Intelligent driving method and related apparatus

Abstract

An intelligent driving method and a related apparatus, applicable to the field of intelligent driving. The method comprises: determining at least two pieces of second information on the basis of first information, wherein the first information comprises information of at least one predicted trajectory of each first obstacle around a vehicle, a second obstacle is present among the at least one first obstacle, the first information comprises information of at least two predicted trajectories of the second obstacle, each piece of second information indicates the position of a predicted trajectory, and different second information indicates the positions of different predicted trajectories; and on the basis of each piece of second information and a strategy set corresponding to each piece of second information, determining a planned trajectory of the vehicle and / or decision information of the vehicle, wherein the strategy set comprises feasible driving strategies of the vehicle in a transverse direction and / or a longitudinal direction. When various possible predicted trajectories of an obstacle are fully considered, the trajectory planning process of the vehicle is prevented from being too conservative, so that the safety of a planned trajectory is improved, and the reduction of traffic efficiency is avoided.

Description

A smart driving method and related equipment

[0001] This application claims priority to Chinese Patent Application No. 202410696732.2, filed with the State Intellectual Property Office of China on May 30, 2024, entitled "An Intelligent Driving Method and Related Equipment", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to intelligent driving technology, and more particularly to an intelligent driving method and related equipment. Background Technology

[0003] In the field of intelligent driving, vehicle trajectory planning can be performed based on predicted trajectories of obstacles around the vehicle. For the same obstacle, there may be at least two predicted trajectories. In related technologies, generally only the predicted trajectory with the highest probability for each obstacle is considered, and then the vehicle trajectory is planned based on this highest-probability predicted trajectory. However, the aforementioned method does not consider low-probability events when planning trajectories, resulting in a higher risk of the planned trajectory. Summary of the Invention

[0004] This application provides an intelligent driving method and related equipment, which avoids the vehicle trajectory planning process from being too conservative, while fully considering multiple possible predicted trajectories of obstacles. This not only improves the safety of the planned trajectory but also minimizes the reduction in traffic efficiency.

[0005] This application provides the following technical solution:

[0006] In a first aspect, this application provides an intelligent driving method applicable to the field of intelligent driving. In this method, a first device acquires first information corresponding to at least one first obstacle around a vehicle (hereinafter referred to as "the first vehicle" for convenience). The first information includes information on at least one predicted trajectory of each of the aforementioned at least one first obstacle. At least one second obstacle exists among the aforementioned at least one first obstacle. The first information includes information on at least two predicted trajectories of each second obstacle. After acquiring the information on at least two predicted trajectories included in the first information, the first device can determine at least two pieces of second information based on the first information. Each piece of second information can indicate the position of at least one predicted trajectory. Different pieces of second information indicate the positions of different predicted trajectories among all predicted trajectories corresponding to the first information.

[0007] After determining at least two pieces of second information, the first device can acquire the strategy set corresponding to each piece of second information, and then determine the planned trajectory and / or decision information of the first vehicle based on each piece of second information and the strategy set corresponding to each piece of second information. The strategy set corresponding to each piece of second information includes at least one feasible driving strategy of the first vehicle in the lateral and / or longitudinal directions, and the decision information includes the driving strategy of the first vehicle in the lateral and / or longitudinal directions. For example, the lateral driving strategy may include changing lanes to the left, maintaining a straight line, and changing lanes to the right, and the longitudinal driving strategy may include acceleration and deceleration.

[0008] For example, the first device can be the first vehicle or other devices that are communicatively connected to the first vehicle.

[0009] In this implementation, first information corresponding to the first obstacle is obtained; based on the first information, at least two pieces of second information are determined, each piece of second information indicating the location of the predicted trajectory; then, a set of strategies corresponding to each piece of second information is obtained, each set of strategies including at least one feasible driving strategy for the vehicle in the lateral and / or longitudinal directions; then, based on each piece of second information and the set of strategies corresponding to each piece of second information, the vehicle's planned trajectory and / or decision information are determined, the decision information including the vehicle's driving strategy in the lateral and / or longitudinal directions; since the first information includes information on at least two predicted trajectories of the second obstacle, more information about the obstacle is considered in the process of determining the vehicle's planned trajectory and / or decision information. Instead of considering only the predicted trajectory with the highest probability for each obstacle, the planned trajectory is designed to reduce risk. Furthermore, instead of directly treating the positions occupied by all predicted trajectories of obstacles as locations where the vehicle cannot drive, thus avoiding overly conservative planned trajectories, the vehicle determines feasible driving strategies for each piece of secondary information after obtaining at least two pieces of secondary information. Then, the planned trajectory and / or decision information of the vehicle are comprehensively determined based on the set of strategies corresponding to each piece of secondary information. In other words, the multiple possible predicted trajectories of obstacles are fully considered, while avoiding overly conservative planned trajectories. This not only improves the safety of the planned trajectory but also minimizes the reduction in traffic efficiency.

[0010] In one possible implementation, the first device determines the vehicle's planned trajectory and / or the vehicle's decision information based on each piece of second information and the set of strategies corresponding to each piece of second information. This may include: the first device determining a first planned trajectory of the vehicle corresponding to each piece of second information based on each piece of second information and the set of strategies corresponding to each piece of second information, and then determining the vehicle's planned trajectory based on all the first planned trajectories corresponding to at least two pieces of second information.

[0011] In this implementation, the predicted trajectories of obstacles around the first vehicle are divided into at least two groups. Based on the predicted trajectories of obstacles included in each group, the first vehicle's trajectory is planned to obtain the first planned trajectory. This avoids each first planned trajectory being overly conservative and helps to obtain a more efficient passage trajectory. When determining the final planned trajectory of the first vehicle, all first planned trajectories corresponding to at least two second pieces of information are comprehensively considered, which helps to improve the safety of the final planned trajectory.

[0012] In one possible implementation, after determining the final planned trajectory of the first vehicle, the first device can also determine decision information corresponding to the final planned trajectory of the first vehicle, the aforementioned decision information including the driving strategy of the first vehicle in the lateral and / or longitudinal directions.

[0013] For example, the first device can determine whether the first vehicle has a lateral strategy of going left, straight, or right based on the final planned trajectory of the first vehicle, and can determine whether the first vehicle has a longitudinal strategy of cutting in or yielding based on the planned trajectory of the first vehicle and the predicted trajectory of the surrounding dynamic obstacles, thereby obtaining the driving strategy of the first vehicle in the lateral and / or longitudinal directions.

[0014] In this implementation, not only is the planned trajectory of the first vehicle determined, but also the driving strategy of the vehicle in the lateral and / or longitudinal directions. That is, more information used in the driving process is generated, which not only improves the interpretability of this solution, but also improves the efficiency of this solution in the dimension of information generation.

[0015] In one possible implementation, the first device determines a first planned trajectory of the first vehicle corresponding to each second piece of information based on each second piece of information and the policy set corresponding to each second piece of information. This includes: the first device determining a second planned trajectory of the first vehicle corresponding to each second piece of information based on each second piece of information and the policy set corresponding to each second piece of information; then performing a fusion operation to obtain a reference trajectory based on the second planned trajectories corresponding to at least two pieces of second information; and then the first device generating a first planned trajectory of the first vehicle corresponding to each second piece of information based on each second piece of information and the reference trajectory. The goal of generating the first planned trajectory includes the first planned trajectory conforming to the reference trajectory.

[0016] For example, both the second planned trajectory of the first vehicle and the first planned trajectory of the first vehicle can include trajectory information corresponding to each moment in the multiple moments of the second time period after the current moment. The starting time point of the second time period is the current moment, and the ending time point of the second time period is the first time point. The trajectory information corresponding to each moment includes the position information, speed and acceleration information of the first vehicle at that moment. The position information can include coordinates, and optionally, the position information can also include orientation angle.

[0017] Optionally, the first device can perform trajectory planning for the first vehicle using a preset algorithm based on each piece of second information and the strategy set corresponding to each piece of second information, to obtain at least one candidate planned trajectory corresponding to each piece of second information; the first device selects the second planned trajectory for the first vehicle from the at least one candidate planned trajectory corresponding to each piece of second information. For example, the aforementioned preset algorithm includes, but is not limited to, graph-based search algorithms, random sampling-based algorithms, optimization algorithms, or other types of path trajectory planning algorithms.

[0018] In this implementation, a second planning trajectory corresponding to each second piece of information is first determined based on each second piece of information and the strategy set corresponding to each second piece of information. Based on the second planning trajectories corresponding to at least two pieces of information, a fusion operation is performed to obtain a reference trajectory. Based on the second information and the reference trajectory, a first planning trajectory corresponding to each second piece of information is generated. Then, based on all the first planning trajectories corresponding to at least two pieces of information, the final planning trajectory obtained by the first vehicle is determined. Since the reference trajectory is obtained after performing a fusion operation on the second planning trajectories corresponding to at least two pieces of information, that is, the reference trajectory is obtained based on all the second information in at least two pieces of information, and the goal when generating the first planning trajectory includes that the first planning trajectory needs to fit the reference trajectory, the final planning trajectory used by the first vehicle takes into account all the second information, thereby improving the safety of the final planning trajectory. Furthermore, since each first planning trajectory only needs to fit the reference trajectory, instead of using the reference trajectory as a hard constraint, the determination process of the final planning trajectory obtained by the first vehicle is not overly conservative, which is conducive to improving the efficiency of the first vehicle's driving process.

[0019] In one possible implementation, both the first and second planned trajectories include the trajectory of the first vehicle from the current time to a first time point, and the reference trajectory includes the trajectory of the first vehicle from the current time to a second time point, where the second time point is earlier than the first time point. In this implementation, the reference trajectory includes the trajectory from the current time to the second time point, and the first planned trajectory includes the trajectory from the current time to the first time point. If the second time point and the first time point are the same, then the entire planned trajectory in each first planned trajectory needs to fit the reference trajectory, which will greatly limit the number of locations the first vehicle can explore during trajectory planning. Setting the second time point to be earlier than the first time point allows for the exploration of more locations when planning the trajectory for the first vehicle, resulting in a better-performing planned trajectory and thus improving the user experience for the driver and passengers.

[0020] In one possible implementation, the first device performs a fusion operation to obtain a reference trajectory based on the second planned trajectories corresponding to at least two pieces of second information. This includes: the first device fusing all the second planned trajectories corresponding to at least two pieces of second information to obtain a fused trajectory; and then, based on the at least two pieces of second information and the fused trajectory, determining the collision time point where the fused trajectory collides with an obstacle surrounding the first vehicle. The obstacle surrounding the first vehicle includes at least one first obstacle, and the collision time point is used to determine a second time point, which is earlier than the collision time point. After determining the second time point, the first device can fuse the sub-planned trajectories within each of the second planned trajectories corresponding to at least two pieces of second information to obtain the reference trajectory. Each sub-planned trajectory includes a trajectory from the current time to the second time point within a second planned trajectory.

[0021] In this implementation, all second-planned trajectories are fused to obtain a fused trajectory. The collision time point between the fused trajectory and obstacles around the vehicle is determined, and this second time point is set earlier than the collision time point. Then, the sub-planned trajectories in each second-planned trajectory are fused to obtain a reference trajectory. Each sub-planned trajectory includes the trajectory from the current time to the second time point, thus ensuring the safety of the reference trajectory. Each first-planned trajectory needs to fit the reference trajectory, which also improves the safety of each first-planned trajectory. This is beneficial to improving the safety of the final planned trajectory of the first vehicle, and thus improving the safety of the vehicle driving process.

[0022] In one possible implementation, the factors for determining the second time point further include: a time point where the first distance is greater than or equal to a first preset distance, and / or a time point where the second distance is greater than or equal to a second preset distance, wherein the first distance includes the maximum distance between different second planning trajectories among all second planning trajectories, and the second distance includes the maximum distance between different predicted trajectories among at least two predicted trajectories of the same second obstacle. If the second time point is set too early, meaning the time between the current moment and the second time point is too short, the predicted trajectories of obstacles around the vehicle will not show a clear trend. This results in high similarity between the sub-planned trajectories corresponding to different second information, making the fused reference trajectory less meaningful. The vehicle's trajectory planning process still cannot fully utilize the predicted trajectories of different obstacles in the surrounding environment. Conversely, if the second time point is set too late, meaning the time between the current moment and the second time point is too long, the obtained reference trajectory will over-consider the influence of the predicted trajectories of each obstacle around the vehicle. Influenced by the reference trajectory, each first-planned trajectory will be overly conservative, and the changes between the trajectory before and after the second time point in each first-planned trajectory may be significant. This leads to greater acceleration of the vehicle when executing the trajectory after the second time point, making it impossible to guarantee the feasibility of the final determined driving trajectory. In this implementation, the factors for determining the second time point also include: the time point at which the maximum distance between different second planned trajectories in all second planned trajectories is greater than or equal to the first preset distance, which helps to avoid excessive changes between the trajectory before and after the second time point in each first planned trajectory; and / or the time point at which the maximum distance between different predicted trajectories in at least two predicted trajectories of the same second obstacle is greater than or equal to the second preset distance, which helps to avoid setting the second time point too early. Therefore, the second time point determined by considering various factors, and the reference trajectory determined based on the second time point, guides the final determined driving trajectory. This helps to ensure the safety of the final determined driving trajectory, avoid the final determined driving trajectory from being too conservative, and improve the smoothness of the final determined driving trajectory.

[0023] In one possible implementation, the first device can determine the planned trajectory of the first vehicle every preset time interval. The first device can also obtain a first reference trajectory generated during the previous determination of the planned trajectory of the first vehicle; the first reference trajectory can also be called a historical reference trajectory. The first device fuses the sub-planned trajectories in each of all second planned trajectories corresponding to at least two pieces of second information to obtain a reference trajectory. This can include: the first device fusing the sub-planned trajectories in each of all second planned trajectories corresponding to at least two pieces of second information to obtain a second reference trajectory, and fusing the first reference trajectory and the second reference trajectory to obtain the final reference trajectory.

[0024] In this implementation, the historical reference trajectory generated during the previous determination of the first vehicle's planned trajectory is also incorporated when generating the final reference trajectory. This avoids excessive jumps between two adjacent planned trajectories for the first vehicle, ensuring the continuity between different planned trajectories of the first vehicle, which is beneficial to improving the comfort of the first vehicle during driving.

[0025] In one possible implementation, the first obstacle is selected from multiple obstacles around the first vehicle based on the third planned trajectory of the first vehicle. The planned trajectory of the first vehicle is determined every preset time interval. The third planned trajectory of the first vehicle is obtained based on the previously determined planned trajectory of the first vehicle. The selected first obstacle satisfies at least one of the following conditions: the predicted trajectory of the first obstacle intersects with the third planned trajectory of the first vehicle, or the distance between the predicted trajectory of the first obstacle and the third planned trajectory of the first vehicle is less than or equal to a preset distance.

