessary to determine whether there is a situation where the vehicle's speed at the target time t exceeds the maximum safe passage speed. If so, the decision-making process of the vehicle exceeding the maximum passage speed is penalized. Throughout the entire longitudinal decision-making process, decisions with lateral collision risk will be penalized to optimize the decision, thereby outputting a safe and risk-free decision trajectory. Similarly, if not, no penalty is imposed on the vehicle's decision-making process.

[0083] In one embodiment of this application, the method further includes obtaining the lateral distance between the vehicle and the predicted trajectory of the obstacle adjacent to the vehicle in the lateral distance based on the predicted trajectory of the obstacle adjacent to the vehicle in the lateral distance, and incorporating the lateral distance into the decision-making process. By fully considering all adjacent obstacles that have not entered the ST map, the maximum passage speed is calculated based on the lateral risk estimation model, and the vehicle decision exceeding the maximum passage speed is penalized in the longitudinal decision, thereby achieving safe passage.

[0084] In one embodiment of this application, the pre-configured lateral risk estimation model includes: performing lateral collision risk modeling based on a TTC model or an RSS model; obtaining the collision time TTC = L / (V1+V1') based on the lateral collision risk modeling result; wherein, L is the lateral distance between the vehicle and the vehicle in front, V1 is the lateral speed of the vehicle in front, V1' is the lateral speed of the vehicle, and TTC is used to determine whether it is safe under the current lateral distance and lateral speed.

[0085] TTC (Time To Collision) is a key metric used to measure the time required for a vehicle to travel normally before a collision occurs, and it plays a crucial role in improving vehicle safety.

[0086] RSS (Responsibility Sensitive Safety) aims to transform human concepts of safe driving and the division of accident responsibility into mathematical models and reference parameters for decision-making and control.

[0087] Lateral risks associated with stationary obstacles stem from perception errors and potential starting intentions. However, lateral risks associated with dynamic obstacles are more complex, encompassing not only perception errors but also their complex movement intentions (whether to change lanes, cut in, or turn), speed, size, type, control errors, and so on.

[0088] Generally, the greater the speed of an obstacle, the greater the perceived error. With a fixed steering wheel control error, a higher obstacle speed means even small steering wheel movements will result in greater lateral speed fluctuations for the vehicle. Similarly, if the obstacle is an electric vehicle, bicycle, or pedestrian, its complex motion can be reflected in its agile steering ability. Furthermore, if the obstacle is a large vehicle, its significant size and mass necessitate maintaining a greater lateral safety distance. Therefore, a mathematical model is needed to reasonably represent this obstacle.

[0089] Preferably, in the embodiments of this application, the lateral risk of obstacles is uniformly modeled as a TTC or RSS mathematical relationship between lateral velocity and lateral distance.

[0090] The TTC and RSS models are currently mature collision safety models, and the same models are used in the embodiments of this application for lateral collision risk modeling. Specifically, as follows... Figure 1 As shown, the lateral distance between the vehicle and the vehicle in front is L, and the lateral velocity of the other vehicle is V. L The lateral speed of the vehicle is V. L '.

[0091] TTC=L / (V1+V1')(Formula 1)

[0092] If the Time Tolerance (TTC) is less than the set time threshold, a lateral collision risk is identified. The Responsibility Safety Sensitivity Model (RSS), similar to TTC, is used to determine whether the vehicle is safe given the current lateral distance and lateral speed.

[0093] In one embodiment of this application, the pre-configured lateral risk estimation model further includes: a fixed steering wheel vibration error exists during vehicle operation, and the error is transmitted to the vehicle chassis as the deflection angle of the front wheels.

[0094] θ = f * δ;

[0095] Wherein, f is the transmission ratio from the vehicle steering wheel to the front wheel steering, and δ is the vehicle steering wheel angle;

[0096] The additional lateral velocity is obtained based on the deflection angle of the vehicle's front wheels.

[0097] V L =v*θ

[0098] Wherein, V1 is directly proportional to the vehicle's forward speed v, and the proportionality coefficient is the deflection angle θ of the vehicle's front wheels.

