A vehicle driving planning method and device, electronic equipment and storage medium

Abstract

The present disclosure relates to a vehicle driving planning method, comprising: acquiring road traffic information, path information and position information of a target vehicle; performing driving trajectory planning processing on the target vehicle according to the road traffic information, the path information and the position information, to obtain a first driving trajectory, the first driving trajectory comprising at least two trajectory points and corresponding driving time, driving distance, target speed and target acceleration information of each trajectory point; performing driving safety detection on the first driving trajectory to obtain a detection result; in the case that the detection result is a passing detection result, performing smoothing processing on the first driving trajectory to obtain a second driving trajectory; and controlling the target vehicle to perform driving operation according to the second driving trajectory. The present disclosure proposes a driving planning method combining trajectory planning, safety monitoring and smoothing processing, ensures safe driving through trajectory driving safety detection, and ensures the continuity and smoothness of the generated trajectory through a smoothing algorithm, thereby ensuring the stability and comfort of driving.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving planning algorithm technology, and in particular to a vehicle driving planning method, device, electronic device and storage medium. Background Technology

[0002] With the continuous development of modern transportation, urban road networks are becoming increasingly dense. Cars, as a means of transportation, facilitate daily travel and save a significant amount of travel time. When unfamiliar with road conditions, vehicles often need to plan their routes. In autonomous or assisted driving systems, the following scenario planning module can typically plan a driving route for the vehicle. However, because the planned routes often fail to adequately balance driving safety and passenger comfort, they can easily pose potential hazards to road safety and personal safety. Therefore, how to balance driving safety and passenger comfort when planning vehicle routes has become an urgent technical problem to be solved. Summary of the Invention

[0003] To address at least one of the aforementioned technical problems, this disclosure provides a vehicle driving planning method, apparatus, electronic device, and storage medium.

[0004] On the one hand, this disclosure provides a vehicle driving planning method, including:

[0005] Obtain road traffic information, route information, and the location information of the target vehicle;

[0006] Based on road traffic information, route information and location information, the target vehicle is processed to plan its driving trajectory to obtain a first driving trajectory. The first driving trajectory includes at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point. The driving time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point.

[0007] Based on road traffic information, driving safety is detected on the first driving trajectory, and the detection results are obtained.

[0008] If the detection result is "passed", the first driving trajectory is smoothed to obtain the second driving trajectory;

[0009] Based on the second driving trajectory, control the target vehicle to perform driving operations.

[0010] In an optional embodiment, based on road traffic information, route information, and location information, a driving trajectory planning process is performed on the target vehicle to obtain a first driving trajectory, including:

[0011] Input road traffic information, route information and location information into the trained trajectory point acquisition model to obtain at least two target trajectory points and the driving distance, speed and acceleration information corresponding to each target trajectory point;

[0012] Perform time-series analysis on at least two target trajectory points to obtain the travel time corresponding to each target trajectory point;

[0013] The velocity and acceleration information are smoothed to obtain the target velocity and target acceleration information;

[0014] The first driving trajectory is determined based on at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point.

[0015] In an optional embodiment, the above method further includes:

[0016] Construct a trajectory point acquisition model;

[0017] Acquire vehicle driving data, which includes information on driving distance, speed, and acceleration.

[0018] Based on vehicle driving data, the trajectory point acquisition model is trained to obtain a trained trajectory point acquisition model.

[0019] In an optional embodiment, the velocity and acceleration information are smoothed to obtain target velocity and target acceleration information, including:

[0020] Determine the extreme values ​​of acceleration and jerk of the target vehicle, as well as the difference in travel time between two adjacent target trajectory points;

[0021] Based on the acceleration information, acceleration extreme value and difference of the first target trajectory point among two adjacent target trajectory points, the acceleration information of the second target trajectory point among two adjacent target trajectory points is constrained to obtain the target acceleration information corresponding to the second target trajectory point. The first target trajectory point is earlier than the second target trajectory point.

[0022] Based on the velocity information, acceleration extreme value, and difference of the first target trajectory point among two adjacent target trajectory points, the velocity information of the second target trajectory point among two adjacent target trajectory points is constrained to obtain the target velocity information corresponding to the second target trajectory point.

[0023] In an optional embodiment, based on road traffic information, a driving safety detection is performed on the first driving trajectory to obtain the detection result, including:

[0024] Based on road traffic information, at least one target traffic object and its corresponding predicted trajectory are determined.

[0025] Detect the target traffic object or predicted trajectory, and determine whether it overlaps with the first driving trajectory to obtain the first detection information;

[0026] The system checks whether the first driving trajectory complies with traffic rules to obtain the second detection information.

[0027] The first and second detection information are determined as the detection results.

[0028] In an optional embodiment, the first driving trajectory is smoothed to obtain a second driving trajectory, including:

[0029] Based on at least two target trajectory points, the first driving trajectory is segmented to obtain at least two segmented driving trajectories.

[0030] Each segment of the driving trajectory is smoothed to obtain at least two smooth driving trajectories.

