Path determination method and device, vehicle-mounted controller and vehicle
By obtaining the vehicle's current environmental information and path planning information, evaluating and filtering paths, and using deep learning models for path scoring, the problem of inaccurate path determination in the existing technology is solved, and the accuracy of path determination is improved.
Patent Information
- Application Number
- CN202411159735.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-11-29
Smart Images

Figure CN119043358A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of automobile technology, and in particular, relates to a path determination method, device, vehicle-mounted controller and vehicle. Background Art
[0002] With the rapid development of automobile technology, the complexity and intelligence level of automobile electronic systems are constantly improving, vehicle functions are becoming increasingly rich, and user experience has also been significantly improved. In practical applications, since road conditions are complex and changeable, and emergencies are prone to occur, in order to further improve the user's riding experience and riding safety, the vehicle's driving path can be planned to control the vehicle to drive along the planned driving path.
[0003] However, the existing path determination methods are not comprehensive enough and are difficult to meet actual needs, which reduces the accuracy of path determination. Summary of the invention
[0004] The embodiments of the present application provide a path determination method, device, vehicle-mounted controller and vehicle to solve the problem that the existing path determination method is not comprehensive enough, difficult to meet actual needs, and reduces the accuracy of path determination.
[0005] In a first aspect, an embodiment of the present application provides a path determination method, including:
[0006] When a path planning instruction is detected, current environment information of a current area where the vehicle is located is obtained; the path planning instruction carries the path planning information;
[0007] Determining a plurality of candidate paths based on the current environment information and the path planning information;
[0008] Obtaining path information of each candidate path, and evaluating each candidate path based on the path information to obtain evaluation results of each candidate path in multiple different dimensions;
[0009] Inputting the current environment information and each path information into a trained path scoring model for processing to obtain a path score corresponding to each of the candidate paths; the path scoring model is trained based on a pre-built deep learning model;
[0010] Based on the evaluation results of each candidate path in multiple different dimensions and the path score of each candidate path, each candidate path is screened to obtain a target driving path.
[0011] In a second aspect, an embodiment of the present application provides a path determination device, including:
[0012] A first acquisition unit is used to acquire current environment information of a current area where the vehicle is located when a path planning instruction is detected; the path planning instruction carries the path planning information;
[0013] A first path determination unit, configured to determine a plurality of candidate paths based on the current environment information and the path planning information;
[0014] An evaluation unit, configured to obtain path information of each candidate path, and evaluate each candidate path based on the path information to obtain evaluation results of each candidate path in multiple different dimensions;
[0015] An input unit, used for inputting the current environment information and each path information into a trained path scoring model for processing, so as to obtain a path score corresponding to each candidate path; the path scoring model is trained based on a pre-built deep learning model;
[0016] The second path determination unit is used to screen each of the candidate paths based on the evaluation results of each of the candidate paths in multiple different dimensions and the path scores of each of the candidate paths to obtain a target driving path.
[0017] In a third aspect, an embodiment of the present application provides a vehicle-mounted controller, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a path determination method as described in any one of the first aspects above is implemented.
[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the path determination method as described in any one of the above-mentioned first aspects is implemented.
[0019] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on a vehicle controller, the vehicle controller can execute the path determination method described in any one of the first aspects above.
[0020] In a sixth aspect, an embodiment of the present application provides a vehicle, comprising an on-board controller, wherein the on-board controller is used to execute any one of the path determination methods described in the first aspect.
[0021] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0022] The embodiment of the present application provides a path determination method, which obtains the current environment information of the current area where the vehicle is located when a path planning instruction is detected; the path planning instruction carries the path planning information; based on the current environment information and the path planning information, multiple alternative paths are determined; the path information of each alternative path is obtained, and each alternative path is evaluated based on each path information, and the evaluation results of each alternative path in multiple different dimensions are obtained; the current environment information and each path information are input into the trained path scoring model for processing to obtain the path score corresponding to each alternative path; based on the evaluation results of each alternative path in multiple different dimensions and the path score of each alternative path, each alternative path is screened and processed to obtain the target driving path. Compared with the prior art, the method can accurately determine multiple alternative paths matching the vehicle by combining the current environment information and the path planning information, and then, on the one hand, the situation of each alternative path in different dimensions can be evaluated by combining each path information, and on the other hand, each alternative path can be further evaluated by combining the current environment information and each path information, so as to achieve accurate evaluation of each alternative path. Finally, combining the evaluation results of each alternative path in multiple dimensions and the scores of each path also enables accurate determination of the target driving path, that is, the driving path that best suits the vehicle can be selected from multiple alternative paths, thereby improving the accuracy of path determination. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0024] Figure 1 is a flowchart of an implementation of a path determination method provided in an embodiment of the present application;
[0025] Figure 2 is a flow chart of an implementation of a path determination method provided by another embodiment of the present application;
[0026] Figure 3 is a flow chart of an implementation of a path determination method provided in yet another embodiment of the present application;
[0027] Figure 4 is a structural diagram of a path determination device provided in an embodiment of the present application;
[0028] Figure 5 It is a structural diagram of a vehicle-mounted controller provided in one embodiment of the present application. DETAILED DESCRIPTION
[0029] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0030] In practical applications, with the rapid development of automobile technology, the complexity and intelligence level of automobile electronic systems continue to improve, vehicle functions are increasingly rich, and user experience has also been significantly improved. In practical applications, since road conditions are complex and changeable, and emergencies are prone to occur, in order to further improve the user's riding experience and riding safety, the vehicle's driving path can be planned to control the vehicle to travel along the planned driving path. For example, taking emergencies such as traffic accidents or natural disasters as an example, how to determine the vehicle's evacuation path to control the vehicle's emergency evacuation along the evacuation path to ensure the safety of the occupants has become an important technical challenge.
[0031] Existing path determination methods usually simply consider whether the road traffic is congested to plan the above path, which is not comprehensive enough and difficult to meet actual needs, thereby reducing the accuracy of path determination.
[0032] Based on this, in all embodiments of the present application, a path determination method is provided to improve the accuracy of path determination. It should be noted that the specific implementation process of the above path determination method can be found in detail. Figure 1 to Figure 3 The corresponding method embodiments are not described in detail here.