[0026] For example, the first device acquiring first information corresponding to at least one first obstacle around the first vehicle may include: after acquiring information on one or more predicted trajectories of each of the multiple dynamic obstacles around the first vehicle, the first device may, based on the information on all predicted trajectories of each dynamic obstacle and the third planned trajectory of the first vehicle, filter out at least one first obstacle from the multiple dynamic obstacles around the first vehicle, thereby acquiring the first information corresponding to at least one first obstacle around the first vehicle.

[0027] In this implementation, the vehicle's trajectory is planned every preset time interval. Based on the previously determined planned trajectory, a third planned trajectory is determined. Then, multiple obstacles around the vehicle are filtered based on the third planned trajectory. The predicted trajectories of the filtered first obstacles intersect with the third planned trajectory, or the distance between the predicted trajectories of the filtered first obstacles and the third planned trajectory is less than or equal to a preset distance. That is, dynamic obstacles with the possibility of interaction with the vehicle are first filtered from multiple obstacles around the vehicle. Then, the vehicle's trajectory is planned based on the predicted trajectories of the aforementioned dynamic obstacles with the possibility of interaction with the vehicle. This helps to reduce the computer resources consumed in the trajectory planning process.

[0028] In one possible implementation, during the process of the first device selecting at least one first obstacle from multiple dynamic obstacles around the first vehicle, the first device may further classify the selected at least one first obstacle into first obstacles that have strong interaction with the first vehicle and first obstacles that have weak interaction with the first vehicle. For example, if all predicted trajectories of a certain dynamic obstacle intersect with the third planned trajectory of the first vehicle, then the certain dynamic obstacle can be determined as a first obstacle that has strong interaction with the first vehicle; if all predicted trajectories of a certain dynamic obstacle do not intersect with the third planned trajectory of the first vehicle, and the closest distance between all predicted trajectories of a certain dynamic obstacle and the third planned trajectory of the first vehicle is less than or equal to a preset distance, then the certain dynamic obstacle can be determined as a first obstacle that has weak interaction with the first vehicle. The first device acquiring first information corresponding to at least one first obstacle around the first vehicle may include: the first device acquiring information on all predicted trajectories of the first obstacle that has strong interaction with the first vehicle, and acquiring information on the predicted trajectory with the highest probability value from all predicted trajectories of the first obstacle that has weak interaction with the first vehicle, that is, obtaining the first information, wherein the first information includes information on all predicted trajectories of the first obstacle that has strong interaction with the first vehicle, and information on the predicted trajectory with the highest probability value of the first obstacle that has weak interaction with the first vehicle.

[0029] In one possible implementation, the second information can be represented as a raster map, a way of representing the surrounding environment in a rasterized manner. Each piece of second information can indicate at least one raster occupied by a predicted trajectory, thereby indicating the location of at least one predicted trajectory. For example, the environment surrounding the first vehicle can be rasterized; if a raster is occupied by a predicted trajectory, the raster is filled with color; if a raster is not occupied by a predicted trajectory, the raster is not filled with color.

[0030] This implementation explicitly uses a grid map to represent the predicted trajectory of surrounding obstacles, and has better compatibility with the vehicle trajectory planning process, thus reducing the implementation difficulty of this solution.

[0031] In one possible implementation, the first device determines at least two pieces of second information based on the first information, which may include: the first device determining a set of strategies corresponding to each predicted trajectory based on the information of each predicted trajectory in the information of at least one predicted trajectory included in the first information, wherein the set of strategies corresponding to each predicted trajectory includes at least one feasible driving strategy for the first vehicle in the lateral and / or longitudinal directions determined based on each predicted trajectory; the first device dividing the information of the at least two predicted trajectories included in the first information into at least two groups based on the set of strategies corresponding to each predicted trajectory, wherein each group includes information of at least one predicted trajectory, and the set of strategies corresponding to different predicted trajectories in each group is the same, that is, predicted trajectories corresponding to the same set of strategies are assigned to the same group; then the first device representing each predicted trajectory in a rasterized form based on the information of all predicted trajectories included in each of the at least two groups, optionally also representing static obstacles around the first vehicle in a rasterized form, thereby obtaining a piece of second information; the first device repeats the aforementioned operation to obtain at least two pieces of second information corresponding one-to-one with the at least two groups.

[0032] Secondly, this application provides an intelligent driving device that can be applied in the field of intelligent driving. The device includes: an acquisition module, configured to acquire first information corresponding to at least one first obstacle around the vehicle, the first information including information on at least one predicted trajectory of each first obstacle, and a second obstacle existing among the at least one first obstacle, the first information including information on at least two predicted trajectories of the second obstacle; a determination module, configured to determine at least two pieces of second information based on the first information, wherein the second information indicates the position of the predicted trajectory, and different pieces of second information indicate the position of different predicted trajectories among the predicted trajectories corresponding to the first information; the acquisition module is further configured to acquire a set of strategies corresponding to each piece of second information, the set of strategies corresponding to each piece of second information including at least one feasible driving strategy for the vehicle in the lateral and / or longitudinal directions; the determination module is further configured to determine the planned trajectory of the vehicle and / or the decision information of the vehicle based on each piece of second information and the set of strategies corresponding to each piece of second information, the decision information including the driving strategy of the vehicle in the lateral and / or longitudinal directions.

[0033] In this second aspect, the intelligent driving device is also used to perform the steps performed by the first device in the first aspect and various possible implementations of the first aspect. The meanings of the terms in the second aspect and various possible implementations of the second aspect, as well as the beneficial effects brought about by the various possible implementations, can be referred to the descriptions in the various possible implementations of the first aspect, and will not be repeated here.

[0034] Thirdly, embodiments of this application provide an apparatus including a processor and a memory, the processor being coupled to the memory, the memory being used to store a program; and the processor being used to execute the program in the memory, causing the apparatus to perform the method described in the first aspect above.

[0035] Fourthly, embodiments of this application provide a vehicle including a processor and a memory, the processor being coupled to the memory, the memory being used to store a program; the processor being used to execute the program in the memory, causing the vehicle to perform the method described in the first aspect above.

[0036] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the methods described in the first or second aspect above.

[0037] Sixthly, embodiments of this application provide a computer program product, which includes a program that, when run on a computer, causes the computer to perform the methods described in the first or second aspect above.

[0038] Seventhly, this application provides a chip system including a processor for supporting the implementation of the functions involved in the foregoing aspects, such as transmitting or processing data and / or information involved in the foregoing methods. In one possible design, the chip system further includes a memory for storing program instructions and data necessary for the terminal device or communication device. This chip system may be composed of chips or may include chips and other discrete devices. Attached Figure Description

[0039] Figure 1 is a schematic diagram of an intelligent driving method provided in an embodiment of this application;

[0040] Figure 2 is a schematic diagram of different types of dynamic obstacles around the first vehicle provided in an embodiment of this application;

[0041] Figure 3 is another schematic diagram of different types of dynamic obstacles around the first vehicle provided in the embodiment of this application;

[0042] Figure 4 is a schematic diagram of determining a candidate planning trajectory provided in an embodiment of this application;

[0043] Figure 5 is a schematic diagram of obtaining a reference trajectory by performing a fusion operation based on a second planning trajectory corresponding to at least two second pieces of information, according to an embodiment of this application.

[0044] Figure 6 is a schematic diagram of the first planned trajectory determined based on reference trajectories determined at different second time points, according to an embodiment of this application.

[0045] Figure 7 is a schematic diagram of a traffic scenario where a vehicle passes through an intersection, as provided in an embodiment of this application.

[0046] Figure 8 is a schematic diagram of the speed and acceleration changes of the first vehicle when passing through an intersection according to an embodiment of this application;

[0047] Figure 9 is a structural schematic diagram of an intelligent driving device provided in an embodiment of this application;

[0048] Figure 10 is a schematic diagram of a device provided in an embodiment of this application;

[0049] Figure 11 is a structural schematic diagram of a vehicle provided in an embodiment of this application. Detailed Implementation

[0050] The embodiments of this application will now be described with reference to the accompanying drawings. Obviously, the described embodiments are merely some, and not all, of the embodiments of this application. Those skilled in the art will recognize that, with the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0051] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms are interchangeable where appropriate; this is merely a way of distinguishing objects with the same attributes in the description of embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0052] In the embodiments of this application, "instruction" can include direct and indirect instructions, as well as explicit and implicit instructions. The information indicated by a certain piece of information (hereinafter referred to as instruction information) is called the information to be instructed. In specific implementation, there are many ways to indicate the information to be instructed, such as, but not limited to, directly indicating the information to be instructed, such as the information to be instructed itself or its index. It can also indirectly indicate the information to be instructed by indicating other information, where there is an association between the other information and the information to be instructed; or it can indicate only a part of the information to be instructed, while the other parts are known or pre-agreed upon. For example, the instruction can be implemented by using a pre-agreed (e.g., protocol predefined) arrangement of various information, thereby reducing the instruction overhead to a certain extent. This application does not limit the specific method of instruction. It is understood that for the sender of the instruction information, the instruction information can be used to indicate the information to be instructed; for the receiver of the instruction information, the instruction information can be used to determine the information to be instructed.

[0053] The method provided in this application can be applied to the field of intelligent driving. Optionally, it can be used to determine the planned trajectory of a vehicle and / or the vehicle's decision information, including the vehicle's driving strategy in the lateral and / or longitudinal directions. The vehicle in this application can be a car, truck, motorcycle, bus, ship, airplane, helicopter, recreational vehicle, amusement park vehicle, tram, golf cart, or train, etc., which can be determined in combination with the actual application scenario.

[0054] For example, a vehicle's trajectory can be planned based on the predicted trajectories of obstacles around the vehicle. There may be at least two predicted trajectories for the same obstacle. In related technologies, only the predicted trajectory with the highest probability for each obstacle is generally considered, and the vehicle's trajectory is planned based on the predicted trajectory with the highest probability for each obstacle. However, the trajectory planning based on the aforementioned method does not take into account low-probability events, resulting in a high risk of the planned trajectory.

[0055] To address the aforementioned issues, this application discloses: obtaining first information corresponding to a first obstacle; determining at least two pieces of second information based on the first information, each piece of second information indicating the location of a predicted trajectory; then obtaining a strategy set corresponding to each piece of second information, each strategy set including at least one feasible driving strategy for the vehicle in the lateral and / or longitudinal directions; and then determining the vehicle's planned trajectory and / or decision information based on each piece of second information and the strategy set corresponding to each piece of second information, the decision information including the vehicle's driving strategy in the lateral and / or longitudinal directions; since the first information includes information on at least two predicted trajectories of the second obstacle, that is, in the process of determining the vehicle's planned trajectory and / or decision information... Considering more predicted trajectories for obstacles, rather than just the one with the highest probability for each obstacle, helps reduce the risk of the planned trajectory. Furthermore, instead of directly treating the positions occupied by all predicted obstacle trajectories as locations where the vehicle cannot travel, thus avoiding overly conservative planned trajectories, it determines feasible driving strategies for the vehicle based on each piece of secondary information. Then, based on the strategy set corresponding to each piece of secondary information, it comprehensively determines the vehicle's planned trajectory and / or decision information. In other words, it fully considers multiple possible predicted trajectories for obstacles while avoiding overly conservative planned trajectories, thereby improving the safety of the planned trajectory and minimizing the reduction in traffic efficiency.

[0056] Based on the above description, the detailed implementation process of the intelligent driving method provided in this application will be introduced below. Specifically, please refer to Figure 1, which is a schematic diagram of an intelligent driving method provided in an embodiment of this application. As shown in Figure 1, the intelligent driving method provided in this application may include:

[0057] 101. Obtain first information corresponding to at least one first obstacle around the first vehicle, the first information including information on at least one predicted trajectory of each first obstacle, and a second obstacle exists among the at least one first obstacle, the first information including information on at least two predicted trajectories of the second obstacle.

[0058] For example, the first device can acquire information about at least one predicted trajectory of each of a plurality of dynamic obstacles surrounding the first vehicle, thereby determining first information corresponding to at least one first obstacle surrounding the first vehicle. Optionally, the first device can also acquire a probability value corresponding to each predicted trajectory of each dynamic obstacle.

[0059] The first device can be the first vehicle or other devices that are communicatively connected to the first vehicle, depending on the specific application scenario. Each of the at least one first obstacle can be understood as a dynamic obstacle. The second obstacle is similar to the first obstacle and can also be understood as a dynamic obstacle in the environment surrounding the first vehicle. The concept of a second obstacle is introduced only to indicate that among the multiple predicted trajectories included in the first information, there are at least two predicted trajectories containing the same second obstacle.

[0060] Information about each predicted trajectory for each dynamic obstacle can indicate the position of a predicted trajectory of a dynamic obstacle within a first time interval after the current moment. For example, the first time interval can be 5 seconds, 6 seconds, 8 seconds, 10 seconds, or other durations after the current moment. Optionally, the information about each predicted trajectory also includes a probability value corresponding to the predicted trajectory, which represents the probability that the dynamic obstacle will move along the predicted trajectory. Optionally, the information about each predicted trajectory may also include the size of the dynamic obstacle; for example, the information about the predicted trajectory of a dynamic obstacle may also include the length and width of the dynamic obstacle.

[0061] Information about each predicted trajectory of the first obstacle may include the predicted position of the first obstacle at each time point in multiple time periods within a first time period, and the identification information of the predicted trajectory; exemplarily, the predicted position may include the coordinates of the first obstacle, and optionally, the predicted position may also include the orientation angle of the first obstacle. Optionally, information about a predicted trajectory of the first obstacle may also include the predicted velocity of the first obstacle at each time point in multiple time periods within the first time period.

[0062] Furthermore, the information of each predicted trajectory of the first obstacle can be obtained based on the state information of the first obstacle at multiple time points within the first time period. The state information of the first obstacle at multiple time points within the first time period can include the current state information and the predicted state information of the first obstacle. The state information of the first obstacle can be represented as: State dynamic =[x, y, θ, t, length, width], where, State dynamic The first obstacle's state information is represented by x and y, which represent the two-dimensional coordinates of the first obstacle. θ represents the first obstacle's orientation angle, t represents a time point within the first time period, and length and width represent the length and width of the first obstacle. The predicted position of the first obstacle at each time point within the first time period can be directly obtained from the first obstacle's state information, and the predicted velocity of the first obstacle at each time point within the first time period can be calculated based on x, y, and t in the state information.