[0031] At least two smooth driving trajectories are merged to obtain a second driving trajectory.

[0032] In an optional embodiment, the above method further includes:

[0033] If the test result is a failure, activate the vehicle emergency strategy.

[0034] Secondly, the present invention also provides a vehicle driving planning device, comprising:

[0035] The acquisition module is used to acquire road traffic information, route information, and the location information of the target vehicle;

[0036] The first driving trajectory module is used to perform driving trajectory planning processing on the target vehicle based on road traffic information, route information and location information to obtain the first driving trajectory. The first driving trajectory includes at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point. The driving time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point.

[0037] The safety monitoring module is used to perform driving safety detection on the first driving trajectory based on road traffic information and obtain the detection results;

[0038] The second driving trajectory module is used to smooth the first driving trajectory to obtain the second driving trajectory when the detection result is "pass".

[0039] The control module is used to control the target vehicle to perform driving operations based on the second driving trajectory.

[0040] Thirdly, the present invention also provides an electronic device, comprising:

[0041] processor;

[0042] Memory used to store processor-executable instructions;

[0043] The processor is used to execute instructions to implement the vehicle driving planning method described above.

[0044] Fourthly, the present invention also provides a storage medium that, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to execute the above-described vehicle driving planning method.

[0045] Fifthly, the present invention also provides a computer program product comprising a computer program stored in a readable storage medium, wherein at least one processor of a computer device reads from the readable storage medium and executes the computer program, causing the device to perform the above-described vehicle driving planning method.

[0046] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0047] Implementing this disclosure will have the following beneficial effects:

[0048] The system acquires road traffic information, route information, and the location information of the target vehicle. Based on these information, it plans the target vehicle's trajectory to obtain a first trajectory. This first trajectory includes at least two target trajectory points and the corresponding travel time, distance, speed, and acceleration information for each point. The travel time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point. Based on the road traffic information, the system performs a driving safety test on the first trajectory to obtain the test result. If the test result is satisfactory, the system smooths the first trajectory to obtain a second trajectory. Based on the second trajectory, the system controls the target vehicle to perform driving operations.

[0049] This application, by including at least two target trajectory points in the first driving trajectory and the corresponding driving time, driving distance, target speed, and target acceleration information for each target trajectory point, can indicate the time, distance, speed, and acceleration information required for vehicle driving, so as to more accurately control the vehicle to drive along the trajectory; by proposing a driving planning method that combines trajectory planning, safety monitoring, and smoothing processing, it can plan the trajectory while ensuring safe driving through trajectory driving safety detection; and through a smoothing algorithm, it ensures that the generated trajectory is continuous and smooth, ensuring the stability and comfort of driving.

[0050] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0051] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. The accompanying drawings are incorporated in and constitute a part of this specification, illustrating embodiments consistent with this disclosure, and are used together with the description to explain the principles of this disclosure, and do not constitute an improper limitation of this disclosure.

[0052] Figure 1 This is a schematic diagram of an implementation environment according to an exemplary embodiment;

[0053] Figure 2 This is a flowchart illustrating a vehicle driving planning method according to an exemplary embodiment;

[0054] Figure 3 This is a flowchart illustrating an embodiment of obtaining a first driving trajectory;

[0055] Figure 4 This is a flowchart illustrating a model training method according to an exemplary embodiment;

[0056] Figure 5 This is a flowchart illustrating a method for obtaining detection results according to an exemplary embodiment;

[0057] Figure 6 This is a flowchart illustrating an embodiment of obtaining a second driving trajectory;

[0058] Figure 7 This is a schematic diagram illustrating a driving planning framework according to an exemplary embodiment;

[0059] Figure 8 This is a schematic diagram illustrating a front-end planning framework according to an exemplary embodiment;

[0060] Figure 9 This is a block diagram illustrating a vehicle driving planning device according to an exemplary embodiment;

[0061] Figure 10 This is a block diagram illustrating an electronic device for vehicle driving planning according to an exemplary embodiment. Detailed Implementation

[0062] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0063] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0064] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. Like reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise. The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0065] In this document, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0066] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0067] In related technologies, vehicle driving planning is usually implemented using rule-based algorithms, such as curve fitting and optimization algorithms. These methods can generate planning information under some set rules. However, rule-based algorithms have complex logic, many manually set rules that are difficult to match effectively, and they are not flexible enough to ensure driving safety and meet comfort requirements.

[0068] To simplify the driving planning process and ensure that the planning results meet comfort requirements while guaranteeing driving safety, this disclosure provides a vehicle driving planning method.

[0069] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an application environment according to an exemplary embodiment, such as... Figure 1 As shown, the application environment may include server 01 and terminal 02.

[0070] In an optional embodiment, server 01 can be used for computational processing in a vehicle driving planning method. Specifically, server 01 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0071] In an optional embodiment, terminal 02 can perform calculations in conjunction with the vehicle driving planning method of server 01. Specifically, terminal 02 can be, but is not limited to, electronic devices such as smartphones, desktop computers, tablets, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, and smart wearable devices. Optionally, the operating system running on the electronic device can be, but is not limited to, Android, iOS, Linux, Windows, and Unix systems.