[0033] In one embodiment of the present application, in order to achieve rapid response of the vehicle, the system in the vehicle related to the above-mentioned path determination method (such as an automotive electronic system) can adopt a service-oriented architecture (Service-Oriented Architecture, SOA), split the different functional units of the above-mentioned system into independent services, and connect and communicate through standardized interfaces. By achieving loose coupling and combinable extensions between services, the flexibility and scalability of the above-mentioned system can be improved. At the same time, the use of a service-oriented architecture can also achieve rapid integration and collaborative work between the above-mentioned system and other systems (such as traffic management systems and rescue systems, etc.) to achieve information sharing and optimal use of resources, thereby meeting the needs in different planning scenarios (such as emergency evacuation scenarios (traffic accidents, natural disasters, etc. and conventional driving scenarios) while ensuring the safety of passengers.
[0034] In practical applications, service-oriented architecture is a component model that splits different functional units of an application (called services) and connects them through well-defined interfaces and protocols between these services. The interface is defined in a neutral way and should be independent of the hardware platform, operating system, and programming language that implements the service. This allows services built in a variety of systems to interact in a unified and universal way.
[0035] See also Figure 1 , Figure 1 This is a flow chart of a path determination method provided in an embodiment of the present application. In the embodiment of the present application, the execution subject of the path determination method is a vehicle controller. The vehicle controller may be a domain controller.
[0036] It should be noted that in all embodiments of the present application, the system related to the path determination method in the vehicle can adopt a service-oriented architecture.
[0037] like Figure 1 As shown, the path determination method provided in an embodiment of the present application may include S101 to S104, which are described in detail as follows:
[0038] In S101, when a path planning instruction is detected, current environment information of a current area where the vehicle is located is obtained; the path planning instruction carries path planning information.
[0039] In one implementation of the embodiment of the present application, the on-board controller can receive in real time the path planning instructions autonomously sent by the autonomous driving system through the autonomous driving system to which it is wirelessly connected.
[0040] In another implementation of the embodiment of the present application, the vehicle-mounted controller can also receive in real time the path planning instructions sent by the user terminal through the user terminal to which it is wirelessly connected.
[0041] In another implementation of the embodiment of the present application, the vehicle-mounted controller can also receive in real time the path planning instructions sent by the external emergency maneuvering center through the external emergency maneuvering center to which it is wirelessly connected.
[0042] In another implementation of the embodiment of the present application, when the vehicle-mounted controller detects that the user in the vehicle performs a preset operation, such as clicking a preset control in the vehicle, the vehicle-mounted controller can detect the path planning instruction sent by the user.
[0043] In some possible embodiments, in order to facilitate subsequent analysis and tracing, the vehicle controller may record the received path planning instructions in detail.
[0044] In this embodiment, the log information recorded in the log includes but is not limited to: the source of the above-mentioned path planning instructions, the time of receipt, the analysis result, the verification result and the subsequent processing, etc. Such log information is of great significance for the maintenance, optimization and accident investigation of the system related to the path determination method in the vehicle.
[0045] In an embodiment of the present application, when the on-board controller detects a path planning instruction, it indicates that the vehicle's driving path needs to be planned at this time. Therefore, in order to achieve accurate planning of the driving path, the on-board controller can determine that the vehicle is in path planning mode at this time and obtain the current environmental information of the current area where the vehicle is located.
[0046] The current environment information includes, but is not limited to: road images, obstacle information, and weather conditions in the current area.
[0047] The road image refers to an image including lane lines and traffic signs in the current area.
[0048] Obstacle information includes but is not limited to: the position, size and movement status of other vehicles in the current area; the position, size and movement status of pedestrians in the current area, etc.
[0049] Weather conditions include but are not limited to: temperature, humidity, and visibility in the current area.
[0050] In one implementation of the embodiment of the present application, the vehicle controller can obtain the road image, obstacle information and weather conditions of the above current area in real time through various vehicle sensors connected to it by wireless communication. The vehicle sensors include but are not limited to: radar, camera, infrared sensor, temperature sensor and humidity sensor.
[0051] It should be noted that in the embodiment of the present application, the path planning instruction carries the path planning information. The path planning information includes, but is not limited to, the destination, the reason for the vehicle planning, and the urgency of the planning. Exemplarily, the reasons for the vehicle planning may include: regular driving, the user needs to rush to the destination urgently, a fire, flood or traffic accident occurs in the current area, etc. Regular driving specifically refers to the normal driving of the vehicle to the destination in the absence of any special circumstances (such as the user needs to rush to the destination urgently, a fire, flood or traffic accident occurs in the current area, etc.).
[0052] The planning urgency is used to characterize the urgency of the planning reason of the vehicle.
[0053] In one embodiment of the present application, after the vehicle controller detects the path planning instruction, in order to ensure the authenticity and reliability of the path planning instruction, so as to avoid the vehicle controller executing steps S101 to S105 based on a false path planning instruction, thereby avoiding wasting resources and increasing the energy consumption of the vehicle, the vehicle controller can specifically verify the path planning instruction through the following steps, which are described in detail as follows:
[0054] When the path planning instruction is detected, the integrity and format of the path planning instruction are detected to obtain a first detection result;
[0055] Acquiring identity information of a sender of the path planning instruction, and performing a legitimacy check on the identity information to obtain a second test result;
[0056] If the first detection result is that the verification is passed, and the second detection result is that the verification is passed, the current environment information is obtained.
[0057] In this embodiment, when the vehicle controller detects the path planning instruction, the vehicle controller can detect the integrity of the path planning instruction based on the preset rules and detect the format of the path planning instruction based on the preset format, thereby obtaining a first detection result of the path planning instruction. The preset rules and the preset format can be set according to actual needs and are not limited here.
[0058] In some possible embodiments, the preset rule may include: the type of the instruction complies with the set type and the length of the instruction complies with the set length. The set type and the set length may be determined according to actual needs and are not limited here.
[0059] In this embodiment, when the vehicle-mounted controller detects a path planning instruction, it can determine the type and length of the path planning instruction, compare the type with the set type to determine whether the type is the set type, and compare the length with the set length to determine whether the length is the set length.