[0063] In one implementation, the at least one first obstacle can be selected from multiple dynamic obstacles around the first vehicle based on the predicted trajectory of each dynamic obstacle. For example, step 101 may include: after acquiring information about one or more predicted trajectories of each dynamic obstacle among multiple dynamic obstacles around the first vehicle, the first device can select at least one first obstacle from the multiple dynamic obstacles around the first vehicle based on the information of all predicted trajectories of each dynamic obstacle and the third planned trajectory of the first vehicle, thereby obtaining first information corresponding to the at least one first obstacle around the first vehicle. The selected first obstacle satisfies at least one of the following conditions: the predicted trajectory of the first obstacle intersects with the third planned trajectory of the first vehicle, or the distance between the predicted trajectory of the first obstacle and the third planned trajectory of the first vehicle is less than or equal to a preset distance; for example, the preset distance can be 5 meters, 8 meters, 10 meters, or other lengths, which can be determined based on the actual application scenario.

[0064] Regarding the specific implementation of the first device acquiring the third planned trajectory of the first vehicle, optionally, the first device can determine the planned trajectory of the first vehicle every preset time interval. The third planned trajectory of the first vehicle can be obtained based on the previously determined planned trajectory of the first vehicle. For example, if the previously determined planned trajectory of the first vehicle specifically takes the form of a main branch trajectory plus multiple branch trajectories, the first device can select one branch trajectory from the multiple branch trajectories, and determine the main branch trajectory plus the selected branch trajectory as the third planned trajectory. The aforementioned selected branch trajectory can be randomly selected from the multiple branch trajectories, or it can be the one with the highest score among the multiple branch trajectories, or it can be selected from the multiple branch trajectories according to other trajectories. The concepts of main branch trajectory, branch trajectory, and the score value of the branch trajectory will be described in subsequent steps and will not be elaborated here. Alternatively, if the previously determined planned trajectory of the first vehicle specifically takes the form of a single trajectory, the previously determined planned trajectory of the first vehicle can be directly used as the third planned trajectory of the first vehicle.

[0065] In this embodiment, the vehicle's trajectory is planned every preset time interval. A third planned trajectory is determined based on the previously determined planned trajectory of the vehicle. Then, multiple obstacles around the vehicle are filtered based on the third planned trajectory. The predicted trajectories of the filtered first obstacles intersect with the third planned trajectory, or the distance between the predicted trajectories of the filtered first obstacles and the third planned trajectory is less than or equal to a preset distance. That is, dynamic obstacles with the possibility of interaction with the vehicle are first filtered from multiple obstacles around the vehicle. Then, the vehicle's trajectory is planned based on the predicted trajectories of the aforementioned dynamic obstacles with the possibility of interaction with the vehicle. This helps to reduce the computer resources consumed in the trajectory planning process of the vehicle.

[0066] Alternatively, after acquiring information on one or more predicted trajectories for each of the multiple dynamic obstacles surrounding the first vehicle, as well as the probability value corresponding to each predicted trajectory, the first device can acquire the predicted trajectory with the highest probability for each dynamic obstacle. Based on the predicted trajectory with the highest probability for each dynamic obstacle, the first device can perform trajectory planning for the first vehicle, thereby obtaining the third planned trajectory for the first vehicle.

[0067] Alternatively, after acquiring information on one or more predicted trajectories of each of the multiple dynamic obstacles surrounding the first vehicle, the first device can perform trajectory planning for the first vehicle based on all the predicted trajectories of each dynamic obstacle, thereby obtaining the third planned trajectory of the first vehicle. It should be understood that the specific implementation method of the first device obtaining the third planned trajectory of the first vehicle can be flexibly determined in combination with the actual application scenario.

[0068] Regarding the specific implementation of the first device selecting at least one first obstacle from multiple dynamic obstacles around the first vehicle, for ease of description, any one of the multiple dynamic obstacles around the first vehicle will be referred to as the target dynamic obstacle. For example, after obtaining information on one or more predicted trajectories of each dynamic obstacle among the multiple dynamic obstacles around the first vehicle, the first device can determine whether there is an intersection between all the predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle based on the information on all the predicted trajectories of the target dynamic obstacle. If at least one predicted trajectory among all the predicted trajectories of the target dynamic obstacle intersects with the third planned trajectory of the first vehicle, then the target dynamic obstacle can be identified as the first obstacle.

[0069] If there is no intersection between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle, the nearest distance between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle can be obtained. It is then determined whether the aforementioned nearest distance is less than or equal to a preset distance. If the aforementioned nearest distance is less than or equal to the preset distance, the target dynamic obstacle can be identified as the first obstacle. If the nearest distance is greater than the preset distance, the target dynamic obstacle can be identified as an obstacle that does not interact with the first vehicle, and the target dynamic obstacle can be discarded. Then, it is determined whether the next dynamic obstacle is the first obstacle.

[0070] Alternatively, if there is no intersection between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle, the target dynamic obstacle can be directly discarded, and then it can be determined whether the next dynamic obstacle is the first obstacle.

[0071] Alternatively, instead of determining whether there is an intersection between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle, it can be directly determined whether the nearest distance between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle is less than or equal to a preset distance. If the aforementioned nearest distance is less than or equal to the preset distance, the target dynamic obstacle can be identified as the first obstacle; if the nearest distance is greater than the preset distance, the target dynamic obstacle can be discarded, and then it can be determined whether the next dynamic obstacle is the first obstacle.

[0072] The first device can perform the above operation on each of the multiple dynamic obstacles around the first vehicle, thereby enabling the selection of at least one first obstacle from the multiple dynamic obstacles around the first vehicle.

[0073] Regarding the specific implementation of the first device acquiring first information corresponding to at least one first obstacle around the first vehicle, optionally, during the process of the first device selecting at least one first obstacle from multiple dynamic obstacles around the first vehicle, the first device may further classify the selected at least one first obstacle into first obstacles with strong interaction with the first vehicle and first obstacles with weak interaction with the first vehicle; for example, if all predicted trajectories of the target dynamic obstacle intersect with the third planned trajectory of the first vehicle, the target dynamic obstacle can be determined as a first obstacle with strong interaction with the first vehicle; if there is no intersection between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle, and the closest distance between all predicted trajectories of the target dynamic obstacle and the third planned trajectory of the first vehicle is less than or equal to a preset distance, the target dynamic obstacle can be determined as a first obstacle with weak interaction with the first vehicle. The first device acquiring first information corresponding to at least one first obstacle around the first vehicle may include: the first device acquiring information on all predicted trajectories of the first obstacle that has strong interaction with the first vehicle, and acquiring information on the predicted trajectory with the highest probability value from all predicted trajectories of the first obstacle that has weak interaction with the first vehicle, that is, obtaining the first information, wherein the first information includes information on all predicted trajectories of the first obstacle that has strong interaction with the first vehicle, and information on the predicted trajectory with the highest probability value of the first obstacle that has weak interaction with the first vehicle.

[0074] To better understand this solution, please refer to Figure 2. Figure 2 is a schematic diagram of different types of dynamic obstacles around the first vehicle provided in this application embodiment. Figure 2 shows the first obstacle with strong interaction with the first vehicle, the first obstacle with weak interaction with the first vehicle, and the obstacle with no interaction with the first vehicle. In Figure 2, rectangles of different shades represent the dynamic obstacles around the first vehicle. The darkest rectangle represents the first obstacle with strong interaction with the first vehicle. As shown in Figure 2, the predicted trajectory of each darkest rectangle intersects with the third planned trajectory of the first vehicle. Rectangles with medium shades represent the first obstacle with weak interaction with the first vehicle. If any predicted trajectory of a dynamic obstacle is located within the gray fan-shaped area in Figure 2, it means that the distance between the predicted trajectory of the dynamic obstacle and the third planned trajectory of the first vehicle is less than or equal to a preset distance. Rectangles with medium shades are located within this gray area. The lightest rectangle represents the obstacle with no interaction with the first vehicle. It should be understood that the examples in Figure 2 are only for the convenience of understanding this solution and are not intended to limit this solution.

[0075] Please refer to Figure 3. Figure 3 is another schematic diagram of different types of dynamic obstacles around the first vehicle provided in the embodiment of this application. The vehicle in Figure 3 represents the first vehicle in this application. Figure 3 shows multiple dynamic obstacles around the vehicle, such as multiple pedestrians, multiple bicycles, and multiple other vehicles. The dynamic obstacles that interact with the vehicle are circled by ellipses. It should be understood that the examples in Figure 3 are only for the convenience of understanding this solution and are not intended to limit this solution.

[0076] Alternatively, the first device can directly acquire information on all predicted trajectories of each first obstacle to obtain the first information. That is, the first information may include information on all predicted trajectories of each first obstacle around the first vehicle. The specific information included in the first information can be determined in combination with the actual application scenario.

[0077] Optionally, in step 101, the first device may also acquire the state information of each static obstacle among at least one static obstacle around the first vehicle, wherein the state information of the static obstacle includes at least the position of each static obstacle, and optionally, the state information of the static obstacle may also include the size of the static obstacle.

[0078] For example, the state information of a static obstacle can be represented as: State static =[x, y, θ, length,, width]], where, State static This represents the state information of a static obstacle. x and y represent the two-dimensional coordinates of the static obstacle, θ represents the orientation angle of the static obstacle, and length and width represent the length and width of the static obstacle.

[0079] 102. Based on the first information, determine at least two pieces of second information, wherein the second information indicates the position of the predicted trajectory, and different pieces of second information indicate the positions of different predicted trajectories in the predicted trajectories corresponding to the first information.

[0080] For example, after obtaining information on at least two predicted trajectories included in the first information (optionally, also including the probability value corresponding to each predicted trajectory), the first device can determine at least two pieces of second information based on the first information. Each piece of second information can indicate the position of at least one predicted trajectory, and different pieces of second information indicate the positions of different predicted trajectories among all predicted trajectories corresponding to the first information.

[0081] Optionally, the second information can be represented in the form of a raster map, which is a rasterized representation of the surrounding environment. Each piece of second information can indicate at least one raster occupied by the predicted trajectory, thereby indicating the location of at least one predicted trajectory. For example, the environment around the first vehicle can be rasterized. If a raster is occupied by a predicted trajectory, the raster is filled with color. If a raster is not occupied by a predicted trajectory, the raster is not filled with color.

[0082] Optionally, if the probability value corresponding to each predicted trajectory is also obtained in step 101, then each piece of second information may also indicate the probability that each grid cell is occupied. Optionally, if the state information of static obstacles around the first vehicle is also obtained in step 101, then each piece of second information may also indicate the grid cells occupied by all static obstacles around the first vehicle.

[0083] For example, different occupancy probabilities of a grid can be represented by filling it with different colors. For instance, a red grid represents a high occupancy probability, while a blue grid represents a low occupancy probability. Alternatively, different shades of the same color can be used to represent different occupancy probabilities of a grid. For instance, a grid filled entirely with red represents that the grid is occupied, with a darker shade of red indicating a higher occupancy probability and a lighter shade of red indicating a lower occupancy probability. The specific implementation method can be determined based on the actual application scenario.

[0084] For example, each piece of second information can be represented as a three-dimensional grid map. The three-dimensional grid map can represent the occupation of the grid in both spatial and temporal dimensions. The three-dimensional grid map includes an x-axis, a y-axis, and a t-axis. The plane constructed by the x-axis and y-axis represents a two-dimensional map around the first vehicle, and the t-axis represents time. For example, the interval of the constructed grid map in the time dimension can be 0.5s, 1s, or other lengths. The specific representation of the grid map can be determined in combination with the actual application scenario.

[0085] Regarding the specific implementation of the first device determining at least two pieces of second information based on the first information, in one case, each piece of second information indicates the location of a predicted trajectory (optionally, it also indicates the probability of the predicted trajectory). Optionally, each piece of second information also indicates the location of static obstacles around the first vehicle. Then step 102 may include: after obtaining the information of at least two predicted trajectories included in the first information, the first device can represent each predicted trajectory in a rasterized form. Optionally, it can also represent the static obstacles around the first vehicle in a rasterized form, thereby obtaining one piece of second information. The first device repeats the aforementioned steps at least twice to obtain at least two pieces of second information corresponding one-to-one with the at least two predicted trajectories.

[0086] In another scenario, step 102 may include: the first device determining a set of strategies corresponding to each predicted trajectory based on the information of each predicted trajectory in the information of at least one predicted trajectory included in the first information, wherein the set of strategies corresponding to each predicted trajectory includes at least one feasible driving strategy for the first vehicle in the lateral and / or longitudinal directions determined based on each predicted trajectory; for example, the lateral driving strategy may include changing lanes to the left, maintaining a straight course, and changing lanes to the right, and the longitudinal driving strategy may include accelerating and decelerating.

[0087] The first device, based on the strategy set corresponding to each predicted trajectory, divides the information of at least two predicted trajectories obtained in step 101 into at least two groups. Each group includes information of at least one predicted trajectory, and the strategy sets corresponding to different predicted trajectories in each group are the same. That is, predicted trajectories corresponding to the same strategy set are assigned to the same group. It should be noted that since each strategy set includes at least one driving strategy, the fact that the strategy sets corresponding to two different predicted trajectories are the same means that all driving strategies included in the strategy sets corresponding to the two different predicted trajectories are the same. For example, strategy set 1 corresponding to predicted trajectory 1 includes 3 feasible driving strategies, and strategy set 2 corresponding to predicted trajectory 2 includes 3 feasible driving strategies. If the 3 feasible driving strategies included in strategy set 1 and the 3 feasible driving strategies included in strategy set 2 are the same, it means that strategy set 1 and strategy set 2 are the same. It should be understood that the example here is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0088] The first device represents each predicted trajectory in a rasterized form based on information from all predicted trajectories included in each of at least two groups. Optionally, it also represents static obstacles around the first vehicle in a rasterized form, thereby obtaining a second piece of information. Optionally, if the probability value corresponding to each predicted trajectory is obtained in step 101, and different predicted trajectories included in the same group occupy the same grid, the probability values ​​corresponding to the different predicted trajectories can be weighted and summed to obtain the occupancy probability of the grid. The first device repeats the aforementioned operation to obtain at least two pieces of second information corresponding one-to-one with at least two groups.

[0089] In this embodiment, a grid map is used to represent the predicted trajectory of surrounding obstacles, which has better compatibility with the vehicle trajectory planning process and reduces the implementation difficulty of this solution.

[0090] Optionally, after obtaining the information of the multiple predicted trajectories included in the first information, the first device can also determine the third information, which indicates the position of all the predicted trajectories included in the first information; optionally, if the state information of static obstacles around the first vehicle is also obtained in step 101, the third information also indicates the position of the static obstacles around the first vehicle.

[0091] For example, the third information can be represented as a grid map. The first device can then represent each predicted trajectory in a gridded form based on the information of all predicted trajectories included in the first information. Optionally, it can also represent static obstacles around the first vehicle in a gridded form to obtain the third information. Optionally, if the probability value corresponding to each predicted trajectory is obtained in step 101, and different predicted trajectories occupy the same grid in the information of all predicted trajectories included in the first information, the probability values ​​corresponding to the different predicted trajectories can be weighted and summed to obtain the occupancy probability of that grid.