[0072] For example, road traffic information, route information, and the location information of the target vehicle are input on terminal 02. Server 01 obtains the road traffic information, route information, and target vehicle location information from terminal 02. Based on the road traffic information, route information, and location information, the server performs trajectory planning processing on the target vehicle to obtain a first trajectory. The first trajectory includes at least two target trajectory points and the corresponding travel time, travel distance, target speed, and target acceleration information for each target trajectory point. The travel time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point. Based on the road traffic information, the server performs a driving safety test on the first trajectory to obtain a test result. If the test result is satisfactory, the server smooths the first trajectory to obtain a second trajectory. Based on the second trajectory, the server controls the target vehicle to perform driving operations. Finally, the driving result is transmitted back to terminal 02.

[0073] In addition, it should be noted that, Figure 1 The example shown is merely one application environment provided by this disclosure; in practical applications, other application environments may also be included.

[0074] In the embodiments described in this specification, the server 01 and the terminal 02 can be directly or indirectly connected through wired or wireless communication, and this disclosure does not impose any restrictions.

[0075] Figure 2 This is a flowchart illustrating a vehicle driving planning method according to an exemplary embodiment, such as... Figure 2 As shown, the vehicle driving planning method includes the following:

[0076] Step S201: Obtain road traffic information, route information, and the location information of the target vehicle.

[0077] In this embodiment of the disclosure, the road traffic information and route information can be obtained by collecting road traffic information through sensors mounted on the vehicle, or by reading data from the GPS or Beidou navigation system, or by obtaining route information given by the route planning module; the location information of the target vehicle can be obtained by obtaining map positioning information in real time.

[0078] Step S202: Based on road traffic information, route information and location information, perform driving trajectory planning processing on the target vehicle to obtain a first driving trajectory. The first driving trajectory includes at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point. The driving time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point.

[0079] In this embodiment of the disclosure, such as Figure 3As shown, based on road traffic information, route information, and location information, the target vehicle's driving trajectory is planned to obtain the first driving trajectory, which includes the following:

[0080] Step S301: Input the road traffic information, route information and location information into the trained trajectory point acquisition model to obtain at least two target trajectory points and the driving distance, speed and acceleration information corresponding to each target trajectory point.

[0081] In this embodiment, obstacle detection information and its predicted trajectory information, road environment information, and target vehicle state information are first extracted from road traffic information, path information, and location information. The obstacle predicted trajectory information and target vehicle state information are updated over time. If the cycle of driving planning calculation is T, and the time step of the trajectory is Δt, then in each cycle, the trained trajectory point acquisition model is called to calculate T / Δt-1 times, resulting in T / Δt-1 target trajectory points. When calculating the first target trajectory point, the current state of the target vehicle is used as the state information input. Based on the obstacle detection information and its predicted trajectory information, road environment information, and target vehicle state information, the first target trajectory point and the corresponding driving distance, speed, and acceleration information are calculated. The calculation of subsequent target trajectory points uses the previously calculated target trajectory point as the state information input, and combines the obstacle detection information and its predicted trajectory information, and road environment information to calculate subsequent target trajectory points and the corresponding driving distance, speed, and acceleration information for each subsequent target trajectory point. The calculation result is as follows: target trajectory point P(s,v,a), where s is the driving distance, v is the speed, and a is the acceleration.

[0082] Step S302: Perform time series analysis on at least two target trajectory points to obtain the travel time corresponding to each target trajectory point.

[0083] In this embodiment of the disclosure, since the output of at least two target trajectory points does not include a time sequence, a time sequence needs to be added to the at least two target trajectory points to indicate the travel time of each target trajectory point. The travel time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point. Optionally, the time interval is Δt, the time step of the trajectory. The travel time of the first target trajectory point is based on the initial time of the trajectory planning. The travel time of the next target trajectory point is increased by one time step Δt, and so on, until the travel time of the last target trajectory point is determined.

[0084] Step S303: Smooth the velocity and acceleration information to obtain the target velocity and target acceleration information.

[0085] In this embodiment of the disclosure, the velocity and acceleration information are smoothed to obtain target velocity and target acceleration information, including the following:

[0086] Step S3031: Determine the extreme values ​​of acceleration and jerk of the target vehicle, as well as the difference in travel time between two adjacent target trajectory points.

[0087] In this embodiment of the disclosure, the maximum and minimum acceleration of the target vehicle for safe driving, as well as the maximum and minimum jerk, are used to determine the extreme values ​​of acceleration and jerk of the target vehicle; the difference is determined by subtracting the travel time corresponding to two adjacent target trajectory points.

[0088] Step S3032: Based on the acceleration information, jerk extreme value and difference of the first target trajectory point among two adjacent target trajectory points, constrain the acceleration information of the second target trajectory point among two adjacent target trajectory points to obtain the target acceleration information corresponding to the second target trajectory point. The first target trajectory point is earlier than the second target trajectory point.