[0060] At the same time, the on-board controller can detect whether the format of the path planning instruction is a preset format.
[0061] Afterwards, when the on-board controller detects that the type of the path planning instruction is the set type and the length of the path planning instruction is the set length, it can determine that the result of the integrity test of the path planning instruction is verification passed, that is, the path planning instruction complies with the preset rules; when the on-board controller detects that the type of the path planning instruction is the set type and the length of the path planning instruction is not the set length, it can determine that the result of the integrity test of the path planning instruction is verification failed, that is, the path planning instruction does not comply with the preset rules; when the on-board controller detects that the type of the path planning instruction is not the set type and the length of the path planning instruction is the set length, it can determine that the result of the integrity test of the path planning instruction is verification failed, that is, the path planning instruction does not comply with the preset rules; when the on-board controller detects that the type of the path planning instruction is not the set type and the length of the path planning instruction is not the set length, it can determine that the result of the integrity test of the path planning instruction is verification failed, that is, the path planning instruction does not comply with the preset rules.
[0062] When the on-board controller detects that the format of the path planning instruction is a preset format, it can determine that the result of detecting the format of the path planning instruction is verification passed.
[0063] In this embodiment, the first test result includes but is not limited to: verification passed and verification failed. Among them, the verification passed in the first test result means that the integrity and format of the path planning instruction have been verified, that is, the path planning instruction complies with the preset rules, and the format of the path planning instruction complies with the preset format. The verification failed in the first test result means that the integrity and / or format verification of the path planning instruction have failed, that is, the path planning instruction does not comply with the preset rules and / or the format of the path planning instruction does not comply with the preset format.
[0064] In this embodiment, the on-board controller may also obtain the identity information of the sender of the path planning instruction, and perform a legitimacy check on the identity information, thereby obtaining a second detection result of the path planning instruction.
[0065] Specifically, the vehicle controller can compare the acquired identity information with the set identity information to determine whether the identity information matches the set identity information, thereby obtaining a second detection result. The set identity information includes but is not limited to: the identity information of the vehicle owner and the identity information of the driver using the vehicle.
[0066] It should be noted that the second test result includes, but is not limited to: verification passed and verification failed. Among them, the verification passed in the second test result means that the legitimacy of the sender's identity information has been verified, that is, the sender's identity information is consistent with the set identity information. The verification failed in the second test result means that the legitimacy of the sender's identity information has not been verified, that is, the sender's identity information is inconsistent with the set identity information.
[0067] In this embodiment, when the on-board controller detects that the first detection result of the path planning instruction is verification passed, and the second detection result is verification passed, it means that the verification of the path planning instruction has passed, that is, the path planning instruction is authentic and reliable, that is to say, it is indeed necessary to plan the vehicle's driving path at this time. Therefore, the on-board controller can obtain the current environmental information of the current area where the vehicle is located.
[0068] In one embodiment of the present application, when the on-board controller detects that the first detection result of the path planning instruction is verification failure, and / or the second detection result is verification failure, it means that the verification of the path planning instruction has not been fully passed, that is, the path planning instruction does not have authenticity and / or reliability, that is, the path planning instruction received by the on-board controller at this time is false. Therefore, the on-board controller can output a prompt message for prompting the receipt of a false path planning instruction, so that relevant personnel can promptly conduct relevant investigations on the false path planning instruction.
[0069] In one implementation of this embodiment, the vehicle-mounted controller may output the above-mentioned prompt information through a vehicle-mounted microphone wirelessly connected thereto.
[0070] In another implementation of this embodiment, the vehicle-mounted controller may also display the above-mentioned prompt information through a vehicle-mounted display screen connected thereto wirelessly.
[0071] In another embodiment of the present application, since the vehicle-mounted controller obtains the above-mentioned current environmental information through different vehicle-mounted sensors, and different vehicle-mounted sensors have different measurement accuracy and error characteristics, the vehicle-mounted controller can fuse the information other than the road image in the above-mentioned current environmental information obtained, so that the data collected by different vehicle-mounted sensors are integrated into a unified and comprehensive environmental perception result, that is, the current environmental information after fusion processing is obtained.
[0072] In this embodiment, the vehicle controller may format the information except the road image in the above-mentioned current environment information to obtain different information with the same format. After that, the vehicle controller may splice the above-mentioned different information with the same format to obtain a unified and comprehensive environment perception result.
[0073] In another embodiment of the present application, since different vehicle-mounted sensors may be affected by various factors in actual use, such as temperature, humidity, electromagnetic interference, etc., which may cause certain errors in the data collected by different vehicle-mounted sensors, the vehicle-mounted controller may calibrate the above-obtained current environmental information one by one to improve the accuracy and consistency of the data, thereby obtaining the calibrated current environmental information.
[0074] In this embodiment, the vehicle controller can obtain the target threshold corresponding to each vehicle sensor according to the pre-stored correspondence between different vehicle sensors and amplitude thresholds, and calculate the sum / difference between the data collected by each vehicle sensor and its corresponding target threshold, so as to realize the calibration processing of the data collected by different vehicle sensors, and then the vehicle controller can determine each sum / difference as the calibrated current environment information. Among them, the target threshold corresponding to each vehicle sensor is used to describe the fluctuation value of the data collected by each vehicle sensor.
[0075] In S102, a plurality of candidate paths are determined based on the current environment information and the path planning information.
[0076] In the embodiment of the present application, after the vehicle controller obtains the current environment information and path planning information of the current area where the vehicle is located, the vehicle controller can determine multiple alternative paths based on the current environment information and path planning information. Each alternative path includes but is not limited to: main roads, secondary roads, auxiliary roads, and possible safe haven areas.
[0077] Specifically, the onboard controller can determine the road traffic conditions in the current area where the vehicle is located based on the current environmental information; at the same time, the onboard controller can determine the connectivity and passability of all possible paths based on the current environmental information and path planning information, thereby obtaining all possible passable paths for the vehicle. Afterwards, the onboard controller can determine multiple alternative paths from all possible passable paths based on the road traffic conditions in the area, that is, the drivable paths of the vehicle under the current road traffic conditions.