[0092] 103. Obtain the strategy set corresponding to each second piece of information, wherein the strategy set corresponding to each second piece of information includes at least one feasible driving strategy for the first vehicle in the lateral and / or longitudinal directions.

[0093] For example, after determining at least two pieces of second information, the first device can also obtain a set of strategies corresponding to each piece of second information. The set of strategies corresponding to each piece of second information includes the positions of all predicted trajectories indicated by each piece of second information, and at least one feasible driving strategy for the first vehicle in the lateral and / or longitudinal directions.

[0094] Furthermore, in one case, if each piece of second information indicates a predicted trajectory location, then in step 103, the first device needs to determine at least one feasible driving strategy corresponding to each piece of second information based on the location of a predicted trajectory indicated by each piece of second information (optionally, also based on the location of static obstacles around the first vehicle indicated by each piece of second information).

[0095] In another scenario, if each piece of second information corresponds to a group in step 102, and each group includes information on at least one predicted trajectory, since the policy sets corresponding to different predicted trajectories in the same group are all the same, that is, the policy sets corresponding to all predicted trajectories indicated by a second piece of information are all the same, then the policy set corresponding to each piece of second information can be directly obtained based on the policy set corresponding to each predicted trajectory determined in step 102.

[0096] To better understand this scheme, for example, the predicted trajectory 1 corresponds to policy set 1, the predicted trajectory 2 corresponds to policy set 2, and the predicted trajectory 3 corresponds to policy set 3. Since policy sets 1, 2, and 3 are all the same, the information of predicted trajectory 1, 2, and 3 is grouped into the same group. Thus, the second information 1 indicates predicted trajectory 1, 2, and 3. Therefore, any one of policy sets 1, 2, and 3 can be determined as the policy set corresponding to the second information 1. It should be understood that this example is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0097] 104. Based on each piece of second information and the set of strategies corresponding to each piece of second information, determine the planned trajectory of the first vehicle and / or the decision information of the first vehicle, wherein the decision information includes the driving strategies of the first vehicle in the lateral and / or longitudinal directions.

[0098] Optionally, step 104 may include: the first device determining a first planned trajectory for the first vehicle corresponding to each second piece of information based on each second piece of information and the strategy set corresponding to each second piece of information; and then determining the final planned trajectory for the first vehicle based on all the first planned trajectories corresponding to at least two second pieces of information. In this embodiment, the predicted trajectories of obstacles around the vehicle are divided into at least two groups, and the first vehicle is trajectory planned based on the predicted trajectories of obstacles included in each group to obtain the first planned trajectory of the first vehicle. This avoids each first planned trajectory being overly conservative and is conducive to obtaining a more efficient passage trajectory. When determining the final planned trajectory of the first vehicle, all the first planned trajectories corresponding to at least two second pieces of information are comprehensively considered, which is conducive to improving the safety of the final planned trajectory.

[0099] Regarding the specific implementation method of the first device determining the first planned trajectory of the first vehicle corresponding to each second piece of information, in one implementation method, the first device can determine the second planned trajectory of the first vehicle corresponding to each second piece of information based on each second piece of information and the strategy set corresponding to each second piece of information; perform a fusion operation to obtain a reference trajectory based on all the second planned trajectories corresponding to at least two pieces of second information; and then generate the first planned trajectory of the first vehicle corresponding to each second piece of information based on each second piece of information and the reference trajectory, wherein the goal of generating the first planned trajectory includes the first planned trajectory conforming to the reference trajectory.

[0100] For example, both the second planned trajectory of the first vehicle and the first planned trajectory of the first vehicle can include trajectory information corresponding to each moment in a second time period after the current moment, where the starting point of the second time period is the current moment and the ending point of the second time period is the first time point. The trajectory information corresponding to each moment includes the position information, velocity, and acceleration information of the first vehicle at that moment. The position information can include coordinates, and optionally, the position information can also include an orientation angle. For example, the trajectory information corresponding to each moment can be represented as: P node = [x, y, yaw, t, v, a], where x and y represent the coordinates in the trajectory information, yaw represents the orientation angle in the trajectory information, t represents the time corresponding to the trajectory information, v represents the velocity in the trajectory information, and a represents the acceleration in the trajectory information. It should be understood that the example here is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0101] The length of the second time segment can be the same as the length of the first time segment, or it can be different from the length of the first time segment. For example, the length of the second time segment can be 5 seconds, 6 seconds, or other durations, depending on the specific application scenario. The time interval between two adjacent moments in the multiple moments included in the second time segment can be a preset duration, such as 0.2 seconds, 0.5 seconds, or other durations, depending on the actual application scenario.

[0102] For example, regarding the specific implementation process of the first device determining the second planned trajectory of the first vehicle corresponding to each second piece of information, the first device can perform trajectory planning for the first vehicle using a preset algorithm based on each second piece of information and the strategy set corresponding to each second piece of information, to obtain at least one candidate planned trajectory corresponding to each second piece of information; the first device selects the second planned trajectory of the first vehicle from the at least one candidate planned trajectory corresponding to each second piece of information.

[0103] For example, the aforementioned preset algorithms include, but are not limited to, graph-based search algorithms, random sampling-based algorithms, optimization algorithms, or other types of path trajectory planning algorithms. Furthermore, graph-based search algorithms may include the A* algorithm, random sampling-based algorithms may include the rapidly exploring random tree (RRT) algorithm, and optimization algorithms may include simulated annealing, tabu search, etc. The specific preset algorithm used can be determined based on the actual application scenario.

[0104] To further understand this scheme, the following describes the specific implementation process of the two steps, "obtaining at least one candidate planning trajectory corresponding to each second piece of information" and "selecting a second planning trajectory from at least one candidate planning trajectory corresponding to each second piece of information", taking the preset algorithm as a random sampling algorithm as an example.

[0105] For ease of description, the current time is referred to as time a, the next time is referred to as time a+1, and the time immediately after that is referred to as time a+2. For the determination process of any candidate planned trajectory, for example, the first device can sample a feasible acceleration 1 of the first vehicle in the lateral direction and a feasible acceleration 2 of the first vehicle in the longitudinal direction. Based on acceleration 1, acceleration 2 and the starting position of the first vehicle at time a, a reachable position of the first vehicle at time a+1 is determined, and the aforementioned reachable position of the first vehicle at time a+1 is taken as a child node. The first device can repeat the above steps multiple times to sample multiple feasible accelerations of the first vehicle in the lateral and longitudinal directions, obtain multiple reachable positions of the first vehicle at time a+1, and take the multiple reachable positions of the first vehicle at time a+1 as multiple child nodes. The first device can evaluate each of the aforementioned multiple sub-nodes to obtain the cost value corresponding to each sub-node. Then, it selects the sub-node with the smallest cost value from the multiple sub-nodes as the starting node at time a+1 (which can also be understood as the starting position of the first vehicle at time a+1). Based on the position of the first vehicle at time a+1, the first device repeats the aforementioned operation to obtain a position selected from the multiple reachable positions of the first vehicle at time a+2, which is the starting position of the first vehicle at time a+2. By analogy, it can obtain the positions of the first vehicle at multiple times in the second time period after the current time, thereby obtaining a candidate planned trajectory of the first vehicle and the total cost value corresponding to the aforementioned candidate planned trajectory.

[0106] To more intuitively understand this solution, please refer to Figure 4, which is a schematic diagram of determining a candidate planning trajectory provided by an embodiment of this application. The candidate planning trajectory includes the positions of multiple times within a second time period after the current time. Figure 4 shows the process of determining two trajectory points at t=0.5s and t=1.0s, taking the time interval between two adjacent times in the aforementioned multiple times as an example of 0.5s. The first device can sample feasible accelerations in the horizontal and vertical directions. Based on the starting position of the first vehicle at t=0s and the sampled acceleration, multiple sub-nodes (i.e., multiple reachable positions) of the first vehicle at t=0.5s are determined. The multiple sub-nodes of the first vehicle at t=0.5s are evaluated to obtain the cost value corresponding to each sub-node. Then, the sub-node with the smallest cost value is selected from the multiple sub-nodes as the starting position at t=0.5s. The first device samples feasible accelerations of the first vehicle in the lateral and longitudinal directions again. Based on the starting position of the first vehicle at t = 0.5s and the sampled accelerations, multiple sub-nodes of the first vehicle at t = 1.0s are determined. The multiple sub-nodes of the first vehicle at t = 1.0s are evaluated to obtain the cost value corresponding to each sub-node. Then, the sub-node with the smallest cost value is selected from the multiple sub-nodes as the starting position at t = 1.0s. This process is repeated to obtain multiple positions included in the candidate planned trajectory. It should be understood that the example in Figure 4 is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0107] Optionally, the cost value corresponding to each child node can be obtained based on at least one of the following for each child node: heuristic cost, obstacle cost, lane centerline cost, driving cost, or other cost. Optionally, if the cost value corresponding to each child node is obtained based on at least two of the heuristic cost, obstacle cost, lane centerline cost, and driving cost, then the first device can perform a weighted summation of the aforementioned at least two costs corresponding to each child node to obtain the cost value corresponding to that child node.

[0108] For example, after determining the set of strategies corresponding to the second information, the first device can determine at least one target location point corresponding to each strategy in the aforementioned set of strategies, and can obtain multiple target location points corresponding to the set of strategies, wherein a target location point represents a location point that the first vehicle can reach at the end time of the second time period after the current time.

[0109] For example, the strategy set includes strategy 1 and strategy 2. Strategy 1 is to maintain straight ahead and accelerate to overtake, while strategy 2 is to maintain straight ahead and decelerate to yield. Target position points 1 and 2 correspond to strategy 1, and target position points 3 and 4 correspond to strategy 2. Target position point 1 represents the position that the first vehicle can reach at the end of the second time period if it maintains straight ahead and accelerates to overtake using acceleration 1. Target position point 2 represents the position that the first vehicle can reach at the end of the second time period if it maintains straight ahead and accelerates to overtake using acceleration 2. Target position point 3 represents the position that the first vehicle can reach at the end of the second time period if it maintains straight ahead and decelerates to yield using acceleration 3. Target position point 4 represents the position that the first vehicle can reach at the end of the second time period if it maintains straight ahead and decelerates to yield using acceleration 4. It should be understood that this example is only for the convenience of understanding the relationship between the strategies and target position points and is not intended to limit this scheme.

[0110] The first device can calculate the heuristic cost corresponding to each child node based on one of the target location points corresponding to all target location points in the policy set. To further understand this scheme, the formula for calculating the heuristic cost of each child node is disclosed below: C heuristic =k s ×ds 2 +k d ×dd 2 +k t ×dt

[0111] Among them, C heuristic k represents the heuristic cost corresponding to a child node. s d represents the vertical distance between the child node and a target location, dd represents the horizontal distance between the child node and the target location, dt represents the distance between the first time point corresponding to the child node and the end time of the second time period, the first time point corresponding to the child node indicates that the child node is the reachable position of the first vehicle at the first time point, and k represents the distance between the first vehicle and the target location. s k d and k t These represent three hyperparameters. It should be understood that this example is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0112] It should be noted that when determining the positions of multiple time points included in the same candidate predicted trajectory, it is necessary to generate the heuristic cost of the child nodes multiple times. The same target location point is used to calculate the heuristic cost in the process of determining the same candidate predicted trajectory.

[0113] For example, the first device can determine the obstacle cost corresponding to each child node based on the second information. To further understand this solution, the formula for calculating the obstacle cost of each child node is disclosed below:

[0114] Among them, C obs D represents the obstacle cost of a child node. obs p represents the distance between the child node and the nearest obstacle in the second information. obs D represents the probability of the nearest obstacle being present in the second information. For example, the second information can be represented as a grid map. obs This can represent the distance between the child node and the nearest grid cell occupied by an obstacle in the second information. The nearest grid cell can be either a grid cell occupied by the predicted trajectory of a dynamic obstacle or a grid cell occupied by a static obstacle. obs This represents the occupancy probability of the most recent grid cell. It should be understood that this example is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0115] For example, the first device can determine the lane centerline cost corresponding to each child node based on the distance between each child node and the lane centerline. To further understand this solution, the formula for calculating the obstacle cost of each child node is disclosed below: C lane =d lane 2

[0116] Among them, C lane d represents the lane centerline cost of a child node. lane This represents the distance between the child node and the lane centerline. It should be understood that this example is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0117] For example, since each sub-node is obtained based on the starting position of the first vehicle at a certain moment, the sampled lateral acceleration, and the sampled longitudinal acceleration, the lateral and longitudinal acceleration corresponding to each sub-node can be obtained. The first device can determine the driving cost corresponding to each sub-node based on the absolute value of the lateral acceleration and the absolute value of the longitudinal acceleration corresponding to each sub-node. Specifically, the larger the absolute value of the lateral acceleration corresponding to each sub-node, the larger the driving cost corresponding to that sub-node; the smaller the absolute value of the lateral acceleration corresponding to each sub-node, the smaller the driving cost corresponding to that sub-node. Similarly, the larger the absolute value of the longitudinal acceleration corresponding to each sub-node, the larger the driving cost corresponding to that sub-node; the smaller the absolute value of the longitudinal acceleration corresponding to each sub-node, the smaller the driving cost corresponding to that sub-node.

[0118] Optionally, after obtaining the cost value of each child node among multiple child nodes, the first device can store each child node and its corresponding cost value in an open list. The multiple child nodes are then sorted based on their cost values. Using an open list to store multiple child nodes and their corresponding cost values ​​improves the efficiency of the process of "searching for the child node with the smallest cost value" and ensures that high-quality child nodes (i.e., the child node with the smallest cost value) are searched first. It should be noted that the aforementioned open list can also be replaced with an array, linked list, or other data storage structure, depending on the specific application scenario.

[0119] Optionally, at least one candidate planning trajectory corresponding to the second information corresponds one-to-one with at least one target location point.

[0120] Optionally, the first device selects a second planning trajectory from at least one candidate planning trajectory corresponding to each second piece of information. This may include: the first device selects the second planning trajectory with the lowest total value from at least one candidate planning trajectory that corresponds one-to-one with at least one target location point, which is to obtain the second planning trajectory corresponding to a second piece of information.

[0121] Alternatively, the first device may randomly select a candidate planning trajectory from at least one candidate planning trajectory, and use the randomly selected candidate planning trajectory as the second planning trajectory, etc. The specific implementation method can be determined in combination with the actual application scenario.

[0122] It should be noted that the above description of the specific implementation process of the two steps of "obtaining at least one candidate planning trajectory corresponding to each second information" and "selecting a second planning trajectory from at least one candidate planning trajectory corresponding to each second information" using the preset algorithm as a random sampling algorithm is only to prove the feasibility of this solution. In this application embodiment, the specific implementation of the preset algorithm when it is other types of algorithms will not be described in detail.