[0089] In this embodiment of the present disclosure, the first target trajectory point can be represented as P. i-1 =(t i-1 s i-1 , vi-1 a i-1 The second target trajectory point can be represented as P. i =(t i s i , vi a i Constrain the acceleration information of the second target trajectory point among two adjacent target trajectory points to limit the acceleration of the second target trajectory point. The constraint method can be: a) i-1 +j min *(t i -t i-1 )≤a i ≤a i-1 +j max *(t i -t i-1 ), where j min For minimum jerk, j max For the maximum jerk, (t) i -t i-1 ) represents the difference in travel time between two adjacent target trajectory points.

[0090] Step S3033: Based on the velocity information, acceleration extreme value and difference of the first target trajectory point among two adjacent target trajectory points, constrain the velocity information of the second target trajectory point among two adjacent target trajectory points to obtain the target velocity information corresponding to the second target trajectory point.

[0091] In this embodiment of the disclosure, the velocity information of the second target trajectory point among two adjacent target trajectory points is constrained to limit the velocity of the second target trajectory point. The constraint method can be: v i-1 +a min* (t i -t i-1 )≤v i ≤v i-1 +a max *(t i -r i-1 ), where a min For minimum acceleration, a max For the maximum acceleration, (t) i -t i-1 ) represents the difference in travel time between two adjacent target trajectory points.

[0092] Based on the above, the embodiments of this disclosure constrain the acceleration and speed information of the second target trajectory point according to the acceleration information, acceleration extreme value, and difference in travel time between the first target trajectory point and the first target trajectory point, and the speed information, acceleration extreme value, and difference in travel time between the two adjacent target trajectory points. This can prevent the potential safety hazards caused by sudden acceleration or speed changes during vehicle operation. Compared with the method of directly constraining the acceleration or speed information of the second target trajectory point through acceleration extreme value or speed extreme value, the embodiments of this disclosure can avoid excessive acceleration or speed difference between adjacent target trajectory points, thereby increasing the comfort of vehicle operation.

[0093] Step S304: Determine the first driving trajectory based on at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point.

[0094] In this embodiment of the disclosure, connecting at least two target trajectory points forms the first driving trajectory, which includes the driving time, driving distance, target speed, and target acceleration information corresponding to each target trajectory point.

[0095] Based on the above, the embodiments of this disclosure, by inputting the collected information into the trained trajectory point acquisition model, can more easily obtain the driving distance, speed, and acceleration information corresponding to each target trajectory point without setting complex rules, thus simplifying the complexity of the algorithm; by adding timing to at least two target trajectory points, the execution order of each target trajectory point can be determined, further enriching the content included in the target trajectory point information and facilitating vehicle execution.

[0096] In an optional embodiment, such as Figure 4 As shown, the above method also includes:

[0097] Step S401: Construct a trajectory point acquisition model.

[0098] In this embodiment, the trajectory point acquisition model is preferably a radial basis function neural network (RBN) model. The RBN model consists of three layers: an input layer, a hidden layer, and an output layer. The input layer has m parameters, primarily including the target vehicle's location information and state information determined from road traffic information, path information, and the target vehicle's location information, as well as the location information of obstacles and their predicted trajectory information. The hidden layer has k neurons, with a Gaussian activation function. The output layer has n output parameters, mainly related to the target trajectory point, including the distance traveled to the target trajectory point, speed, and acceleration information.

[0099] Step S402: Obtain vehicle driving data, which includes driving distance, driving speed, and driving acceleration label information.

[0100] In this embodiment, the vehicle driving data is preferably sampled from experienced drivers in various road scenarios. The data samples are screened, and data that affects safety or user experience, as well as invalid data, are removed. The screened data is labeled, taking into account both safety and user experience. Labeling methods may include adding tags for driving distance, driving speed, and driving acceleration, and assigning weights to different driving distances, speeds, and accelerations. Data with better safety and comfort are assigned higher weights, while data with poorer safety and comfort are assigned lower weights.

[0101] Step S403: Based on vehicle driving data, train the trajectory point acquisition model to obtain a trained trajectory point acquisition model.

[0102] In this embodiment, tagged vehicle driving data is input into a pre-constructed trajectory point acquisition model, and the model is iteratively trained. The purpose of training is to obtain the center, variance, and weights from the hidden layer to the output layer of the neural network algorithm, thereby obtaining the key parameters of the model. The trained model is tested using evaluation data; if it meets the performance indicators, the training task is completed; otherwise, training is repeated.

[0103] As can be seen from the above, the embodiments of this disclosure provide a large amount of reliable training data for model training by collecting driving data from excellent drivers; the trajectory planned by the algorithm based on neural network algorithm is closer to the driving behavior of excellent drivers, making the driving trajectory smooth, comfortable and efficient, and making the driving of autonomous vehicles easier for other drivers to understand.