[0078] In one embodiment of the present application, in order to improve the accuracy of determining multiple candidate paths, the vehicle controller can specifically Figure 2 Steps S201 to S203 shown implement step S102, which are described in detail as follows:
[0079] In S201, image recognition processing is performed on the road image to determine lane lines and traffic signs of the road in the current area.
[0080] In this embodiment, since the current environmental information includes a road image, the on-board controller can perform image recognition processing on the road image to determine the lane lines and traffic signs contained in the road image, and further determine the lanes and traffic signs that exist in the current area where the vehicle is located.
[0081] It should be noted that the road image includes multiple frames.
[0082] In practical applications, traffic signs are also called road signs or road traffic signs. Traffic signs specifically refer to road facilities that use words or symbols to convey guidance, restrictions, warnings or instructions.
[0083] In S202, based on the weather condition, the road friction coefficient of the road and the visibility of the current area are determined.
[0084] In this embodiment, since the current environmental information also includes weather conditions, and the weather conditions include but are not limited to the temperature, humidity, and visibility of the current area, the vehicle controller can determine the road friction coefficient of the road based on the temperature and humidity of the current area and the dryness of the road. At the same time, the vehicle controller can also directly determine the visibility of the current area where the vehicle is located based on the visibility in the weather conditions.
[0085] In practical applications, the road friction coefficient is a parameter that measures the friction between the tires and the road surface when the vehicle is driving on the road. For example, the road friction coefficient of a good asphalt or concrete road surface is usually between 0.010 and 0.018; the road friction coefficient of a gravel road surface is between 0.020 and 0.025; the road friction coefficient of a compacted dirt road after rain is between 0.050 and 0.150; the road friction coefficient of a muddy dirt road (in the rainy season or thaw period) is between 0.100 and 0.250; the road friction coefficient of an icy road surface is lower, usually between 0.015 and 0.030.
[0086] Visibility refers to the maximum distance at which a person with normal vision can identify a target object from the background. That is, during the day, with the sky close to the horizon as the background, one can clearly see the outline of a dark target object on the ground with a viewing angle greater than 20 degrees and identify what it is, and at night, one can clearly see the light point of the target light.
[0087] In S203, path planning is performed on the current area based on the lane lines, the traffic signs, the obstacle information, the road friction coefficient, the visibility, and the path planning information to obtain a plurality of the candidate paths.
[0088] In this embodiment, the current environmental information also includes obstacle information, and the obstacle information will also affect the driving of the vehicle. Therefore, the on-board controller can determine multiple alternative paths based on lane lines, traffic signs, obstacle information, road friction coefficient, visibility and path planning information.
[0089] Specifically, the on-board controller can obtain all possible driving paths in the current area based on the lane lines and traffic signs in the current area; then, the on-board controller can perform obstacle avoidance planning for all possible paths based on the obstacle information to obtain various travel paths for the vehicle to bypass obstacles in all driving paths; then, the on-board controller can find multiple target travel paths with the destination from various travel paths based on the destination in the path planning information.
[0090] Since the road friction coefficient and visibility of roads at different locations in an area are not exactly the same, the road friction coefficient and visibility of the above-mentioned road include multiple sub-areas with different road friction coefficients and different visibility. Therefore, the on-board controller can determine the road friction coefficient and visibility corresponding to each target passage based on the position range of each target passage. Based on this, the on-board controller can determine the road friction coefficient range and visibility range required for vehicle driving at this time based on the planning reason and planning urgency in the path planning information, and based on the road friction coefficient range and visibility range, find multiple target paths within the road friction coefficient range and visibility range from each target passage, and determine the multiple target paths as alternative paths.
[0091] In one embodiment of the present application, when the path planning information includes the destination, the planning reason of the vehicle, and the urgency of the planning, in order to further improve the accuracy of determining multiple alternative paths, the vehicle controller can specifically implement step S203 according to the following steps, which are described in detail as follows:
[0092] Inputting the lane line, the traffic sign, the obstacle information and the destination into a trained path planning model for processing to obtain multiple feasible paths; the path planning model is trained by a pre-built neural network model;
[0093] Each of the feasible paths is scored based on the urgency, the planning reason, the road friction coefficient, and the visibility, and a plurality of alternative paths are selected from the plurality of feasible paths based on the scoring results of each feasible path.
[0094] In this embodiment, the vehicle controller can directly input lane lines, traffic signs, obstacle information and destinations into the trained path planning model for path planning, thereby obtaining multiple feasible paths for vehicles to pass. The path planning model is trained by a pre-built neural network model.
[0095] It should be noted that the path planning model can be obtained by training a pre-built neural network model based on a preset sample set. Among them, each sample data in the preset sample set includes sample information (the sample information includes lane lines, traffic signs, obstacle information and destination) and multiple sample paths corresponding to the sample information. When training the pre-built neural network model, the sample information in each sample data is used as the input of the neural network model, and the multiple sample paths corresponding to the sample information in each sample data are used as the output of the neural network model. Through training, the neural network model can learn the correspondence between all possible sample information and sample paths, and use the trained neural network model as the path planning model.
[0096] In some possible embodiments, different planning reasons have different requirements for roads. For example, when the planning reason is fire, the requirement is usually that the road cannot be close to the fire point. When the planning reason is flood, the requirement is usually that the road cannot be close to the terrain lower than the flood site. When the planning reason is regular driving, there is no requirement for the road. When the planning reason is that the user needs to rush to the destination urgently, the requirement is usually that the road surface is normal and continuous (that is, there will be no construction section or impassable section in the middle of the road). Therefore, the on-board controller can plan all possible paths based on the destination, planning reason and the current position of the vehicle. Afterwards, the on-board controller determines multiple feasible paths that the vehicle can pass from all possible paths based on lane lines, traffic signs and obstacle information.
[0097] Specifically, the onboard controller can determine the road section range from the current position of the vehicle to the destination, and then the onboard controller can find all possible paths from the road section range that will not affect the road due to planning reasons and thus hinder the vehicle's travel. For example, when the planning reason is a fire, it usually affects the road near the fire point. Therefore, the onboard controller can exclude the path points near the fire point from the road section range, and determine the paths composed of the remaining path points in the road section range except the above path points as possible paths.