[0123] Optionally, in the process of generating at least one candidate planning trajectory corresponding to at least one target location point, the generation process of candidate planning trajectories corresponding to different target location points can be parallel, which is beneficial to improving the efficiency of the overall acquisition process of at least one candidate planning trajectory corresponding to at least one target location point, and thus beneficial to improving the efficiency of the overall acquisition process of the second planning trajectory corresponding to each second information, and beneficial to improving the efficiency of trajectory planning for vehicles.

[0124] The first device performs the above operations based on each piece of second information and the strategy set corresponding to each piece of second information, thereby obtaining a second planned trajectory corresponding to each of the at least two pieces of second information. Optionally, in the process of obtaining all second planned trajectories corresponding to at least two pieces of second information, the process of obtaining second planned trajectories corresponding to different pieces of second information can be parallelized to improve the efficiency of obtaining all second planned trajectories corresponding to at least two pieces of second information, which is beneficial to improving the efficiency of trajectory planning for vehicles.

[0125] Regarding the specific implementation process of the first device obtaining the reference trajectory based on all the second planned trajectories corresponding to at least two second pieces of information, for example, the second planned trajectory and the first planned trajectory of the first vehicle both include the trajectory of the first vehicle from the current time to the first time point, and the reference trajectory includes the trajectory of the vehicle from the current time to the second time point. The determination method of the reference trajectory will be introduced in three cases below.

[0126] Scenario 1:

[0127] For example, the second time point is the same as the first time point. The first device performs a fusion operation to obtain a reference trajectory based on the second planned trajectory corresponding to at least two pieces of second information. This may include: the first device directly fuses all the second planned trajectories corresponding to at least two pieces of second information to obtain the reference trajectory.

[0128] Scenario 2:

[0129] For example, the second time point is earlier than the first time point. The first device performs a fusion operation to obtain a reference trajectory based on the second planned trajectory corresponding to at least two pieces of second information. This may include: the first device determining the second time point, and then fusing the sub-planned trajectories in each of the second planned trajectories corresponding to at least two pieces of second information to obtain the reference trajectory. The sub-planned trajectories in each second planned trajectory include the trajectory from the current time to the second time point in each second planned trajectory.

[0130] For example, all second planning trajectories corresponding to at least two second pieces of information include second planning trajectory 1, second planning trajectory 2, second planning trajectory 3, and second planning trajectory 4. The first time point is the 5th second after the current time, and the second time point is the 3rd second after the current time. Then, all sub-planning trajectories corresponding to at least two pieces of second information can include sub-planning trajectory 1, sub-planning trajectory 2, sub-planning trajectory 3, and sub-planning trajectory 4. Sub-planning trajectory 1 includes the trajectory within the next 3 seconds after the current time in second planning trajectory 1, sub-planning trajectory 2 includes the trajectory within the next 3 seconds after the current time in second planning trajectory 2, sub-planning trajectory 3 includes the trajectory within the next 3 seconds after the current time in second planning trajectory 3, and sub-planning trajectory 4 includes the trajectory within the next 3 seconds after the current time in second planning trajectory 4. The reference trajectory is obtained by fusing sub-planning trajectory 1, sub-planning trajectory 2, sub-planning trajectory 3, and sub-planning trajectory 4. It should be understood that the example here is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0131] For example, the first device may fuse the sub-planning trajectories in each of the second planning trajectories corresponding to at least two pieces of second information. This fusion may involve the first device employing a probabilistic fusion method to fuse all sub-planning trajectories corresponding to at least two pieces of second information. Different sub-planning trajectories among the all sub-planning trajectories corresponding to at least two pieces of second information may be assigned different probability values ​​or the same probability value, depending on the specific application scenario.

[0132] Optionally, the probability value of each sub-planning trajectory can be the probability value of the second information corresponding to the sub-planning trajectory. The probability value of the second information represents the probability value of the predicted trajectory appearing in the second information. If the second information includes only one predicted trajectory, the probability value of the second information represents the probability value of the aforementioned predicted trajectory. If the second information includes at least two predicted trajectories, the probability value of the second information can be obtained by weighted summation of the probability values ​​of the at least two predicted trajectories.

[0133] To better understand this solution, please refer to Figure 5. Figure 5 is a schematic diagram of a method for obtaining a reference trajectory by performing a fusion operation based on a second planning trajectory corresponding to at least two pieces of second information, according to an embodiment of this application. Figure 5 includes three sub-schematic diagrams: left, middle, and right. The left sub-schematic diagram of Figure 5 shows two second planning trajectories corresponding to at least two pieces of second information. The middle sub-schematic diagram of Figure 5 shows a segment of a sub-planning trajectory extracted from each of the two second planning trajectories to obtain two sub-planning trajectories. The reference trajectory is obtained by probabilistically fusing the two sub-planning trajectories. P1 and P2 in Figure 5 represent the probabilities of the two sub-planning trajectories, respectively. The right sub-schematic diagram of Figure 5 shows that each second planning trajectory includes the trajectory from the current time to the first time point, and the reference trajectory includes the trajectory from the current time to the second time point. The second time point is earlier than the first time point. It should be understood that the examples in Figure 5 are only for the convenience of understanding this solution and are not intended to limit this solution.

[0134] Optionally, the first device determining the second time point may include: the first device fusing all second planned trajectories corresponding to at least two pieces of second information to obtain a fused trajectory; for example, the aforementioned fusion method may be probabilistic fusion. Then, the first device can determine the collision time point where the fused trajectory collides with obstacles around the vehicle based on at least two pieces of second information and the fused trajectory. The obstacles around the vehicle include at least one of the aforementioned first obstacles; optionally, the obstacles around the vehicle also include static obstacles; the aforementioned collision time point is used to determine the second time point, and the second time point needs to be earlier than the collision time point.

[0135] Optionally, the factors for determining the second time point also include: the time point where the first distance is greater than or equal to the first preset distance, and / or the time point where the second distance is greater than or equal to the second preset distance; the first distance includes the maximum distance between different second planning trajectories among all second planning trajectories, and the second distance includes the maximum distance between different predicted trajectories among at least two predicted trajectories of the same second obstacle.

[0136] To further understand this scheme, the process of determining the second time point is described below with reference to the formula. For example, the first device performs probabilistic fusion on all the second planned trajectories corresponding to the at least two pieces of second information to obtain the fused trajectory; in the third information in the form of a grid map, the collision time point T between the fused trajectory and the obstacles around the vehicle is determined. collision Therefore, based on the fused trajectory at the collision time point T collision The speed of the obstacle that collides with the trajectory of the first vehicle after merging with its own at the time of collision T. collision Based on the predicted velocity at the location and the collision time, determine the latest time. The second time point needs to be earlier than the aforementioned latest time. To avoid collisions.

[0137] Since each second planned trajectory includes trajectory information from multiple times within a second time period after the current time, the first device can also sequentially acquire the first distance corresponding to each of the aforementioned multiple times, determine whether the first distance is greater than or equal to a first preset distance, and acquire the time point when the first distance is greater than or equal to the first preset distance.

[0138] For example, in, State represents the maximum distance at time t among all different second planning trajectories corresponding to at least two pieces of second information, where time t is a moment within the second time interval following the current moment. thr This represents the first preset distance. It should be understood that this example is only for the convenience of understanding this scheme.

[0139] The first device can also acquire the maximum distance (i.e., the second distance) between at least two different predicted trajectories of each of the at least one second obstacle. When it is determined that the maximum distance between the different predicted trajectories of any second obstacle is greater than or equal to a second preset distance, the device acquires the time point at which the second distance is greater than or equal to the second preset distance.

[0140] For example, in, The maximum distance between at least two different predicted trajectories representing the same second obstacle at time t, where time t is a moment in the second time interval after the current moment. thr This represents the second preset distance. It should be understood that this example is only for the convenience of understanding this scheme.

[0141] The first device can be based on the latest time. The time point when the first distance is greater than or equal to the first preset distance and the time point when the second distance is greater than or equal to the second preset distance. Determine the second time point; to further understand this scheme, the formula used to determine the second time point can be as follows: stk geo +k object =1

[0142] Where, k geo and k object There are two hyperparameters. It should be understood that the examples in the above formulas are only for the convenience of understanding this scheme; for example, the first device can also be based on the collision time point T. collisionThe latest time determined Confirmed as the second time point; for example, the first device can also... This is determined to be the second time point, which is also the latest time. The time point when the first distance is greater than or equal to the first preset distance The minimum value between these two points is determined as the second time; for example, the first device can also... This is determined to be the latest time point, which is also the latest time. The time point when the second distance is greater than or equal to the second preset distance The minimum value between them is determined as the second time, etc., and the specific value can be determined based on the actual application scenario.

[0143] In this embodiment, all second planned trajectories are fused to obtain a fused trajectory. The collision time point between the fused trajectory and obstacles around the vehicle is determined, and this second time point is set earlier than the collision time point. Then, the sub-planned trajectories in each second planned trajectory are fused to obtain a reference trajectory. Each sub-planned trajectory includes the trajectory from the current time to the second time point, thereby ensuring the safety of the reference trajectory. Each first planned trajectory needs to fit the reference trajectory, which also improves the safety of each first planned trajectory. This is beneficial to improving the safety of the final planned trajectory of the first vehicle, and thus improving the safety of the vehicle driving process.

[0144] If the second time point is set too early, meaning the time between the current moment and the second time point is too short, the predicted trajectories of obstacles around the vehicle will not show a clear trend. This results in high similarity between the sub-planned trajectories corresponding to different second information, making the fused reference trajectory less meaningful. The vehicle's trajectory planning process still cannot fully utilize the predicted trajectories of different obstacles in the surrounding environment. Conversely, if the second time point is set too late, meaning the time between the current moment and the second time point is too long, the obtained reference trajectory will over-consider the influence of the predicted trajectories of each obstacle around the vehicle. Influenced by the reference trajectory, each first-planned trajectory will be overly conservative, and the changes between the trajectory before and after the second time point in each first-planned trajectory may be significant. This leads to greater acceleration of the vehicle when executing the trajectory after the second time point, making it impossible to guarantee the feasibility of the final determined driving trajectory. In this application, the factors for determining the second time point also include: the time point at which the maximum distance between different second planned trajectories in all second planned trajectories is greater than or equal to the first preset distance, thereby helping to avoid excessive changes between the trajectory before and after the second time point in each first planned trajectory; and / or, the time point at which the maximum distance between different predicted trajectories in at least two predicted trajectories of the same second obstacle is greater than or equal to the second preset distance, thereby helping to avoid setting the second time point too early. Therefore, the second time point determined by considering various factors, and the reference trajectory determined based on the second time point guiding the final determined driving trajectory, is beneficial to ensuring the safety of the final determined driving trajectory, avoiding the final determined driving trajectory from being too conservative, and improving the smoothness of the final determined driving trajectory.

[0145] To more intuitively understand the aforementioned concepts, please refer to Figure 6. Figure 6 is a schematic diagram of the first planned trajectory determined based on reference trajectories determined at different second time points according to an embodiment of this application. Figure 6 shows the situations when using second time point 1, second time point 2, and second time point 3, respectively. Because second time point 1 is too early, the vehicle's trajectory planning process cannot fully utilize the predicted trajectories of different obstacles in the surrounding environment, and the vehicle's planned trajectory will prematurely rely on only the predicted trajectory in one second piece of information. Because second time point 3 is too late, the vehicle's acceleration is too large when executing the trajectory after the second time point. When using second time point 2, it is beneficial to ensure the safety of the finally determined driving trajectory while also improving the smoothness of the finally determined driving trajectory. It should be understood that the examples in Figure 6 are only for the convenience of understanding this solution and are not intended to limit this solution.

[0146] Scenario 3:

[0147] For example, the first device can determine the planned trajectory of the first vehicle every preset time interval. The first device can also obtain a first reference trajectory generated during the previous determination of the planned trajectory of the first vehicle. The first reference trajectory can also be called a historical reference trajectory. The reference trajectory determined through the above-mentioned situation 1 or situation 2 is determined as the second reference trajectory. The first reference trajectory and the second reference trajectory are fused to obtain the final reference trajectory. Since the historical reference trajectory generated during the previous determination of the planned trajectory of the first vehicle is also fused when generating the final reference trajectory, excessive jumps between two adjacent planned trajectories of the first vehicle are avoided, ensuring the continuity between different planned trajectories of the first vehicle, which is conducive to improving the comfort of the first vehicle during driving.

[0148] To further understand this scheme, the formula for fusing the first and second reference trajectories to obtain the final reference trajectory is as follows:

[0149] in, This represents the reference trajectory that the first device finally obtains when generating the planned trajectory of the first vehicle in the current (i.e., the i-th) iteration. P(i-1) represents the first reference trajectory generated by the first device during the (i-1)th determination of the planned trajectory of the first vehicle. The weight parameters, P(i) represents the second reference trajectory generated by the first device in the current (i.e., the i-th) generation of the planned trajectory of the first vehicle through either case 1 or case 2 above. The weight parameters; optionally, P(i-1) is less than P(i). It should be understood that the example here is only for the convenience of understanding this scheme and is not intended to limit this scheme.

[0150] Regarding the specific implementation process of the first device generating a first planned trajectory for the first vehicle corresponding to each piece of second information based on each piece of second information and a reference trajectory, in one implementation, the first device can optimize the second planned trajectory corresponding to each piece of second information based on each piece of second information and the reference trajectory to obtain the first planned trajectory for the first vehicle corresponding to each piece of second information. The objective of generating the first planned trajectory includes ensuring that the first planned trajectory conforms to the reference trajectory; optionally, factors considered when generating the first planned trajectory may also include driving risk and / or driving comfort.

[0151] For example, the first device can input any second piece of information, a reference trajectory, and a second planned trajectory corresponding to the second piece of information (hereinafter referred to as the "target second planned trajectory") into the optimizer to obtain the first planned trajectory corresponding to the second piece of information output by the optimizer (hereinafter referred to as the "target first planned trajectory"). Optionally, the input of the optimizer may also include other second planned trajectories besides the target second planned trajectory among all the second planned trajectories corresponding to at least two pieces of information. For example, the optimizer may employ an L-BFGS optimizer, a cplex optimizer, or other types of optimizers. The L-BFGS optimizer is an optimization algorithm using limited memory, and the cplex optimizer is a mathematical optimization algorithm. The specific optimization algorithm can be determined based on the actual application scenario. The first device can perform the aforementioned operation for each of the at least two pieces of information to obtain the first planned trajectory corresponding to each of the at least two pieces of information.