[0104] Step S203: Based on road traffic information, perform driving safety detection on the first driving trajectory and obtain the detection results.

[0105] In this embodiment of the disclosure, such as Figure 5 As shown, based on road traffic information, a driving safety test is performed on the first driving trajectory, and the test results include the following:

[0106] Step S501: Based on road traffic information, determine at least one target traffic object and its corresponding predicted trajectory.

[0107] In this embodiment of the disclosure, at least one target traffic object is extracted from road traffic information, and the trajectory of the target traffic object is predicted to obtain the predicted trajectory corresponding to the target traffic object. Optionally, the target traffic object may be pedestrians or vehicles around the target vehicle, or pedestrians or vehicles in the direction of travel of the target vehicle.

[0108] Step S502: Detect the target traffic object or predicted trajectory and determine whether it overlaps with the first driving trajectory to obtain the first detection information.

[0109] In this embodiment of the disclosure, detecting a target traffic object or a predicted trajectory may involve detecting at least one frame of an image containing the target traffic object or its predicted trajectory, detecting whether the target traffic object or its predicted trajectory in the at least one frame of the image overlaps with a first driving trajectory, and obtaining first detection information.

[0110] Step S503: Detect whether the first driving trajectory meets the traffic rules and obtain the second detection information.

[0111] In this embodiment of the disclosure, the traffic rules can be driving regulations within traffic rules. When the traffic rule is a driving regulation, the system detects whether the first driving trajectory meets the driving regulation to obtain second detection information.

[0112] Step S504: Determine the first detection information and the second detection information as the detection result.

[0113] In this embodiment of the disclosure, the first detection information and the second detection information are combined into a detection result.

[0114] Based on the above, this embodiment of the present disclosure can verify whether a collision with other vehicles or pedestrians will occur during the trajectory execution process by detecting whether the target traffic object or the predicted trajectory overlaps with the first driving trajectory. This not only ensures the safety of the driver but also helps to improve the overall driving safety of the road. By detecting whether the first driving trajectory meets traffic rules, it can verify whether the relevant traffic rules will be followed during the trajectory execution process, thus avoiding the trajectory planning from violating traffic regulations.

[0115] Step S204: If the detection result is "passed", smooth the first driving trajectory to obtain the second driving trajectory.

[0116] In this embodiment of the disclosure, such as Figure 6 As shown, the first driving trajectory is smoothed to obtain the second driving trajectory, which includes the following:

[0117] Step S601: Based on at least two target trajectory points, the first driving trajectory is segmented to obtain at least two segmented driving trajectories.

[0118] In this embodiment of the disclosure, the first driving trajectory is divided into at least two segmented driving trajectories based on at least two target trajectory points. The endpoint of each segmented driving trajectory is a target trajectory point among the at least two target trajectory points.

[0119] Step S602: Smooth each segment of the driving trajectory to obtain at least two smooth driving trajectories.

[0120] In this embodiment of the disclosure, a smoothing algorithm is used to smooth each segment of the driving trajectory, ensuring that each segment meets driving comfort requirements and reducing sharp turns in each segment, resulting in at least two smooth driving trajectories. These at least two smoothed driving trajectories still contain at least two target trajectory points.

[0121] Step S603: Merge at least two smooth driving trajectories to obtain a second driving trajectory.

[0122] In this embodiment of the disclosure, at least two smooth driving trajectories are merged according to the driving sequence to obtain a second driving trajectory.

[0123] As can be seen from the above, by segmenting the first driving trajectory, the subsequent smoothing process can be more accurate; by smoothing each segment of the driving trajectory, the possible sharp turns and other operations in each segment of the driving trajectory can be reduced, improving the comfort of the driving process and making the trajectory closer to the driving operation of an excellent driver.

[0124] Step S205: Control the target vehicle to perform driving operations according to the second driving trajectory.

[0125] In this embodiment of the disclosure, the target vehicle is controlled to travel along the second driving trajectory based on the trajectory path included in the second driving trajectory, at least two target trajectory points, and the driving time, driving distance, driving speed, and acceleration information corresponding to each target trajectory point.

[0126] In an optional embodiment, the above method further includes:

[0127] If the test result is a failure, activate the vehicle emergency strategy.

[0128] In this embodiment of the disclosure, the detection failure includes: a first detection result indicating that the target traffic object or its predicted trajectory overlaps with the first driving trajectory, and / or a second detection result indicating that the first driving trajectory does not comply with traffic rules. In the event of a detection failure, activating the vehicle emergency strategy includes: activating the vehicle takeover function to remind the driver to take over the vehicle; if the takeover function is not available, then an emergency stop is initiated.

[0129] As can be seen from the above, the embodiments of this disclosure, by activating the vehicle emergency strategy when the detection result is a failure, can promptly remind the driver to control the vehicle when the safety detection fails, thus preventing traffic accidents; and by performing an emergency stop in the absence of a takeover function, it can avoid the potential traffic safety hazards caused by the vehicle continuing to drive, thereby improving the safety of driving plans.