[0098] Afterwards, since in actual applications vehicles need to comply with traffic regulations and drive accordingly according to lane lines and traffic signs, the on-board controller can eliminate paths that do not comply with traffic regulations from all the possible paths mentioned above based on lane lines and traffic signs.
[0099] Afterwards, since obstacles on the road (such as other vehicles, etc.) may also hinder the vehicle's travel, the on-board controller can determine a feasible path that will not collide with obstacles from various paths that comply with traffic regulations based on the acquired obstacle information.
[0100] In practical applications, since the road friction coefficient will affect the driving of the vehicle, the visibility will affect the driving sight of the vehicle, and the urgency of the planning will affect the driving time and speed of the vehicle, in this embodiment, the on-board controller can determine the road friction coefficient range and visibility range that will not affect the planning of the vehicle based on the urgency. Afterwards, since the road friction coefficient and visibility of the roads at different positions in an area are not exactly the same, that is, the road friction coefficient and visibility of the above-mentioned roads include multiple sub-areas with different road friction coefficients and different visibility, the on-board controller can determine the road friction coefficient and visibility corresponding to each feasible path based on the position range of each feasible path. Based on this, the on-board controller can find multiple target feasible paths within the road friction coefficient range and visibility range from the road friction coefficient and visibility corresponding to each feasible path, and determine the multiple target feasible paths as multiple alternative paths.
[0101] In S103, the path information of each candidate path is obtained, and each candidate path is evaluated based on the path information to obtain evaluation results of each candidate path in multiple different dimensions.
[0102] In an embodiment of the present application, after obtaining multiple alternative paths, the on-board controller can obtain the path information of each alternative path, and evaluate each alternative path based on the path information, to obtain evaluation results of each alternative path in multiple different dimensions, rather than only evaluating each alternative path based on the degree of road congestion.
[0103] It should be noted that the path information of each alternative path includes but is not limited to: the road condition information of each alternative path, the straight-line distance and actual driving distance between the starting point and the end point of each alternative path, the predicted driving time of each alternative path, and the road safety level of each alternative path.
[0104] The traffic information of each alternative route is used to describe the traffic conditions of each alternative route, including but not limited to: traffic flow, road congestion level, road surface conditions, etc.
[0105] The road safety level of each alternative route is used to characterize the safety of each road of each alternative route, the safety of the surrounding environment, and possible natural disaster risks (such as fire, etc.).
[0106] The multiple different dimensions include but are not limited to: road condition dimension, time dimension, distance dimension and road safety dimension.
[0107] In this embodiment, the vehicle controller can calculate the predicted travel time of each alternative route according to the actual travel distance of each alternative route and the expected speed of the vehicle, wherein the expected speed specifically refers to the speed that the vehicle is expected to reach under the corresponding planning reason.
[0108] In one embodiment of the present application, the on-board controller can evaluate the road condition information corresponding to each alternative path based on the road condition evaluation table, so as to obtain the evaluation results of each alternative path in the road condition dimension; the on-board controller can evaluate the straight-line distance and actual driving distance between the starting point and the end point of each alternative path based on the distance evaluation table, so as to obtain the evaluation results of each alternative path in the distance dimension; the on-board controller can evaluate the predicted driving time of each alternative path based on the time evaluation table, so as to obtain the evaluation results of each alternative path in the time dimension; the on-board controller can evaluate the road safety level of each alternative path based on the road safety evaluation table, so as to obtain the evaluation results of each alternative path in the road safety dimension.
[0109] It should be noted that the above-mentioned road condition assessment table can be composed of various detailed rules for assessing the road conditions in real life; the above-mentioned distance assessment table can be composed of various detailed rules for assessing the straight-line distance and actual driving distance of the vehicle when evacuating in real life; the above-mentioned time assessment table can be composed of various detailed rules for assessing the time of the vehicle when evacuating in real life; the above-mentioned road safety assessment table can be composed of various detailed rules for assessing the safety level of the road on which the vehicle is traveling when evacuating in real life.
[0110] In another embodiment of the present application, in order to improve the evaluation accuracy of the evaluation results of each candidate path in multiple different dimensions, the vehicle controller can specifically implement step S103 according to the following steps, which are described in detail as follows:
[0111] Evaluate each of the road condition information to obtain a first result of each of the candidate paths in terms of road condition;
[0112] Evaluate each of the straight-line distances and each of the actual driving distances to obtain a second result of each of the alternative paths in the distance dimension;
[0113] Evaluate each of the predicted travel times to obtain a third result of each of the alternative routes in the time dimension;
[0114] The safety level of each road is evaluated to obtain a fourth result of each alternative path in the road safety dimension.
[0115] In this embodiment, for any alternative path, the vehicle controller can evaluate the road condition dimension of the alternative path based on the road condition information in the path information of the alternative path, that is, evaluate the road traffic condition of the alternative path in the road condition dimension based on the road condition information in the path information of the alternative path, thereby obtaining the first result of the alternative path. The road traffic condition is used to describe the road passability of the alternative path.
[0116] It should be noted that the road accessibility includes but is not limited to: difficult, medium, good and easy, and the first result includes but is not limited to: difficult to pass, relatively difficult to pass, passable and easy to pass, etc.
[0117] In some possible embodiments, for any alternative path, when the on-board controller detects that the traffic flow of the alternative path is large, the road congestion is high, and the road surface condition is poor, the on-board controller may determine that the road traffic condition of the alternative path is difficult to pass, that is, the first result of the alternative path is difficult to pass. When the on-board controller detects that the traffic flow of the alternative path is large, the road congestion is high, and the road surface condition is medium, or the traffic flow of the alternative path is medium, the road congestion is high, and the road surface condition is poor, or the traffic flow of the alternative path is medium, the road congestion is medium, and the road surface condition is poor, or the traffic flow of the alternative path is medium, the road congestion is medium, and the road surface condition is medium, the on-board controller may determine that the road traffic condition of the alternative path is medium to pass, that is, the first result of the alternative path is difficult to pass. When the onboard controller detects that the traffic flow of the alternative path is medium, the road congestion is medium, and the road surface condition is good, or that the traffic flow of the alternative path is large, the road congestion is medium, and the road surface condition is good, the onboard controller can determine that the road traffic condition of the alternative path is that the road passability is good, that is, the first result of the alternative path is passable. When the onboard controller detects that the traffic flow of the alternative path is small, the road congestion is small, and the road surface condition is good, or that the traffic flow of the alternative path is small, the road congestion is small, and the road surface condition is medium, the onboard controller can determine that the road traffic condition of the alternative path is that the road passability is easy, that is, the first result of the alternative path is easy to pass.