[0099] As demonstrated by the TTC and RSS models, modeling the lateral velocity of the vehicle and obstacles is crucial. However, vehicles traveling normally in a straight line do not generate lateral velocity. Therefore, embodiments of this application propose modeling the potential lateral velocity using the vehicle's forward speed. Considering a fixed steering wheel vibration error that may exist during vehicle operation, this error is transmitted to the vehicle chassis and manifests as the front wheel deflection angle, as shown in the following formula:

[0100] θ=f*δ (Formula 2)

[0101] Where f is the transmission ratio from the vehicle's steering wheel to the front wheels, which is determined by mechanical characteristics and generally fluctuates around a fixed value, which can be assumed to be this fixed value. δ is the vehicle's steering wheel angle.

[0102] Due to the front wheel deflection angle, additional lateral velocity is generated, as shown in V in the diagram above. L As shown,

[0103] V L =v*θ (Formula 3)

[0104] That is, the lateral velocity is directly proportional to the vehicle's forward velocity, and the proportionality coefficient is the front wheel deflection angle θ.

[0105] It is important to note that the front wheel deflection angle needs to be adjusted according to the obstacle's maneuverability and the intended action:

[0106] For vehicles with agile steering, such as bicycles and electric vehicles, the steering wheel vibration error is greater and their steering intention is more unpredictable. These can be uniformly modeled as a larger front wheel rotation angle, resulting in a greater lateral velocity. Therefore, a larger lateral safety distance or a reduced vehicle speed is required when passing such obstacles. This model can calculate reasonable lateral safety distances or passing speeds for different types of obstacles, exhibiting human-like behavior.

[0107] In one embodiment of this application, the pre-configured lateral risk estimation model further includes: using the distance between the obstacle outline and the vehicle body as the lateral distance; and setting a time threshold according to the obstacle type, wherein the time threshold is the shortest time calculated based on the current lateral speed and lateral distance using a TTC or RSS model.

[0108] In practical implementation, the lateral distance is the distance between the obstacle's outline and the vehicle's body. Generally, perception errors can be considered, and this absolute distance can be reduced to a certain extent.

[0109] Time threshold: The collision time is the shortest time calculated based on the current lateral velocity and lateral distance using the TTC or RSS model. If this time is greater than the set time threshold, there is no risk of collision; otherwise, there is a risk.

[0110] It's important to note that time thresholds should be set according to obstacle type. For obstacles such as large vehicles, pedestrians, and two-wheeled vehicles, human driving habits tend to maintain a greater safe distance or appropriately reduce the speed when passing them. This characteristic can be achieved by setting a larger time threshold for these types of obstacles.

[0111] In one embodiment of this application, the pre-configured lateral risk estimation model further includes: calculating the maximum safe longitudinal speed of the vehicle based on the longitudinal velocity v of the obstacle, the lateral distance L, and the time threshold TTC.

[0112]

[0113] The TTC time threshold and the front wheel angle of the other vehicle are determined according to the type of obstacle, and the deflection angle of the front wheel of the vehicle is an adjustable parameter.

[0114] Among them, v 障碍物 θ 障碍物 Lateral velocity and V 障碍物 The forward speed of the vehicle is directly proportional to θ. 障碍物 The proportionality factor is the front wheel deflection angle θ.

[0115] Among them, f 自 This is the transmission ratio from the vehicle's steering wheel to the front wheels, which can be assumed to be a fixed value, δ. 自This refers to the steering wheel angle of the vehicle.

[0116] Given the longitudinal speed, lateral distance, and time threshold of the obstacle and other vehicles, the maximum safe speed of the vehicle can be calculated using formulas (1, 2, and 3), which is the longitudinal speed. The TTC time threshold and the front wheel angle of the other vehicle vary depending on the type of obstacle, and the steering wheel angle of the vehicle is also an adjustable parameter.

[0117] This application embodiment also provides a decision-making device 300, such as Figure 3 As shown, a schematic diagram of the decision-making device in an embodiment of this application is provided. The decision-making device 300 includes at least: a traversal processing module 310, a projection calculation module 320, a speed calculation module 330, and a decision module 340, wherein:

[0118] In one embodiment of this application, the traversal processing module 310 is specifically used to: traverse obstacles and the predicted trajectories of the obstacles to obtain obstacles appearing on the vehicle path planning.

[0119] Obstacle information can be acquired through the vehicle's localization and perception module and transmitted to downstream modules. The predicted trajectory of the obstacle is determined based on its distance from the vehicle and its speed. It can be understood that obstacles include, but are not limited to, vehicles in front, stationary obstacles, and other slow-moving road users.

[0120] By traversing the aforementioned obstacles and their trajectory information, the obstacles appearing in the vehicle's path planning can be obtained. It can be understood that these obstacles include all possible obstacles.