[0130] In the above embodiments, this disclosure, by inputting the collected information into a trained trajectory point acquisition model, can more easily obtain the driving distance, speed, and acceleration information corresponding to each target trajectory point without setting complex rules, thus simplifying the complexity of the algorithm. By constraining the acceleration and speed information of the second target trajectory point, it can prevent the potential safety hazards caused by sudden acceleration or speed changes during vehicle driving. Compared with the method of directly constraining the acceleration or speed information of the second target trajectory point through acceleration extreme values ​​or speed extreme values, it can avoid excessively large acceleration or speed differences between two adjacent target trajectory points, increasing the comfort of vehicle driving. By detecting whether the target traffic object or predicted motion trajectory overlaps with the first driving trajectory, it can verify whether there will be a collision with other vehicles or pedestrians during trajectory execution, which not only ensures its own driving safety but also helps to improve the overall driving safety of the road. By smoothing each segment of the driving trajectory, it can improve the comfort of the driving process and make the trajectory closer to the driving operation of an excellent driver.

[0131] In one specific implementation method, the technical solution of this application is implemented as follows:

[0132] Figure 7 It is a driving planning framework. For example... Figure 7 As shown, the implementation process adopts a driving planning framework combining front-end planning, a safety module, and back-end planning. The front-end module uses a neural network algorithm, which is simple in structure and highly efficient. The safety module performs collision detection based on the trajectory information calculated by the front-end to ensure safe driving. The back-end module uses a smoothing algorithm to ensure that the generated trajectory is continuous and smooth, ensuring driving stability and comfort.

[0133] The preprocessing module acquires the target vehicle's location information (i.e., map positioning information), road traffic information (i.e., perception and prediction information), and path information provided by path planning in real time, integrating these inputs into the data types required by the front-end planning module. After receiving the input from the preprocessing module, the front-end planning module calculates trajectory 1 using a neural network model and sends trajectory 1 to the safety module for collision detection and traffic rule checks. If the checks pass, trajectory 1 is sent to the back-end planning module for smoothing and finally generates trajectory 2, which is then published. If trajectory 1 fails the safety check, feedback is sent to the front-end module for recalculation. If multiple consecutive frames fail to generate a trajectory that meets the safety check, the safety module reports an error status to the system and initiates emergency braking, takeover requests, and other strategies.

[0134] The aforementioned front-end planning module based on a neural network algorithm collects driving data from experienced drivers to train a neural network model, making the output trajectory of the driving plan closer to that of a human driver. This improves driving comfort, smoothness, and traffic efficiency, and the driving behavior is more easily understood by other human drivers. Figure 8 As shown, the front-end planning module includes a scheduling module, a neural network model, an input module, a constraint module, and a post-processing module. The input module extracts three types of data information based on map positioning information, perception prediction information, and path information provided by path planning: target traffic object information and its predicted trajectory information, road environment information, and vehicle input status information. Among these, the predicted trajectory of the target traffic object and the vehicle status information are dynamically updated according to the calculation time sequence.

[0135] The aforementioned neural network model preferably uses a radial basis function (RBF) neural network. Practical applications include, but are not limited to, RBF neural network algorithms. A RBF neural network consists of three layers: an input layer, hidden layers, and an output layer. The input layer has *m* parameters, primarily including vehicle positioning information, state information, obstacle positioning information, and predicted trajectory information. The hidden layer has *k* neurons, with a Gaussian activation function. There are *n* output parameters, mainly information about the planned target trajectory points, including travel distance, speed, and acceleration.

[0136] The purpose of neural network training is to obtain the center of the basis functions, variance (width), and weights from the hidden layers to the output layer of the neural network algorithm. First, sufficient driving data samples from excellent drivers in various road scenarios must be collected. These samples are then filtered, removing data that might affect safety or user experience, as well as invalid data. The processed data is labeled, taking into account both safety and user experience. For ease of training, the data is normalized. The data is then input into the designed algorithm model to begin iterative training, obtaining the model's key parameters. The trained model is then tested using evaluation data. If the performance metrics are met, the training task is complete; otherwise, training is re-established.

[0137] The scheduling module performs time-series scheduling. If the planning module's calculation cycle is T and the trajectory's time step is Δt, then within each planning cycle, the neural network model algorithm is called to calculate T / Δt-1 times. Except for the first trajectory point calculation, which uses the vehicle's current state as input, the calculations for other points use the information from the previous calculated trajectory point as input. The calculation result P(s, v, a) is sent to the post-processing module and assigned a time step ti, resulting in the trajectory point Pi = (t... i s i v i a i ).

[0138] The constraint module will constrain the velocity and acceleration information of Pi's trajectory points. When i > 0, the previous calculated trajectory point is as follows:

[0139] P i-1 =(t i-1 s i-1 v i-1 a i-1 )

[0140] Based on the previously calculated trajectory point and the established constraints, the velocity v and acceleration a in the Pi data are constrained as follows:

[0141]

[0142] After each calculation is completed, the corresponding trajectory points are output to the post-processing module. The post-processing module sends the latest calculated trajectory points to the neural network algorithm for the next round of calculation. After completing T / Δt-1 times, a complete planned trajectory Γ is obtained and the trajectory Γ is sent to the security module.