[0118] Based on this, the on-board controller can obtain the first result of each alternative path in the road condition dimension.
[0119] In this embodiment, for any alternative path, the vehicle-mounted controller can respectively detect the first range of the straight-line distance in the path information of the alternative path and the second range of the actual driving distance. The vehicle-mounted controller can then evaluate the situation of the alternative path in the distance dimension based on the first range and the second range, thereby obtaining the second result of the alternative path.
[0120] It should be noted that the first range includes but is not limited to: a first setting range, a second setting range, and a third setting range. The second range includes but is not limited to: a fourth setting range, a fifth setting range, and a sixth setting range. The first setting range can be [0, first threshold), the second setting range can be [first threshold, second threshold), the third setting range can be [second threshold, third threshold), the fourth setting range can be [0, fourth threshold), the fifth setting range can be [fourth threshold, fifth threshold), and the sixth setting range can be [fifth threshold, sixth threshold). Among them, the first threshold is less than the second threshold, the second threshold is less than the third threshold, the fourth threshold is less than the fifth threshold, and the fifth threshold is less than the sixth threshold.
[0121] The second result includes, but is not limited to: farther distance, medium distance, closer distance, etc.
[0122] In some possible embodiments, for any alternative path, when the on-board controller detects that the first range of the straight-line distance of the alternative path is the first set range and the second range of the actual driving distance is the fourth set range, the on-board controller may determine that the second result of the alternative path is a short distance. When the on-board controller detects that the first range of the straight-line distance of the alternative path is the second set range and the second range of the actual driving distance is the fifth set range, or that the first range of the straight-line distance of the alternative path is the first set range and the second range of the actual driving distance is the fifth set range, or that the first range of the straight-line distance of the alternative path is the second set range and the second range of the actual driving distance is the fourth set range, the on-board controller may determine that the second result of the alternative path is a medium distance. When the vehicle controller detects that the first range of the straight-line distance of the alternative path is the third set range and the second range of the actual driving distance is the sixth set range, or that the first range of the straight-line distance of the alternative path is the third set range and the second range of the actual driving distance is the fifth set range, or that the first range of the straight-line distance of the alternative path is the second set range and the second range of the actual driving distance is the sixth set range, the vehicle controller can determine that the second result of the alternative path is a longer distance.
[0123] Based on this, the on-board controller can obtain the second result of each alternative path in the distance dimension.
[0124] In this embodiment, for any alternative path, the vehicle-mounted controller can detect the time range of the predicted driving time in the path information of the alternative path, and then the vehicle-mounted controller can evaluate the situation of the alternative path in the time dimension based on the above time range, so as to obtain the third result of the alternative path.
[0125] It should be noted that the time range includes but is not limited to: a first preset range, a second preset range, and a third preset range. The first preset range can be [0, first set time), the second preset range can be [first set time, second set time), and the third preset range can be [second set time, third set time). Among them, the first set time is less than the second set time, and the second set time is less than the third set time.
[0126] The third result includes but is not limited to: longer time, medium time, shorter time, etc.
[0127] In some possible embodiments, for any alternative path, when the on-board controller detects that the time range of the predicted driving time of the alternative path is within the first preset range, the on-board controller can determine that the third result of the alternative path in the time dimension is a shorter time; when the on-board controller detects that the time range of the predicted driving time of the alternative path is within the second preset range, the on-board controller can determine that the third result of the alternative path in the time dimension is a medium time; when the on-board controller detects that the time range of the predicted driving time of the alternative path is within the third preset range, the on-board controller can determine that the third result of the alternative path in the time dimension is a longer time.
[0128] Based on this, the on-board controller can obtain the third result of each alternative path in the time dimension.
[0129] In this embodiment, for any alternative path, the vehicle controller can directly evaluate the road safety dimension of the alternative path based on the road safety level in the path information of the alternative path, thereby obtaining the fourth result of the alternative path.
[0130] It should be noted that road safety levels include but are not limited to: high risk, medium risk, low risk and safe.
[0131] Correspondingly, the fourth result includes but is not limited to: high risk, medium risk, low risk and safe, etc.
[0132] In some possible embodiments, for any alternative path, when the on-board controller detects that the road safety level of the alternative path is high risk, the on-board controller may determine that the fourth result of the alternative path on the road safety dimension is high risk; when the on-board controller detects that the road safety level of the alternative path is medium risk, the on-board controller may determine that the fourth result of the alternative path on the road safety dimension is medium risk; when the on-board controller detects that the road safety level of the alternative path is low risk, the on-board controller may determine that the fourth result of the alternative path on the road safety dimension is low risk; when the on-board controller detects that the road safety level of the alternative path is safe, the on-board controller may determine that the fourth result of the alternative path on the road safety dimension is safe.
[0133] Based on this, the on-board controller can obtain the fourth result of each alternative path in the road safety dimension.
[0134] In S104, the current environment information and each path information are input into a trained path scoring model for processing to obtain a path score corresponding to each of the alternative paths; the path scoring model is trained based on a pre-built deep learning model.
[0135] In the embodiment of the present application, in order to further evaluate each candidate path so as to obtain a more accurate target driving path later, the vehicle controller can input the current environment information and each path information into the trained path scoring model for scoring, and obtain the path score corresponding to each candidate path. The path scoring model is trained based on a pre-built deep learning model.