[0121] In one embodiment of this application, the projection calculation module 320 is specifically used to: project obstacles appearing on the vehicle path planning onto the ST map, and calculate the lateral distance from the predicted trajectory of the obstacle near the vehicle to the planned path of the vehicle. The ST map is used to characterize the relationship between the planned path length and time, and the obstacle near the vehicle is obtained through screening.

[0122] Following the methods described in related technologies, obstacles appearing in the vehicle's path planning are projected onto the ST map. Based on the projection results from the ST map, the lateral distances from the predicted trajectories of other nearby obstacles to the planned path of the vehicle are calculated. It is important to note that these nearby obstacles are pre-selected using rules, which reduces the computational load compared to not selecting obstacles.

[0123] In related technologies, the longitudinal decision-making process only considers obstacles appearing in the ST diagram. This is equivalent to treating obstacles as two-dimensional objects in the ST space of the vehicle path, where S represents path length and T represents time, thus ignoring the lateral distance dimension of the obstacles.

[0124] The above process not only needs to consider obstacles in the ST diagram, but also obstacles that are laterally close. By directly incorporating lateral distance into the decision-making process, the lateral distance between obstacles and the path is restored, thus possessing the three-dimensional spatiotemporal joint decision-making characteristics under the vehicle path.

[0125] In one embodiment of this application, the speed calculation module 330 is specifically used to: calculate the maximum speed at which the vehicle can safely pass through at the target time t based on a pre-configured lateral risk estimation model, according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0126] like Figure 1 As shown, the lateral distance between the vehicle and the vehicle in front is L, and the lateral velocity of the vehicle in front is V. L The lateral speed of the vehicle is V. L Based on a pre-configured lateral risk estimation model, and according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle, the decision penalty module 340 is specifically used to: when calculating the penalty value of the vehicle at the target time t, increase the penalty value for the vehicle exceeding the maximum safe passage speed at the target time t, and output the decision trajectory.

[0127] When calculating the COST penalty value for the vehicle at the target time t, the vehicle's speed at the target time t should be increased by more than V. t The COST penalty value is used as the lateral risk COST penalty value and added to the vehicle's decision-making process.

[0128] Because obstacle prediction trajectories are taken into account, calculations based on lateral distance and travel speed can be accurate to the actual moment when the obstacle and the vehicle are in contact. This allows for decisions on decelerating at reasonable times and locations to avoid lateral risks, thus exhibiting certain spatiotemporal joint planning characteristics.

[0129] In one embodiment of this application, a determination module is further included, used for...

[0130] Determine whether there is a situation where the speed of a vehicle exceeds the maximum safe passage speed of the vehicle at the target time t. If so, penalize the decision-making process of the vehicle that exceeds the maximum passage speed.

[0131] Otherwise, no penalty will be imposed on the decision-making process of the vehicle itself.

[0132] In one embodiment of this application, a lateral distance module is also included for...

[0133] Based on the predicted trajectories of obstacles that are laterally close to the vehicle, the lateral distance between the vehicle and the predicted trajectories of the obstacles that are laterally close to the vehicle is obtained, and the lateral distance is added to the decision-making process.

[0134] In one embodiment of this application, the speed calculation module 330 is further used for

[0135] Lateral collision risk modeling is performed based on the TTC model or RSS model;

[0136] Based on the lateral collision risk modeling results, the collision time TTC = L / (V1+V1') is obtained;

[0137] Wherein, L is the lateral distance between the vehicle and the vehicle in front, V1 is the lateral speed of the vehicle in front, V1' is the lateral speed of the vehicle, and TTC is used to determine whether it is safe at the current lateral distance and lateral speed.

[0138] In one embodiment of this application, the speed calculation module 330 is further used for

[0139] There is a fixed steering wheel vibration error during vehicle operation, and this error is transmitted to the vehicle chassis and manifests as a deflection angle of the front wheels.

[0140] θ = f * δ;

[0141] Wherein, f is the transmission ratio from the vehicle steering wheel to the front wheel steering, and δ is the vehicle steering wheel angle;

[0142] The additional lateral velocity is obtained based on the deflection angle of the vehicle's front wheels.

[0143] V1=v*θ

[0144] Wherein, V1 is directly proportional to the vehicle's forward speed v, and the proportionality coefficient is the deflection angle θ of the vehicle's front wheels.