[0143] The safety module primarily includes collision detection and emergency strategies to enhance the safety of the neural network model's calculated trajectory. Collision detection is performed on trajectory 1 generated by the front-end planning module. If successful, trajectory 1 is sent to the back-end planning module for smoothing. If the collision detection fails and there is a collision risk, the status is fed back to the front-end planning module for replanning. If several planning attempts all indicate a collision risk, the emergency strategy is activated. If a takeover function is available, the driver is prompted to take over; otherwise, an emergency stop is initiated.

[0144] The backend planning module, based on the results of the frontend planning module, uses relevant smoothing algorithms to smooth and optimize the frontend trajectory, ultimately obtaining a trajectory that meets comfort and stability requirements, and then publishes the trajectory to the downstream trajectory tracking and control module.

[0145] Figure 9This is a block diagram illustrating a vehicle driving planning device according to an exemplary embodiment, with reference to... Figure 9 The device includes an acquisition module 901, a first driving trajectory module 902, a safety monitoring module 903, a second driving trajectory module 904, and a control module 905, wherein...

[0146] The acquisition module 901 is used to acquire road traffic information, route information, and the location information of the target vehicle.

[0147] The first driving trajectory module 902 is used to perform driving trajectory planning processing on the target vehicle based on road traffic information, route information and location information to obtain the first driving trajectory. The first driving trajectory includes at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point. The driving time indicates the time required for the target vehicle to travel from the starting point to each target trajectory point.

[0148] The safety monitoring module 903 is used to perform driving safety detection on the first driving trajectory based on road traffic information and obtain the detection results;

[0149] The second driving trajectory module 904 is used to smooth the first driving trajectory to obtain the second driving trajectory when the detection result is a pass.

[0150] The control module 905 is used to control the target vehicle to perform driving operations according to the second driving trajectory.

[0151] In an optional embodiment, the first driving trajectory module 902 includes:

[0152] The input module is used to input road traffic information, path information and location information into the trained trajectory point acquisition model to obtain at least two target trajectory points and the driving distance, speed and acceleration information corresponding to each target trajectory point;

[0153] The time series analysis module is used to perform time series analysis on at least two target trajectory points to obtain the travel time corresponding to each target trajectory point.

[0154] The velocity smoothing module is used to smooth the velocity and acceleration information to obtain the target velocity and target acceleration information;

[0155] The first driving trajectory determination module is used to determine the first driving trajectory based on at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point.

[0156] In an optional embodiment, the above-described apparatus further includes:

[0157] Build modules are used to construct trajectory point acquisition models;

[0158] The driving data module is used to acquire vehicle driving data, which carries labels for driving distance, driving speed, and driving acceleration.

[0159] The training module is used to train the trajectory point acquisition model based on vehicle driving data, so as to obtain a trained trajectory point acquisition model.

[0160] In an optional embodiment, the smoothing processing module described above includes:

[0161] The extreme value acquisition module is used to determine the extreme values ​​of acceleration and jerk of the target vehicle, as well as the difference in travel time between two adjacent target trajectory points;

[0162] The first constraint module is used to constrain the acceleration information of the second target trajectory point between two adjacent target trajectory points based on the acceleration information, acceleration extreme value and difference of the first target trajectory point between two adjacent target trajectory points, so as to obtain the target acceleration information corresponding to the second target trajectory point, wherein the first target trajectory point is earlier than the second target trajectory point;

[0163] The second constraint module is used to constrain the velocity information of the second target trajectory point based on the velocity information, acceleration extreme value and difference of the first target trajectory point among two adjacent target trajectory points, so as to obtain the target velocity information corresponding to the second target trajectory point.

[0164] In an optional embodiment, the security monitoring module 903 includes:

[0165] The target traffic object module is used to determine at least one target traffic object and its corresponding predicted motion trajectory based on road traffic information.

[0166] The first detection module is used to detect target traffic objects or predict their movement trajectories, determine whether they overlap with the first driving trajectory, and obtain first detection information.

[0167] The second detection module is used to detect whether the first driving trajectory meets traffic rules and obtain second detection information.

[0168] The detection result determination module is used to determine the first detection information and the second detection information as the detection result.

[0169] In an optional embodiment, the second driving trajectory module 904 includes:

[0170] The segmentation module is used to segment the first driving trajectory based on at least two target trajectory points to obtain at least two segmented driving trajectories.

[0171] The trajectory smoothing module is used to smooth each segment of the driving trajectory to obtain at least two smooth driving trajectories.

[0172] The merging module is used to merge at least two smooth driving trajectories to obtain a second driving trajectory.

[0173] In an optional embodiment, the above-described apparatus further includes:

[0174] The emergency module is used to activate the vehicle's emergency strategy in the event that the test result is a failure.