[0136] It should be noted that the path scoring model can be obtained by training a pre-built deep learning model based on a preset sample set. Among them, each sample data in the preset sample set includes sample environment information and sample path information corresponding to each of the multiple sample candidate paths, as well as the sample environment information and the sample path information corresponding to each of the multiple sample candidate paths. When training the pre-built deep learning model, the sample environment information in each sample and the sample path information corresponding to each of the multiple sample candidate paths are used as the input of the deep learning model, and the sample environment information in each sample and the sample path information corresponding to each of the multiple sample candidate paths are used as the output of the deep learning model. Through training, the deep learning model can learn all possible sample environment information and multiple sample candidate paths, and the corresponding relationship between each path score, and use the trained deep learning model as the path scoring model.
[0137] In S105, each of the candidate paths is screened based on the evaluation results of each of the candidate paths in multiple different dimensions and the path score of each of the candidate paths to obtain a target driving path.
[0138] In an embodiment of the present application, after obtaining the evaluation results of each alternative path in multiple different dimensions and the path score of each alternative path, the on-board controller can determine the target driving path from multiple alternative paths based on the evaluation results of each alternative path in multiple different dimensions and the path scores of each alternative path.
[0139] It should be noted that the path planning of existing navigation apps and the like usually simply plans the vehicle's driving path based on the road congestion level. Unlike the prior art, the present application first determines multiple alternative paths based on the current environmental information and path planning information. Afterwards, the present application can also evaluate the conditions of each alternative path in different dimensions (not just the road congestion level in the prior art) based on the various path information, so that the on-board controller can clearly understand the evaluation conditions of each alternative path in different dimensions. At the same time, the on-board controller can also score each alternative path based on the current environmental information and each path information again, so as to further determine the feasibility of each alternative path under the environmental information and path information, so that the on-board controller can select the driving path that best suits the vehicle from multiple alternative paths.
[0140] In one embodiment of the present application, the vehicle controller may first determine at least one of the candidate paths whose evaluation results in multiple different dimensions meet the corresponding set dimension requirements as an optional path; then, the vehicle controller may determine the optional path corresponding to the maximum path evaluation value in the path score corresponding to at least one optional path as the target driving path. The set dimension requirements corresponding to each dimension can be set according to actual needs and are not limited here.
[0141] Exemplarily, when the dimensions include road condition dimension, distance dimension, time dimension and road safety dimension, the set dimensional requirement corresponding to the road condition dimension may be: the road passability is medium, general or low; the set dimensional requirement corresponding to the distance dimension may be: medium distance or short distance; the set dimensional requirement corresponding to the time dimension may be: medium time or short time; the set dimensional requirement corresponding to the road safety dimension may be: medium risk, low risk or safe.
[0142] In another embodiment of the present application, in order to improve the user experience and meet the user's evacuation needs, the vehicle controller can specifically implement step S105 according to the following steps, which are described in detail as follows:
[0143] Acquiring planning demand information of users in the vehicle;
[0144] Based on the evaluation results of each of the candidate paths in multiple different dimensions and the scores of each of the paths, a first path that matches the planning requirement information is searched from the multiple candidate paths, and the first path is determined as the target driving path.
[0145] In this embodiment, after obtaining the evaluation results of each candidate path in multiple dimensions and the scores of each path, the onboard controller can also obtain the requirements of the user in the vehicle for the vehicle's driving path in real time, that is, planning demand information. The planning demand information includes but is not limited to: road condition requirements, distance requirements, time requirements, and road safety requirements.
[0146] In one implementation of this embodiment, a user in a vehicle can input the above-mentioned planning requirement information in real time in an input interface of a user terminal wirelessly connected to the vehicle controller, and send the planning requirement information to the vehicle controller. Afterwards, the vehicle controller can receive the planning requirement information sent by the user terminal.
[0147] In another implementation of this embodiment, the user in the vehicle can also input the above-mentioned planning requirement information in real time through the display screen in the vehicle. Based on this, the vehicle controller can receive the planning requirement information input by the user in real time through the display screen connected to the vehicle controller wirelessly.
[0148] In this embodiment, after obtaining the user's planning requirement information, the vehicle controller can first obtain multiple optional paths with path scores greater than the set score from each candidate path. The set score can be determined according to actual needs and is not limited here.
[0149] Afterwards, the on-board controller can compare the planning requirement information with the evaluation results of each optional path in multiple different dimensions one by one, and determine the first path with the greatest match between the evaluation results of each optional path in multiple different dimensions and the planning requirement information as the vehicle's final target driving path.
[0150] From the above, it can be seen that a path determination method provided by an embodiment of the present application obtains current environmental information of the current area where the vehicle is located when a path planning instruction is detected; the path planning instruction carries path planning information; based on the current environmental information and the path planning information, multiple alternative paths are determined; the path information of each alternative path is obtained, and each alternative path is evaluated based on each path information to obtain evaluation results of each alternative path in multiple different dimensions; the current environmental information and each of the path information are input into a trained path scoring model for processing to obtain a path score corresponding to each alternative path; each alternative path is screened based on the evaluation results of each alternative path in multiple different dimensions and the path score of each alternative path to obtain a target driving path. Compared with the prior art, this method can combine the current environment information and the path planning information to accurately determine multiple alternative paths that match the vehicle. After that, on the one hand, the situation of each alternative path in different dimensions can be evaluated in combination with each path information. On the other hand, each alternative path can be further evaluated in combination with the current environment information and each path information, thereby achieving accurate evaluation of each alternative path. Finally, combining the evaluation results of each alternative path in multiple different dimensions and the scores of each path also achieves accurate determination of the target driving path, that is, the driving path that best matches the vehicle can be selected from multiple alternative paths, thereby improving the accuracy of path determination.
[0151] See also Figure 3 , Figure 3 is a flowchart of a path determination method provided by another embodiment of the present application. Figure 1 Corresponding embodiment, this embodiment may further include S301 to S306 after S101, which are described in detail as follows:
[0152] In S301, the vehicle is controlled to travel based on the target driving path.
[0153] In S302, target environment information and vehicle driving information of a target area where the vehicle is located are obtained.
[0154] In S303, the road condition of the target area is analyzed according to the target environment information to obtain the target road condition of the target area.
[0155] In S304, the state of the vehicle in the target area is analyzed according to the vehicle driving information to obtain the target state of the vehicle.
[0156] In S305, the target driving path is adjusted according to the target road condition and / or the target state.