[0145] In one embodiment of this application, the pre-configured lateral risk estimation model further includes:

[0146] The lateral distance is determined by the distance between the obstacle's outline and the vehicle's body.

[0147] A time threshold is set based on the obstacle type, which is used to calculate the shortest time based on the current lateral velocity and lateral distance using a TTC or RSS model.

[0148] In one embodiment of this application, the speed calculation module 330 is further configured to calculate the maximum safe longitudinal speed of the vehicle based on the longitudinal speed v of the obstacle, the lateral distance L, and the time threshold TTC.

[0149]

[0150] The TTC time threshold and the front wheel angle of the other vehicle are determined according to the type of obstacle, and the deflection angle of the front wheel of the vehicle is an adjustable parameter.

[0151] It is understood that the above-described decision-making device can implement each step of the decision-making method provided in the foregoing embodiments. The relevant explanations of the decision-making method are applicable to the decision-making device and will not be repeated here.

[0152] Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Please refer to it. Figure 4 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for other business operations.

[0153] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0154] Memory is used to store programs. Specifically, programs may include program code, which includes computer operation instructions. Memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0155] The processor reads the corresponding computer program from non-volatile memory into main memory and then runs it, forming a decision-making mechanism at the logical level. The processor executes the program stored in memory and specifically performs the following operations:

[0156] Traverse the obstacles and their predicted trajectories to obtain the obstacles appearing on the vehicle's path planning;

[0157] Traverse the obstacles and their predicted trajectories to obtain the obstacles appearing on the vehicle's path planning;

[0158] Obstacles appearing on the vehicle's path planning are projected onto the ST graph, and the lateral distance from the predicted trajectory of the obstacle adjacent to the vehicle to the planned path of the vehicle is calculated. The ST graph is used to characterize the relationship between the planned path length and time. The obstacles adjacent to the vehicle are obtained after screening.

[0159] Based on a pre-configured lateral risk estimation model, the maximum speed at which the vehicle can safely pass through at the target time t is calculated according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0160] When calculating the penalty value of the vehicle at the target time t, the penalty value is increased if the vehicle's speed at the target time t exceeds the maximum passing speed for safe passage, and the decision trajectory is output.

[0161] The above is as stated in this application. Figure 2 The method executed by the decision-making device disclosed in the illustrated embodiments can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0162] The electronic device can also perform Figure 2 The method for the decision-making device to execute, and to realize the decision-making device in Figure 1The functions of the embodiments shown are not described in detail here.

[0163] This application also proposes a computer-readable storage medium that stores one or more programs, the programs including instructions that, when executed by an electronic device including multiple applications, enable the electronic device to perform... Figure 2 The method executed by the decision-making device in the illustrated embodiment is specifically used to perform:

[0164] Traverse the obstacles and their predicted trajectories to obtain the obstacles appearing on the vehicle's path planning;

[0165] Obstacles appearing on the vehicle's path planning are projected onto the ST graph, and the lateral distance from the predicted trajectory of the obstacle adjacent to the vehicle to the planned path of the vehicle is calculated. The ST graph is used to characterize the relationship between the planned path length and time. The obstacles adjacent to the vehicle are obtained after screening.

[0166] Based on a pre-configured lateral risk estimation model, the maximum speed at which the vehicle can safely pass through at the target time t is calculated according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle.

[0167] When calculating the penalty value for the vehicle at the target time t, the penalty value is increased for the vehicle's speed exceeding the maximum safe passage speed at the target time t, and a decision trajectory is output. Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0168] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0169] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0170] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0171] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0172] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0173] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0174] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0175] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0176] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A decision-making method, wherein, The method includes: Traverse the obstacles and their predicted trajectories to obtain the obstacles appearing on the vehicle's path planning; Obstacles appearing on the vehicle's path planning are projected onto the ST graph, and the lateral distance from the predicted trajectory of the obstacle adjacent to the vehicle to the planned path of the vehicle is calculated. The ST graph is used to characterize the relationship between the planned path length and time. The obstacles adjacent to the vehicle are obtained after screening. Based on a pre-configured lateral risk estimation model, the maximum speed at which the vehicle can safely pass through at the target time t is calculated according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle. The pre-configured horizontal risk estimation model includes: Lateral collision risk modeling is performed based on the TTC model or RSS model; Based on the lateral collision risk modeling results, the collision time TTC = L / (V1+V1') is obtained; Wherein, L is the lateral distance between the vehicle and the vehicle in front, V1 is the lateral speed of the vehicle in front, V1' is the lateral speed of the vehicle, and TTC is used to determine whether it is safe at the current lateral distance and lateral speed; The pre-configured horizontal risk estimation model also includes: There is a fixed steering wheel vibration error during vehicle operation, and this error is transmitted to the vehicle chassis and manifests as a deflection angle of the front wheels. θ = f * δ; Wherein, f is the transmission ratio from the vehicle steering wheel to the front wheel steering, and δ is the vehicle steering wheel angle; The additional lateral velocity is obtained based on the deflection angle of the vehicle's front wheels. V1=v*θ Wherein, V1 is directly proportional to the vehicle's forward speed v, and the proportionality coefficient is the deflection angle θ of the vehicle's front wheels; When calculating the penalty value of the vehicle at the target time t, the penalty value is increased if the vehicle's speed at the target time t exceeds the maximum passing speed for safe passage, and the decision trajectory is output.