[0152] To further understand this scheme, the process of generating the first planned trajectory corresponding to each piece of second information is described in detail below, using formulas. Here, we take the use of B-spline curves to represent the second planned trajectory of the first vehicle as an example. The second planned trajectory of the first vehicle can be represented as follows: Where 3≤d≤n, the second planned trajectory may include multiple trajectory points. This can be achieved by analyzing multiple trajectory points included in the second planned trajectory. The optimization method is used to optimize the entire second planning trajectory. The objective of the aforementioned optimization process can be expressed by the following formula:

[0153] Where J represents the total cost of the first target trajectory (i.e., the trajectory obtained after optimizing the second target trajectory); C swarm The aggregate cost representing the first planned trajectory of the target is used to reflect the fit between the first planned trajectory of the target and the reference trajectory; C obs C represents the driving risk cost of the first planned trajectory towards the target. smooth This represents the comfort cost of the first planned trajectory representing the target.

[0154] Where, p k p represents the probability value of the predicted trajectory appearing in the second information corresponding to the target second planned trajectory. If the second information includes only one predicted trajectory, then p k p represents the probability value of one of the aforementioned predicted trajectories. If the second information includes at least two predicted trajectories, then the probability values ​​of these at least two predicted trajectories can be weighted and summed to obtain p. k dis ke represents the distance between the trajectory point at time t in the second planned trajectory of the target and the trajectory point at time t in the reference trajectory. t As a hyperparameter, dis o c represents the distance between the trajectory point before optimization and the trajectory point after optimization at time t, which is also the distance between the trajectory point at time t in the second planned trajectory of the target and the trajectory point at time t in the first planned trajectory of the target. c and b1 are both hyperparameters.

[0155] Among them, C obs D represents the driving risk cost of the first planned trajectory towards the target. obs p represents the distance between a trajectory point in the first planned trajectory of the target and the nearest obstacle in the second information. This nearest obstacle can be a predicted trajectory of a dynamic obstacle or a static obstacle. obs b2 represents the probability of the nearest obstacle appearing. b2 is a hyperparameter. The smaller the distance between the trajectory point in the first planned trajectory and the nearest obstacle in the second information, the greater the driving risk cost of the first planned trajectory. obs The larger the value, the better.

[0156] Among them, C smooth The comfort cost of representing the primary planning trajectory towards the target. C represents the rate of change of acceleration corresponding to each trajectory point in the first planned trajectory of the target. The larger the rate of change of acceleration, the lower the comfort level of the first planned trajectory of the target. smooth The larger the value of C, the smaller the rate of change of the aforementioned acceleration, and the higher the comfort level of the first planned trajectory of the target. smooth The smaller the value, the better.

[0157] The first device can input the second information, the reference trajectory, the target second planned trajectory, and the above formula into the optimizer to obtain the optimized planned trajectory (i.e., the target first planned trajectory) output by the optimizer. It should be noted that the examples in the above formula are only to prove the feasibility of this solution. For example, the comfort cost of the target first planned trajectory can also be obtained by calculating the acceleration corresponding to each trajectory point. The specific formula used can be determined in combination with the actual application scenario, and is not limited here.

[0158] In another implementation, after obtaining the reference trajectory, the first device can also re-plan the trajectory of the first vehicle using a preset algorithm based on each second piece of information, the strategy set corresponding to each second piece of information, and the reference trajectory, to obtain at least one re-planned candidate trajectory corresponding to each second piece of information. In the aforementioned process of re-planning the trajectory of the first vehicle, the constraint that each candidate trajectory needs to conform to the reference trajectory is added. The first device can select the first planned trajectory corresponding to each second piece of information from the at least one re-planned candidate trajectory corresponding to each second piece of information.

[0159] In this embodiment, a second planning trajectory corresponding to each second piece of information is first determined based on each second piece of information and the strategy set corresponding to each second piece of information. A reference trajectory is obtained by performing a fusion operation based on the second planning trajectories corresponding to at least two second pieces of information. A first planning trajectory corresponding to each second piece of information is generated based on the second information and the reference trajectory. Then, the final planning trajectory obtained by the first vehicle is determined based on all the first planning trajectories corresponding to at least two second pieces of information. Since the reference trajectory is obtained by performing a fusion operation on the second planning trajectories corresponding to at least two second pieces of information, that is, the reference trajectory is obtained based on all the second information in at least two pieces of information, and the goal of generating the first planning trajectory includes that the first planning trajectory needs to fit the reference trajectory, the planning trajectory finally determined by the first vehicle takes into account all the second information, thereby improving the safety of the final determined planning trajectory. Moreover, each first planning trajectory only needs to fit the reference trajectory, rather than using the reference trajectory as a hard constraint, which avoids the determination process of the final planning trajectory obtained by the first vehicle being overly conservative, which is conducive to improving the efficiency of the first vehicle's driving process.

[0160] Furthermore, the reference trajectory includes the trajectory from the current time to the second time point, and the first planned trajectory includes the trajectory from the current time to the first time point. If the second time point and the first time point are the same, then the entire planned trajectory in each first planned trajectory needs to fit the reference trajectory, which will greatly limit the location points that the first vehicle can explore during the trajectory planning process. Setting the second time point earlier than the first time point is beneficial to exploring more location points when planning the trajectory for the first vehicle, which is conducive to obtaining a better planned trajectory and thus improving the user experience for the driver and passengers.

[0161] In another implementation, the first device performs trajectory planning for the first vehicle using a preset algorithm based on each second piece of information and the strategy set corresponding to each second piece of information. After obtaining at least one candidate planned trajectory corresponding to each second piece of information, the first planned trajectory of the first vehicle can also be directly selected from the at least one candidate planned trajectory corresponding to each second piece of information.

[0162] Regarding the specific implementation method of determining the final planned trajectory of the first vehicle based on all first planned trajectories corresponding to at least two second pieces of information by the first device, in one implementation method, the first device can obtain the score value of each first planned trajectory among all first planned trajectories corresponding to at least two second pieces of information; and select the planned trajectory of the first vehicle from all first planned trajectories corresponding to at least two second pieces of information based on the score value of each first planned trajectory.

[0163] Optionally, the first device can obtain the total value of the first planning trajectory corresponding to each second piece of information generated in the above steps. For example, the total value of each first planning trajectory can be J as described above. The first device generates a score value for each first planning trajectory based on the total value of each first planning trajectory and the probability value of the second piece of information corresponding to each first planning trajectory. Optionally, the higher the total value of each first planning trajectory, the lower the score value of the first planning trajectory can be; the higher the probability value of the second piece of information corresponding to each first planning trajectory, the higher the score value of the first planning trajectory can be. The first device selects the first planning trajectory with the highest score value from all first planning trajectories corresponding to at least two pieces of second information, and determines the first planning trajectory with the highest score value as the planning trajectory of the first vehicle.

[0164] Optionally, the determination factor for the score of each first planned trajectory also includes the degree of similarity between each first planned trajectory and the previously determined planned trajectory of the first vehicle. Specifically, the more similar each first planned trajectory is to the previously determined planned trajectory of the first vehicle, the higher the score of that first planned trajectory can be; conversely, the less similar each first planned trajectory is to the previously determined planned trajectory of the first vehicle, the lower the score of that first planned trajectory can be.

[0165] For example, the first device can calculate the jump cost of each first planned trajectory, and update the total value of each first planned trajectory using the jump cost of each first planned trajectory to obtain the updated total value of each first planned trajectory; wherein, the lower the similarity between each first planned trajectory and the previously determined planned trajectory of the first vehicle, the higher the jump cost of each first planned trajectory; the first device can generate a score value for each first planned trajectory based on the updated total value of each first planned trajectory and the probability value of the second information corresponding to each first planned trajectory. The formula for calculating the jump cost of each first planned trajectory can be as follows:

[0166] in, C represents the distance between the trajectory point of the first planned trajectory at time t and the trajectory point of the previously determined planned trajectory of the first vehicle at time t. The larger the distance, the lower the similarity between the first planned trajectory and the previously determined planned trajectory of the first vehicle. consist The larger the value, the better. It should be understood that the example here is only for the convenience of understanding this scheme.

[0167] In this embodiment, the total cost of executing each first planned trajectory and the probability value of the second information corresponding to each first planned trajectory are comprehensively considered to obtain a score value for each first planned trajectory. Then, based on the score value of each first planned trajectory, the planned trajectory of the first vehicle is selected from multiple first planned trajectories. As the predicted trajectories of dynamic obstacles in the surrounding environment become increasingly clear, the aforementioned method is more conducive to selecting a planned trajectory determined based on a predicted trajectory with a higher probability value. Therefore, when the first vehicle travels along this planned trajectory, it is beneficial to avoid dynamic obstacles, thus improving the safety of the first vehicle's driving process. Furthermore, the travel cost of this planned trajectory is low, reflecting the safety and / or comfort of the planned trajectory, thereby improving the overall evaluation of the first vehicle's driving process and enhancing the user experience. Optionally, the jump cost of each first planned trajectory is also considered when generating the total cost of each first planned trajectory. This helps to improve the similarity between the final planned trajectory and the previously determined planned trajectory, thereby avoiding excessive jumps between adjacent planned routes and improving the continuity of the first vehicle's driving process.

[0168] In another implementation, the first device may also perform a fusion operation based on all the first planned trajectories corresponding to at least two second pieces of information to obtain the main branch trajectory in the planned trajectory of the first vehicle; wherein each first planned trajectory includes the trajectory of the first vehicle from the current time to the first time point, and the main branch trajectory may include the trajectory of the first vehicle from the current time to the third time point; the planned trajectory of the first vehicle also includes multiple branches corresponding one-to-one with all the first planned trajectories, and each branch includes the trajectory from the third time point to the first time point in a first planned trajectory.

[0169] The specific implementation of "the first vehicle performs a fusion operation to obtain the main branch trajectory based on all the first planned trajectories corresponding to at least two second pieces of information" can be found in the description of "the first vehicle performs a fusion operation to obtain the reference trajectory based on all the second planned trajectories corresponding to at least two second pieces of information". The difference is that the second planned trajectory in the above description is replaced with the first planned trajectory, the second time point in the above description is replaced with the third time point, and the reference trajectory in the above description is replaced with the main branch trajectory. It will not be elaborated here.

[0170] In another implementation, the first device can also perform a weighted summation of all the first planned trajectories corresponding to at least two second pieces of information to obtain the planned trajectory of the first vehicle, etc. The specific implementation method can be determined in combination with the actual application scenario.

[0171] Optionally, the first device can generate the planned trajectory of the first vehicle every preset time interval. It should be noted that the second time point (optionally, also including the third time point) is regenerated each time the planned trajectory of the first vehicle is generated, that is, the second time point (optionally including the third time point) is dynamically updated.

[0172] Optionally, the first device can also perform a weighted fusion of the previously generated planned trajectory of the first vehicle and the currently generated planned trajectory of the first vehicle to obtain the final planned trajectory of the first vehicle in this iteration, thereby avoiding excessive jumps between adjacent planned routes and improving the continuity of the first vehicle's driving process.

[0173] Optionally, step 104 further includes: after determining the final planned trajectory of the first vehicle, the first device may also determine decision information corresponding to the final planned trajectory of the first vehicle, the aforementioned decision information including the driving strategy of the first vehicle in the lateral and / or longitudinal directions.

[0174] For example, the first device can determine whether the first vehicle has a lateral strategy of going left, straight, or right based on the final planned trajectory of the first vehicle, and can determine whether the first vehicle has a longitudinal strategy of cutting in or yielding based on the planned trajectory of the first vehicle and the predicted trajectory of the surrounding dynamic obstacles, thereby obtaining the driving strategy of the first vehicle in the lateral and / or longitudinal directions.

[0175] In this embodiment, not only is the planned trajectory of the first vehicle determined, but also the driving strategy of the vehicle in the lateral and / or longitudinal directions, that is, more information used in the driving process is generated, which not only improves the interpretability of this solution, but also improves the efficiency of this solution in the dimension of information generation.

[0176] Alternatively, step 104 may also include: the first device inputting each piece of second information and the set of policies corresponding to each piece of second information into a machine learning model to obtain the planned trajectory of the first vehicle and / or the decision information of the first vehicle output by the machine learning model. The machine learning model can be a model that has undergone training operations; for example, the machine learning model can be a convolutional neural network, a support vector machine, an attention-based neural network, or other types of machine learning models.

[0177] In this embodiment, since the first information includes information on at least two predicted trajectories of the second obstacle, more predicted trajectories of the obstacle are considered in the process of determining the vehicle's planned trajectory and / or decision information, rather than only considering the predicted trajectory with the highest probability for each obstacle. This helps to reduce the risk of the planned trajectory. In addition, instead of directly taking the positions occupied by all the predicted trajectories of the obstacle as the positions where the vehicle cannot drive, the planned trajectory is not too conservative. Instead, a feasible driving strategy for the vehicle is determined for each piece of second information, and then the vehicle's planned trajectory and / or decision information are comprehensively determined based on the strategy set corresponding to each piece of second information. That is, it fully considers the multiple possible predicted trajectories of the obstacle and avoids the planned trajectory being too conservative. This not only improves the safety of the planned trajectory but also minimizes the reduction in traffic efficiency.

[0178] To more intuitively understand the beneficial effects of the method provided in this application, experiments were conducted in two different traffic scenarios, which are described below. In the narrow passage traffic scenario, the social vehicles interacting with the first vehicle have two driving intentions: accelerating to pass and decelerating to yield. These two driving intentions correspond to two different predicted trajectories of the social vehicles. Correspondingly, if the social vehicle accelerates to pass, the first vehicle needs to decelerate to yield; if the social vehicle decelerates to yield, the first vehicle needs to accelerate to pass.

[0179] If the driving intention of the other vehicle changes during its journey, such as from accelerating to yielding or vice versa, it will affect the first vehicle. If only the predicted trajectory with the highest probability of the other vehicle is considered for planning the first vehicle's trajectory, when the other vehicle's driving intention changes during its journey, the first vehicle will need to accelerate rapidly or brake suddenly, which is not only unsafe but also results in a poor driving experience. The trajectory planned using the method provided in this application is as follows: the first vehicle first accelerates tentatively, and after the other vehicle shows a clear reaction, the first vehicle then determines whether to slow down to yield or accelerate to pass. This ensures safety, improves the efficiency of the driving process, and enhances the driving experience.

[0180] This application also conducted experiments in a traffic scenario where vehicles pass through an intersection. Referring first to Figure 7, Figure 7 is a schematic diagram of a traffic scenario where vehicles pass through an intersection provided by an embodiment of this application. Figure 7 includes two sub-schematic diagrams, left and right. In the left sub-schematic diagram of Figure 7, the first vehicle is circled by a black ellipse, and in the right sub-schematic diagram of Figure 7, the first vehicle is circled by a white ellipse. The left sub-schematic diagram of Figure 7 uses a two-dimensional method to show the traffic situation at the intersection where the vehicle passes through. The dashed lines in the left sub-schematic diagram represent the predicted trajectory of dynamic obstacles. The right sub-schematic diagram of Figure 7 uses a grid map to show the grid cells occupied by dynamic obstacles at the intersection. Different shades of gray are used to represent the probability of a grid cell being occupied. The darker the color, the greater the probability of a grid cell being occupied. It should be understood that the example in Figure 7 is only for the convenience of understanding this solution.