[0175] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0176] In an exemplary embodiment, an electronic device is also provided, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is used for the instructions to implement the vehicle driving planning method as described in the embodiments of this disclosure.

[0177] Figure 10 This is a block diagram illustrating an electronic device for vehicle driving planning according to an exemplary embodiment. The electronic device may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown, the electronic device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a vehicle driving planning method. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0178] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the electronic device to which the present disclosure is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0179] In an exemplary embodiment, a storage medium is also provided, which, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to execute the vehicle driving planning method of the present disclosure embodiments.

[0180] In an exemplary embodiment, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform the vehicle driving planning method of the present disclosure embodiments.

[0181] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this disclosure can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0182] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0183] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A vehicle travel planning method characterized by comprising: The method comprises: acquiring road traffic information, path information and position information of a target vehicle; performing driving trajectory planning processing on the target vehicle according to the road traffic information, the path information and the position information, to obtain a first driving trajectory; performing driving safety detection on the first driving trajectory based on the road traffic information, to obtain a detection result; performing smoothing processing on the first driving trajectory when the detection result is a pass detection, to obtain a second driving trajectory; controlling the target vehicle to perform a driving operation according to the second driving trajectory; starting a vehicle emergency strategy when the detection result is a fail detection.

2. The method of claim 1, wherein, The first driving trajectory comprises at least two target trajectory points and target speed and target acceleration information corresponding to each target trajectory point; the driving trajectory planning processing on the target vehicle according to the road traffic information, the path information and the position information to obtain the first driving trajectory comprises: inputting the road traffic information, the path information and the position information into a trained trajectory point acquisition model, to obtain the at least two target trajectory points and the target speed and acceleration information corresponding to each target trajectory point; performing time sequence analysis on the at least two target trajectory points, to obtain the driving time corresponding to each target trajectory point; performing smoothing processing on the speed and acceleration information, to obtain the target speed and target acceleration information; determining the first driving trajectory according to the at least two target trajectory points and the driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point.

3. The method of claim 2, wherein, The method further comprises: constructing a trajectory point acquisition model; acquiring vehicle driving data, which carries driving distance, driving speed and driving acceleration label information; training the trajectory point acquisition model based on the vehicle driving data, to obtain the trained trajectory point acquisition model.

4. The method of claim 2, wherein, The smoothing processing on the speed and acceleration information to obtain the target speed and target acceleration information comprises: determining acceleration extreme values and jerk extreme values of the target vehicle, and a difference value of driving times corresponding to adjacent two target trajectory points; constraining acceleration information of a second target trajectory point in the adjacent two target trajectory points according to acceleration information of a first target trajectory point in the adjacent two target trajectory points, the jerk extreme values and the difference value, to obtain the target acceleration information corresponding to the second target trajectory point, the first target trajectory point being earlier than the second target trajectory point; constraining speed information of a second target trajectory point in the adjacent two target trajectory points according to speed information of a first target trajectory point in the adjacent two target trajectory points, the acceleration extreme values and the difference value, to obtain the target speed information corresponding to the second target trajectory point.

5. The method of claim 1, wherein, The driving safety detection on the first driving trajectory based on the road traffic information to obtain the detection result comprises: determining at least one target traffic object and a predicted motion trajectory corresponding to the target traffic object based on the road traffic information; detecting whether the target traffic object or the predicted motion trajectory overlaps with the first driving trajectory to obtain first detection information; detecting whether the first driving trajectory satisfies a traffic rule to obtain second detection information; determining the first detection information and the second detection information as the detection result.

6. The method of claim 1, wherein, The first driving trajectory includes at least two target trajectory points and driving time, driving distance, target speed and target acceleration information corresponding to each target trajectory point, and the smoothing processing on the first driving trajectory to obtain a second driving trajectory includes: segmenting the first driving trajectory based on the at least two target trajectory points to obtain at least two segmented driving trajectories; smoothing each segmented driving trajectory to obtain at least two smooth driving trajectories; and merging the at least two smooth driving trajectories to obtain the second driving trajectory.

7. A vehicle travel planning device characterized by comprising: The apparatus includes: an acquisition module configured to acquire road traffic information, path information and position information of a target vehicle; a first driving trajectory module configured to perform driving trajectory planning processing on the target vehicle based on the road traffic information, the path information and the position information to obtain a first driving trajectory; a safety monitoring module configured to perform driving safety detection on the first driving trajectory based on the road traffic information to obtain a detection result; a second driving trajectory module configured to perform smoothing processing on the first driving trajectory to obtain a second driving trajectory in a case where the detection result is a pass detection result; a control module configured to control the target vehicle to perform driving operation according to the second driving trajectory; and an emergency module configured to start a vehicle emergency strategy in a case where the detection result is a fail detection result.

8. An electronic device, comprising: includes: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the vehicle driving planning method according to any one of claims 1 to 6.

9. A storage medium, characterized by When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device can perform the vehicle driving planning method according to any one of claims 1 to 6.

Images