2. The method as described in claim 1, wherein, The method further includes: Determine whether there is a situation where the speed of a vehicle exceeds the maximum safe passage speed of the vehicle at the target time t. If so, penalize the decision-making process of the vehicle that exceeds the maximum passage speed. Otherwise, no penalty will be imposed on the decision-making process of the vehicle itself.

3. The method as described in claim 2, wherein, The method further includes: Based on the predicted trajectories of obstacles that are laterally close to the vehicle, the lateral distance between the vehicle and the predicted trajectories of the obstacles that are laterally close to the vehicle is obtained, and the lateral distance is added to the decision-making process.

4. The method as described in claim 1, wherein, The pre-configured horizontal risk estimation model also includes: The lateral distance is determined by the distance between the obstacle's outline and the vehicle's body. A time threshold is set based on the obstacle type. The time threshold is the shortest time calculated based on the current lateral speed and lateral distance using a TTC or RSS model.

5. The method as described in claim 4, wherein, The pre-configured horizontal risk estimation model also includes: The maximum safe longitudinal speed of the vehicle is calculated based on the longitudinal velocity v of the obstacle, the lateral distance L, and the time threshold TTC. The TTC time threshold and the front wheel angle of the other vehicle are determined according to the type of obstacle, and the deflection angle of the front wheel of the vehicle is an adjustable parameter.

6. A decision-making device, wherein, The device includes: The traversal processing module is used to traverse obstacles and their predicted trajectories to obtain the obstacles appearing in the vehicle's path planning. The projection calculation module is used to project obstacles appearing on the vehicle path planning onto the ST map, and calculate the lateral distance from the predicted trajectory of the obstacle adjacent to the vehicle to the planned path of the vehicle. The ST map is used to characterize the relationship between the planned path length and time. The obstacle adjacent to the vehicle is obtained after screening. The speed calculation module is used to calculate the maximum speed at which the vehicle can safely pass through at the target time t, based on a pre-configured lateral risk estimation model, according to the speed of the obstacle at the target time t and the lateral distance between the obstacle and the vehicle. The pre-configured horizontal risk estimation model includes: Lateral collision risk modeling is performed based on the TTC model or RSS model; Based on the lateral collision risk modeling results, the collision time TTC = L / (V1+V1') is obtained; Wherein, L is the lateral distance between the vehicle and the vehicle in front, V1 is the lateral speed of the vehicle in front, V1' is the lateral speed of the vehicle, and TTC is used to determine whether it is safe at the current lateral distance and lateral speed; The pre-configured horizontal risk estimation model also includes: There is a fixed steering wheel vibration error during vehicle operation, and this error is transmitted to the vehicle chassis and manifests as a deflection angle of the front wheels. θ = f * δ; Wherein, f is the transmission ratio from the vehicle steering wheel to the front wheel steering, and δ is the vehicle steering wheel angle; The additional lateral velocity is obtained based on the deflection angle of the vehicle's front wheels. V1=v*θ Wherein, V1 is directly proportional to the vehicle's forward speed v, and the proportionality coefficient is the deflection angle θ of the vehicle's front wheels; The decision module is used to increase the penalty value of the vehicle at the target time t when calculating the penalty value of the vehicle, if the vehicle's speed exceeds the maximum safe passage speed of the vehicle at the target time t, and output the decision trajectory.

7. An electronic device, comprising: processor; as well as A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1 to 5.

8. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of applications, cause the electronic device to perform the method of any one of claims 1 to 5.

Images