[0181] The trajectory planned using the method provided in this application is as follows: On the opposite lane of the first vehicle, there are vehicles and bicycles traveling straight and needing to cross the intersection; at the far end of the first vehicle, pedestrians are crossing the intersection. Influenced by the oncoming vehicles, the first vehicle first maintains a low speed and waits for them to pass, then continues to wait for the bicycles to pass. At t = 4.8s, the first vehicle accelerates and passes the pedestrian crossing ahead, completing the left turn. Please refer to Figure 8, which is a schematic diagram of the speed and acceleration changes of the first vehicle when crossing the intersection according to an embodiment of this application. As shown in Figure 8, the speed of the first vehicle continuously increases, while the acceleration of the first vehicle does not change significantly, ensuring the comfort and safety of the first vehicle during the intersection crossing process. It should be understood that the example in Figure 8 is only for the convenience of understanding this solution.

[0182] Based on the embodiments corresponding to Figures 1 to 8, in order to better implement the above-described solutions of the embodiments of this application, related equipment for implementing the above-described solutions is also provided below. Specifically, refer to Figure 9, which is a structural schematic diagram of an intelligent driving device provided in an embodiment of this application. The intelligent driving device 900 includes: an acquisition module 901, configured to acquire first information corresponding to at least one first obstacle around the vehicle, the first information including information on at least one predicted trajectory of each first obstacle, and if a second obstacle exists among the at least one first obstacle, the first information including information on at least two predicted trajectories of the second obstacle; a determination module 902, configured to determine at least two pieces of second information based on the first information, wherein the second information indicates the position of the predicted trajectory, and different pieces of second information indicate the positions of different predicted trajectories in the predicted trajectories corresponding to the first information; the acquisition module 901 is further configured to acquire a strategy set corresponding to each piece of second information, the strategy set corresponding to each piece of second information including at least one feasible driving strategy for the vehicle in the lateral and / or longitudinal directions; the determination module 902 is further configured to determine the planned trajectory of the vehicle and / or the decision information of the vehicle based on each piece of second information and the strategy set corresponding to each piece of second information, the decision information including the driving strategy of the vehicle in the lateral and / or longitudinal directions.

[0183] Optionally, the determining module 902 is specifically used to: determine the first planned trajectory of the vehicle corresponding to each second piece of information based on each second piece of information and the strategy set corresponding to each second piece of information; and determine the planned trajectory of the vehicle based on all the first planned trajectories corresponding to at least two second pieces of information.

[0184] Optionally, the determining module 902 is also used to determine decision information corresponding to the planned trajectory of the vehicle.

[0185] Optionally, the determining module 902 is specifically used to: determine a second planned trajectory for the vehicle corresponding to each second piece of information based on each second piece of information and the strategy set corresponding to each second piece of information; perform a fusion operation to obtain a reference trajectory based on the second planned trajectories corresponding to at least two second pieces of information; and generate a first planned trajectory for the vehicle corresponding to each second piece of information based on each second piece of information and the reference trajectory, wherein the objective of generating the first planned trajectory includes the first planned trajectory conforming to the reference trajectory.

[0186] Optionally, both the first and second planned trajectories include the vehicle's trajectory from the current time to the first time point, and the reference trajectory includes the vehicle's trajectory from the current time to the second time point, where the second time point is earlier than the first time point.

[0187] Optionally, the determining module 902 is specifically used for: fusing all second planning trajectories corresponding to at least two second pieces of information to obtain a fused trajectory; determining the collision time point when the fused trajectory collides with obstacles around the vehicle based on at least two pieces of second information and the fused trajectory, wherein the obstacles around the vehicle include at least one first obstacle, and the second time point is earlier than the collision time point; and fusing the sub-planning trajectories in each of the second planning trajectories to obtain a reference trajectory, wherein the sub-planning trajectories include the trajectories in the second planning trajectories from the current time to the second time point.

[0188] Optionally, the factors for determining the second time point may further include: a time point where the first distance is greater than or equal to the first preset distance, and / or a time point where the second distance is greater than or equal to the second preset distance, wherein the first distance includes the maximum distance between different second planning trajectories among all second planning trajectories, and the second distance includes the maximum distance between different predicted trajectories among at least two predicted trajectories of the same second obstacle.

[0189] Optionally, the first obstacle is selected from multiple obstacles around the vehicle based on the vehicle's third planned trajectory. The vehicle's planned trajectory is determined every preset time interval. The vehicle's third planned trajectory is obtained based on the previously determined vehicle's planned trajectory. The selected first obstacle satisfies at least one of the following conditions: the predicted trajectory of the first obstacle intersects with the vehicle's third planned trajectory, or the distance between the predicted trajectory of the first obstacle and the vehicle's third planned trajectory is less than or equal to a preset distance.

[0190] Optionally, the second information is a raster map, indicating the raster occupied by the predicted trajectory.

[0191] It should be noted that the information interaction and execution process between the modules / units in the intelligent driving device 900 are based on the same concept as the various method embodiments corresponding to Figures 1 to 8 in this application. For details, please refer to the description in the method embodiments shown above in this application, which will not be repeated here.

[0192] The following describes a device provided in an embodiment of this application. Please refer to Figure 10, which is a structural schematic diagram of the device provided in an embodiment of this application. Specifically, the device 1000 includes: a receiver 1001, a transmitter 1002, a processor 1003, and a memory 1004 (wherein the device 1000 may have one or more processors 1003; Figure 10 shows one processor as an example). The processor 1003 may include an application processor 10031 and a communication processor 10032. In some embodiments of this application, the receiver 1001, transmitter 1002, processor 1003, and memory 1004 may be connected via a bus or other means.

[0193] Memory 1004 may include read-only memory and random access memory, and provides instructions and data to processor 1003. A portion of memory 1004 may also include non-volatile random access memory (NVRAM). Memory 1004 stores processor and operation instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof, wherein the operation instructions may include various operation instructions for implementing various operations.

[0194] Processor 1003 controls the operation of the device. In specific applications, the various components of the device are coupled together through a bus system, which may include not only the data bus, but also power buses, control buses, and status signal buses. However, for clarity, all buses in the diagram are referred to as the bus system.

[0195] The methods disclosed in the embodiments of this application can be applied to or implemented by the processor 1003. The processor 1003 can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by the integrated logic circuits in the hardware of the processor 1003 or by instructions in software form. The processor 1003 can be a general-purpose processor, a digital signal processor (DSP), a microprocessor, or a microcontroller, and may further include 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. The processor 1003 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 methods 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 art, 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 1004. Processor 1003 reads the information in memory 1004 and, in conjunction with its hardware, completes the steps of the above method.

[0196] Receiver 1001 can be used to receive input digital or character information, and to generate signal inputs related to device settings and function control. Transmitter 1002 can be used to output digital or character information through the first interface; transmitter 1002 can also be used to send instructions to the disk group through the first interface to modify the data in the disk group; transmitter 1002 may also include a display device such as a display screen.

[0197] In this embodiment, processor 1003 is used to execute the method performed by the first device in the embodiments corresponding to Figures 1 to 8. It should be noted that the specific manner in which the application processor 10031 in processor 1003 executes the aforementioned steps is based on the same concept as the method embodiments corresponding to Figures 1 to 8 in this application, and the resulting technical effects are the same as those in the method embodiments corresponding to Figures 1 to 8 in this application. For details, please refer to the descriptions in the method embodiments shown above in this application; further details will not be repeated here.

[0198] This application also provides a vehicle, as shown in Figure 11. Figure 11 is a structural schematic diagram of a vehicle provided in this application embodiment. The vehicle 100 is configured for fully or partially automated driving mode. For example, the vehicle 100 can control itself while in automated driving mode, and can determine the current state of the vehicle and its surrounding environment through human operation, determine the possible behavior of at least one other vehicle in the surrounding environment, and determine the confidence level corresponding to the probability of other vehicles performing possible behaviors, and control the vehicle 100 based on the determined information. When the vehicle 100 is in automated driving mode, the vehicle 100 can also be set to operate without human interaction.

[0199] Vehicle 100 may include various subsystems, such as a mobility system 102, a sensor system 104, a control system 106, one or more peripheral devices 108, a power supply 110, a computer system 112, and a user interface 116. Optionally, vehicle 100 may include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and component of vehicle 100 may be interconnected via wired or wireless means.

[0200] The mobility system 102 may include components that provide powered motion to the vehicle 100. In one embodiment, the mobility system 102 may include an engine 118, an energy source 119, a transmission 120, and wheels / tires 121.

[0201] Engine 118 can be an internal combustion engine, an electric motor, an air-compressed engine, or other combinations of engines, such as a hybrid engine consisting of a gasoline engine and an electric motor, or a hybrid engine consisting of an internal combustion engine and an air-compressed engine. Engine 118 converts energy source 119 into mechanical energy. Examples of energy source 119 include gasoline, diesel, other petroleum-based fuels, propane, other compressed gas-based fuels, ethanol, solar panels, batteries, and other sources of electricity. Energy source 119 can also provide energy to other systems of vehicle 100. Transmission 120 transmits mechanical power from engine 118 to wheels 121. Transmission 120 may include a gearbox, a differential, and a drive shaft. In one embodiment, transmission 120 may also include other components, such as a clutch. The drive shaft may include one or more axles that can be coupled to one or more wheels 121.

[0202] Sensor system 104 may include several sensors for sensing information about the environment surrounding vehicle 100. For example, sensor system 104 may include a positioning system 122 (which may be a GPS system, a BeiDou system, or another positioning system), an inertial measurement unit (IMU) 124, a radar 126, a laser rangefinder 128, and a camera 130. Sensor system 104 may also include sensors for the internal systems of the monitored vehicle 100 (e.g., an in-vehicle air quality monitor, fuel gauge, oil temperature gauge, etc.). Sensing data from one or more of these sensors can be used to detect objects and their corresponding characteristics (position, shape, orientation, speed, etc.). This detection and identification is a key function for the safe operation of the autonomous vehicle 100.

[0203] The positioning system 122 can be used to estimate the geographical location of the vehicle 100. An IMU 124 is used to sense changes in the position and orientation of the vehicle 100 based on inertial acceleration. In one embodiment, the IMU 124 can be a combination of an accelerometer and a gyroscope. A radar 126 can use radio signals to sense objects in the surrounding environment of the vehicle 100, specifically millimeter-wave radar or lidar. In some embodiments, in addition to sensing objects, the radar 126 can also be used to sense the speed and / or direction of travel of objects. A laser rangefinder 128 can use lasers to sense objects in the environment in which the vehicle 100 is located. In some embodiments, the laser rangefinder 128 may include one or more laser sources, a laser scanner, and one or more detectors, as well as other system components. A camera 130 can be used to capture multiple images of the surrounding environment of the vehicle 100. The camera 130 can be a still camera or a video camera.

[0204] The control system 106 controls the operation of the vehicle 100 and its components. The control system 106 may include various components, including a steering system 132, a throttle 134, a braking unit 136, a computer vision system 140, a trajectory control system 142, and an obstacle avoidance system 144.

[0205] The steering system 132 is operable to adjust the forward direction of the vehicle 100. For example, in one embodiment, it may be a steering wheel system. The throttle 134 controls the operating speed of the engine 118 and thus the speed of the vehicle 100. The braking unit 136 controls the deceleration of the vehicle 100. The braking unit 136 may use friction to slow down the wheels 121. In other embodiments, the braking unit 136 may convert the kinetic energy of the wheels 121 into electrical current. The braking unit 136 may also take other forms to slow down the rotational speed of the wheels 121 to control the speed of the vehicle 100. The computer vision system 140 is operable to process and analyze images captured by the camera 130 to identify objects and / or features in the environment surrounding the vehicle 100. The objects and / or features may include traffic signals, road boundaries, and obstacles. The computer vision system 140 may use object recognition algorithms, Structure from Motion (SFM) algorithms, video tracking, and other computer vision techniques. In some embodiments, the computer vision system 140 may be used to map the environment, track objects, estimate the speed of objects, etc. The route control system 142 is used to determine the driving route and speed of the vehicle 100. In some embodiments, the route control system 142 may include a lateral planning module 1421 and a longitudinal planning module 1422, which are respectively used to combine data from the obstacle avoidance system 144, GPS 122, and one or more predetermined maps to determine the driving route and speed for the vehicle 100. The obstacle avoidance system 144 is used to identify, evaluate, and avoid or otherwise traverse obstacles in the environment of the vehicle 100, which may specifically be physical obstacles and virtual moving bodies that may collide with the vehicle 100. In one example, the control system 106 may add or alternatively include components other than those shown and described. Alternatively, some of the components shown above may be reduced.

[0206] Vehicle 100 interacts with external sensors, other vehicles, other computer systems, or users via peripheral device 108. Peripheral device 108 may include wireless communication system 146, on-board computer 148, microphone 150, and / or speaker 152. In some embodiments, peripheral device 108 provides a means for a user of vehicle 100 to interact with user interface 116. For example, on-board computer 148 may provide information to a user of vehicle 100. User interface 116 may also operate on-board computer 148 to receive user input. On-board computer 148 may be operated via a touchscreen. In other cases, peripheral device 108 may provide a means for vehicle 100 to communicate with other devices located within the vehicle. For example, microphone 150 may receive audio (e.g., voice commands or other audio input) from a user of vehicle 100. Similarly, speaker 152 may output audio to a user of vehicle 100. Wireless communication system 146 may communicate wirelessly with one or more devices, either directly or via a communication network. For example, the wireless communication system 146 may use 3G cellular communication, such as CDMA, EVDO, GSM / GPRS, or 4G cellular communication, such as LTE, or 5G cellular communication. The wireless communication system 146 may utilize a wireless local area network (WLAN) for communication. In some embodiments, the wireless communication system 146 may utilize an infrared link, Bluetooth, or ZigBee to communicate directly with the device. Other wireless protocols, such as various vehicle communication systems, may also be used. For example, the wireless communication system 146 may include one or more dedicated short-range communications (DSRC) devices that can enable public and / or private data communication between the vehicle and / or a roadside station.

[0207] Power source 110 can provide power to various components of vehicle 100. In one embodiment, power source 110 can be a rechargeable lithium-ion or lead-acid battery. One or more such battery packs can be configured to provide power to various components of vehicle 100. In some embodiments, power source 110 and energy source 119 can be implemented together, as is the case in some fully electric vehicles.

[0208] Some or all of the functions of vehicle 100 are controlled by computer system 112. Computer system 112 may include at least one processor 113, which executes instructions 115 stored in a non-transitory computer-readable medium such as memory 114. Computer system 112 may also be multiple computing devices that control individual components or subsystems of vehicle 100 in a distributed manner. Processor 113 may be any conventional processor, such as a commercially available central processing unit (CPU). Alternatively, processor 113 may be a dedicated device such as an application-specific integrated circuit (ASIC) or other hardware-based processor. Although FIG11 functionally illustrates the processor, memory, and other components of computer system 112 in the same block, those skilled in the art will understand that the processor or memory may actually include multiple processors or memories not stored in the same physical housing. For example, memory 114 may be a hard disk drive or other storage medium located in a housing different from that of computer system 112. Therefore, references to processor 113 or memory 114 will be understood to include references to a collection of processors or memories that may or may not operate in parallel. Unlike using a single processor to perform the steps described herein, some components, such as steering and deceleration components, can each have their own processor that performs only calculations related to the component's specific function.

[0209] In all the aspects described herein, processor 113 may be located remotely from vehicle 100 and may communicate wirelessly with vehicle 100. In other aspects, some of the processes described herein are executed on processor 113 located within vehicle 100, while others are executed by remote processor 113, including taking the necessary steps to perform a single operation.

[0210] In some embodiments, memory 114 may contain instructions 115 (e.g., program logic) that can be executed by processor 113 to perform various functions of vehicle 100, including those described above. Memory 114 may also contain additional instructions, including instructions to send data to, receive data from, interact with, and / or control one or more of the mobility system 102, sensor system 104, control system 106, and peripheral devices 108. In addition to instructions 115, memory 114 may also store data such as road maps, route information, vehicle position, direction, speed, and other such vehicle data, as well as other information. This information may be used by vehicle 100 and computer system 112 during operation of vehicle 100 in autonomous, semi-autonomous, and / or manual modes. A user interface 116 is provided to or receives information from a user of vehicle 100. Optionally, user interface 116 may include one or more input / output devices within the set of peripheral devices 108, such as wireless communication system 146, on-board computer 148, microphone 150, and speaker 152.

[0211] Computer system 112 can control the functions of vehicle 100 based on input received from various subsystems (e.g., driving system 102, sensor system 104, and control system 106) and from user interface 116. For example, computer system 112 can utilize input from control system 106 to control steering system 132 to avoid obstacles detected by sensor system 104 and obstacle avoidance system 144. In some embodiments, computer system 112 is operable to provide control over many aspects of vehicle 100 and its subsystems.

[0212] Alternatively, one or more of these components may be installed separately from or associated with vehicle 100. For example, memory 114 may exist partially or completely separately from vehicle 100. The components may be communicatively coupled together in a wired and / or wireless manner.

[0213] Optionally, the above components are merely examples. In practical applications, components in each of the above modules may be added or removed as needed. Figure 11 should not be construed as a limitation on the embodiments of this application. A vehicle traveling on a road, such as vehicle 100 above, can identify objects in its surrounding environment to determine adjustments to its current speed. These objects can be other vehicles, traffic control equipment, or other types of objects. In some examples, each identified object can be considered independently, and based on the object's individual characteristics, such as its current speed, acceleration, and distance from the vehicle, the speed adjustment to be made by the vehicle can be determined.

[0214] Optionally, vehicle 100 or computing devices associated with vehicle 100, such as computer system 112, computer vision system 140, and memory 114 as shown in Figure 11, can predict the behavior of the identified objects based on the characteristics of the identified objects and the state of the surrounding environment (e.g., traffic, rain, ice on the road, etc.). Optionally, each identified object depends on the behavior of each other, so all identified objects can also be considered together to predict the behavior of a single identified object. Vehicle 100 can adjust its speed based on the predicted behavior of the identified objects. In other words, vehicle 100 can determine what steady state the vehicle will need to adjust to (e.g., accelerate, decelerate, or stop) based on the predicted behavior of the objects. In this process, other factors can also be considered in determining the speed of vehicle 100, such as the lateral position of vehicle 100 in the road, the curvature of the road, the proximity of static and dynamic objects, etc. In addition to providing instructions to adjust the speed of the vehicle, the computing device can also provide instructions to modify the steering angle of the vehicle 100 so that the vehicle 100 follows a given trajectory and / or maintains a safe lateral and longitudinal distance from objects near the vehicle 100 (e.g., cars in adjacent lanes on the road).

[0215] In this embodiment, the processor 113 in the vehicle 100 is used to execute the method executed by the first device in the embodiments corresponding to Figures 1 to 8. It should be noted that the specific manner in which the processor 113 executes the aforementioned steps is based on the same concept as the method embodiments corresponding to Figures 1 to 8 in this application, and the resulting technical effects are the same as those in the method embodiments corresponding to Figures 1 to 8 in this application. For details, please refer to the description in the method embodiments shown above in this application, which will not be repeated here.

[0216] This application also provides a computer-readable storage medium storing a program that, when run on a computer, causes the computer to perform the steps executed by the first device in the methods described in the embodiments shown in Figures 1 to 8.

[0217] This application also provides a computer program product, which includes a program that, when run on a computer, causes the computer to perform the steps performed by the first device in the methods described in the embodiments shown in Figures 1 to 8.

[0218] This application also provides a circuit system including a processing circuit configured to perform the steps executed by the first device in the method described in the embodiments shown in Figures 1 to 8 above.

[0219] The first device or intelligent driving device provided in this application embodiment can specifically be a chip, which includes a processing unit and a communication unit. The processing unit can be, for example, a processor, and the communication unit can be, for example, an input / output interface, pins, or circuits. The processing unit can execute computer execution instructions stored in the storage unit to cause the chip to execute the methods described in the embodiments shown in Figures 1 to 8. Optionally, the storage unit is a storage unit within the chip, such as a register or cache. The storage unit can also be a storage unit located outside the chip in the wireless access device, such as a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, such as random access memory (RAM).

[0220] The processor mentioned above can be a general-purpose central processing unit, a microprocessor, an ASIC, or one or more integrated circuits used to control the execution of a program in the first aspect of the method.

[0221] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0222] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CLUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0223] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0224] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

Claims

1. An intelligent driving method, characterized in that, The method includes: Acquire first information corresponding to at least one first obstacle around the vehicle, the first information including information on at least one predicted trajectory of each first obstacle, wherein there is a second obstacle among the at least one first obstacle, the first information including information on at least two predicted trajectories of the second obstacle; Based on the first information, at least two pieces of second information are determined, wherein the second information indicates the position of the predicted trajectory, and different pieces of second information indicate the positions of different predicted trajectories in the predicted trajectories corresponding to the first information; Obtain the strategy set corresponding to each of the second pieces of information, wherein the strategy set corresponding to each of the second pieces of information includes at least one feasible driving strategy for the vehicle in the lateral and / or longitudinal directions; Based on each piece of second information and the set of strategies corresponding to each piece of second information, the planned trajectory of the vehicle and / or the decision information of the vehicle are determined, wherein the decision information includes the driving strategies of the vehicle in the lateral and / or longitudinal directions.

2. The method according to claim 1, characterized in that, The step of determining the planned trajectory of the vehicle and / or the decision information of the vehicle based on each piece of second information and the strategy set corresponding to each piece of second information includes: Based on each piece of second information and the strategy set corresponding to each piece of second information, a first planned trajectory for the vehicle corresponding to each piece of second information is determined; The planned trajectory of the vehicle is determined based on all the first planned trajectories corresponding to the at least two second pieces of information.

3. The method according to claim 2, characterized in that, The step of determining the planned trajectory of the vehicle and / or the decision information of the vehicle based on each piece of second information and the strategy set corresponding to each piece of second information further includes: The decision information corresponding to the planned trajectory of the vehicle is determined.

4. The method according to claim 2 or 3, characterized in that, The step of determining the first planned trajectory of the vehicle corresponding to each piece of second information based on each piece of second information and the strategy set corresponding to each piece of second information includes: Based on each piece of second information and the strategy set corresponding to each piece of second information, a second planned trajectory for the vehicle corresponding to each piece of second information is determined; Based on the second planned trajectory corresponding to the at least two pieces of second information, a fusion operation is performed to obtain a reference trajectory; Based on each of the second pieces of information and the reference trajectory, a first planned trajectory for the vehicle corresponding to each of the second pieces of information is generated. The objective of generating the first planned trajectory includes ensuring that the first planned trajectory fits the reference trajectory.

5. The method according to claim 4, characterized in that, Both the first planned trajectory and the second planned trajectory include the trajectory of the vehicle from the current time to the first time point, and the reference trajectory includes the trajectory of the vehicle from the current time to the second time point, where the second time point is earlier than the first time point.

6. The method according to claim 5, characterized in that, The step of performing a fusion operation to obtain a reference trajectory based on the second planned trajectory corresponding to the at least two pieces of second information includes: All second planning trajectories corresponding to the at least two second pieces of information are fused to obtain the fused trajectory; Based on the at least two second pieces of information and the fused trajectory, determine the collision time point where the fused trajectory collides with the obstacles around the vehicle, wherein the obstacles around the vehicle include the at least one first obstacle, and the second time point is earlier than the collision time point; The sub-planning trajectories in each of the second planning trajectories are merged to obtain the reference trajectory, wherein the sub-planning trajectories include the trajectories in the second planning trajectories from the current time to the second time point.

7. The method according to claim 6, characterized in that, The factors for determining the second time point also include: a time point where the first distance is greater than or equal to the first preset distance, and / or a time point where the second distance is greater than or equal to the second preset distance, wherein the first distance includes the maximum distance between different second planning trajectories among all the second planning trajectories, and the second distance includes the maximum distance between different predicted trajectories among at least two predicted trajectories of the same second obstacle.

8. The method according to any one of claims 1 to 3, characterized in that, The first obstacle is selected from multiple obstacles around the vehicle based on the vehicle's third planned trajectory. The vehicle's planned trajectory is determined every preset time interval. The vehicle's third planned trajectory is obtained based on the previously determined vehicle's planned trajectory. The selected first obstacle satisfies at least one of the following conditions: the predicted trajectory of the first obstacle intersects with the vehicle's third planned trajectory, or the distance between the predicted trajectory of the first obstacle and the vehicle's third planned trajectory is less than or equal to a preset distance.

9. The method according to any one of claims 1 to 3, characterized in that, The second information is a grid map, indicating the grid occupied by the predicted trajectory.

10. An intelligent driving device, characterized in that, The device includes: An acquisition module is configured to acquire first information corresponding to at least one first obstacle around the vehicle, the first information including information on at least one predicted trajectory of each first obstacle, wherein there is a second obstacle among the at least one first obstacle, and the first information includes information on at least two predicted trajectories of the second obstacle; The determining module is configured to determine at least two pieces of second information based on the first information, wherein the second information indicates the position of the predicted trajectory, and different pieces of second information indicate the positions of different predicted trajectories in the predicted trajectories corresponding to the first information; The acquisition module is further configured to acquire a strategy set corresponding to each piece of the second information, wherein the strategy set corresponding to each piece of the second information includes at least one feasible driving strategy for the vehicle in the lateral and / or longitudinal directions. The determining module is further configured to determine the planned trajectory of the vehicle and / or the decision information of the vehicle based on each second piece of information and the strategy set corresponding to each second piece of information, wherein the decision information includes the driving strategy of the vehicle in the lateral and / or longitudinal directions.

11. The apparatus according to claim 10, characterized in that, The determining module is specifically used for: Based on each piece of second information and the strategy set corresponding to each piece of second information, a first planned trajectory for the vehicle corresponding to each piece of second information is determined; The planned trajectory of the vehicle is determined based on all the first planned trajectories corresponding to the at least two second pieces of information.

12. The apparatus according to claim 11, characterized in that, The determining module is further configured to determine the decision information corresponding to the planned trajectory of the vehicle.

13. The apparatus according to claim 11 or 12, characterized in that, The determining module is specifically used for: Based on each piece of second information and the strategy set corresponding to each piece of second information, a second planned trajectory for the vehicle corresponding to each piece of second information is determined; Based on the second planned trajectory corresponding to the at least two pieces of second information, a fusion operation is performed to obtain a reference trajectory; Based on each of the second pieces of information and the reference trajectory, a first planned trajectory for the vehicle corresponding to each of the second pieces of information is generated. The objective of generating the first planned trajectory includes ensuring that the first planned trajectory fits the reference trajectory.

14. The apparatus according to claim 13, characterized in that, Both the first planned trajectory and the second planned trajectory include the trajectory of the vehicle from the current time to the first time point, and the reference trajectory includes the trajectory of the vehicle from the current time to the second time point, where the second time point is earlier than the first time point.

15. The apparatus according to claim 14, characterized in that, The determining module is specifically used for: All second planning trajectories corresponding to the at least two second pieces of information are fused to obtain the fused trajectory; Based on the at least two second pieces of information and the fused trajectory, determine the collision time point where the fused trajectory collides with the obstacles around the vehicle, wherein the obstacles around the vehicle include the at least one first obstacle, and the second time point is earlier than the collision time point; The sub-planning trajectories in each of the second planning trajectories are merged to obtain the reference trajectory, wherein the sub-planning trajectories include the trajectories in the second planning trajectories from the current time to the second time point.

16. The apparatus according to claim 15, characterized in that, The factors for determining the second time point also include: a time point where the first distance is greater than or equal to the first preset distance, and / or a time point where the second distance is greater than or equal to the second preset distance, wherein the first distance includes the maximum distance between different second planning trajectories among all the second planning trajectories, and the second distance includes the maximum distance between different predicted trajectories among at least two predicted trajectories of the same second obstacle.

17. The apparatus according to any one of claims 10 to 12, characterized in that, The first obstacle is selected from multiple obstacles around the vehicle based on the vehicle's third planned trajectory. The vehicle's planned trajectory is determined every preset time interval. The vehicle's third planned trajectory is obtained based on the previously determined vehicle's planned trajectory. The selected first obstacle satisfies at least one of the following conditions: the predicted trajectory of the first obstacle intersects with the vehicle's third planned trajectory, or the distance between the predicted trajectory of the first obstacle and the vehicle's third planned trajectory is less than or equal to a preset distance.

18. The apparatus according to any one of claims 10 to 12, characterized in that, The second information is a grid map, indicating the grid occupied by the predicted trajectory.

19. A device, characterized in that, The method includes a processor coupled to a memory storing program instructions, which, when executed by the processor, implement the method of any one of claims 1 to 9.

20. A vehicle, characterized in that, The method includes a processor coupled to a memory storing program instructions, which, when executed by the processor, implement the method of any one of claims 1 to 9.

21. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 9.

22. A computer program product, characterized in that, The computer program product includes a program that, when run on a computer, causes the computer to perform the method as described in any one of claims 1 to 9.

Images