[0157] In S306, the vehicle is controlled to travel based on the adjusted target driving path.
[0158] In this embodiment, after determining the target driving path of the vehicle, the on-board controller can control the vehicle to travel based on the target driving path to complete the evacuation requirements of the vehicle.
[0159] Afterwards, when the on-board controller is controlling the vehicle to travel based on the target driving path, in order to determine in real time whether the target driving path meets the actual needs, that is, whether the target driving path needs to be adjusted accordingly, the on-board controller can obtain in real time the target environment information and vehicle driving information of the target area where the vehicle is located during the process of the vehicle traveling based on the target driving path.
[0160] The target environment information includes, but is not limited to: target road images, target obstacle information, and target weather conditions in the target area.
[0161] Vehicle driving information includes but is not limited to: vehicle fuel level and vehicle damage level, etc.
[0162] In this embodiment, the vehicle controller can analyze the road conditions of the target area according to the target environment information, that is, determine the traffic conditions and road conditions of the target area based on the target environment information, thereby obtaining the target road conditions of the target area. The target road conditions include but are not limited to: good, medium and poor.
[0163] Specifically, when the on-board controller detects that the traffic condition in the target area is good and the road surface condition is good, the on-board controller can determine that the target road condition in the target area is good; when the on-board controller detects that the traffic condition in the target area is average and the road surface condition is good, or that the traffic condition in the target area is average and the road surface condition is average, the on-board controller can determine that the target road condition in the target area is medium; when the on-board controller detects that the traffic condition in the target area is poor and the road surface condition is poor, or that the traffic condition in the target area is poor and the road surface condition is average, the on-board controller can determine that the target road condition in the target area is poor.
[0164] In this embodiment, the vehicle controller can analyze the state of the vehicle in the target area according to the vehicle driving information, so as to obtain the target state of the vehicle, wherein the target state includes but is not limited to: good, average and poor.
[0165] Specifically, when the on-board controller detects that the vehicle has sufficient fuel and the degree of damage to the vehicle is relatively light, the on-board controller can determine that the target state of the vehicle is good; when the on-board controller detects that the vehicle has medium fuel and the degree of damage to the vehicle is relatively light, or that the vehicle has medium fuel and the degree of damage to the vehicle is medium, the on-board controller can determine that the target state of the vehicle is fair; when the on-board controller detects that the vehicle has medium fuel and the degree of damage to the vehicle is relatively heavy, or that the vehicle has insufficient fuel and the degree of damage to the vehicle is medium, the on-board controller can determine that the target state of the vehicle is poor.
[0166] In this embodiment, after obtaining the target road condition of the target area and the target state of the vehicle, the onboard controller can adjust the target driving path according to the target road condition and / or target state.
[0167] Specifically, when the onboard controller detects that the target road condition is good and the target state is average or poor, the onboard controller can adjust the target driving path based on the target state of the vehicle so that the vehicle can get as close to the destination as possible while still being able to drive.
[0168] When the on-board controller detects that the target road condition is medium or poor and the target state is good, the on-board controller can adjust the target driving path based on the target road condition of the target area so that the vehicle can travel on a safer road.
[0169] Afterwards, the on-board controller can control the vehicle to travel based on the adjusted target driving path.
[0170] As can be seen from the above, the path determination method provided by this embodiment can control the vehicle to travel based on the target driving path after determining the target driving path; and obtain the target environment information and vehicle driving information of the target area where the vehicle is located; determine the target road condition of the target area according to the target environment information; determine the target state of the vehicle according to the vehicle driving information; adjust the target driving path according to the target road condition and / or target state; and control the vehicle to travel based on the adjusted target driving path. When controlling the vehicle to travel based on the target driving path, this method can obtain the target environment information and vehicle driving information of the target area where the vehicle is located during the driving process in real time, so that the road condition and the real-time state of the vehicle can be determined in real time, and then it can be determined in real time whether the target driving path needs to be adjusted according to the determined road condition and the real-time state of the vehicle, and when adjustment is required, the target driving path is adjusted accordingly in real time, so that the vehicle can travel based on the adjusted target driving path, thereby further improving the planning accuracy of the vehicle's driving path and the driving safety of the vehicle.
[0171] In another embodiment of the present application, taking the need for evacuation of a vehicle as an example, the onboard controller can determine the user status of the user in the vehicle in real time during the process of executing steps S101 to S10 and controlling the vehicle to travel based on the target driving path, and output safety warning information and / or evacuation guidance information for prompting the user based on the user status to improve the personal safety of the user in the vehicle. The evacuation guidance information includes but is not limited to information such as guiding the user to find a safe exit and prompting the user to use emergency escape tools.
[0172] In another embodiment of the present application, in the case of emergency evacuation, the user in the vehicle may encounter the danger of being unable to evacuate by himself. At this time, the vehicle controller can trigger the emergency help function, that is, send a help signal to the external rescue system through the vehicle communication system.
[0173] It should be noted that the above-mentioned help signal includes but is not limited to the vehicle's location information, user status and required rescue resources, so that the external rescue system can quickly locate and launch rescue operations.
[0174] In another embodiment of the present application, the vehicle controller can also establish a real-time communication link with external organizations such as traffic management systems, fire protection systems, and medical systems. Through this link, the vehicle controller can obtain the latest road conditions, traffic control measures, rescue resources and other key information in real time, providing strong support for the target driving path of the vehicle.
[0175] For example, if a vehicle needs to rush to a destination (such as a hospital) in an emergency, in the case of traffic congestion, the on-board controller can request the traffic management system to adjust the timing of traffic lights or open an emergency passage; if a fire or accident occurs and the vehicle needs to be evacuated, the on-board controller can request the fire department corresponding to the fire system to provide fire extinguishing and rescue support; if there are injuries, the on-board controller can request the medical department corresponding to the medical system to provide emergency medical assistance.
[0176] In another embodiment of the present application, the vehicle controller may have an information sharing function. For example, if a vehicle needs to be evacuated, the vehicle controller may share information such as the real-time location of the vehicle, the status of the vehicle, and the user information of the users in the vehicle with an external organization, so that the external organization can better understand the evacuation situation and make corresponding decisions.
[0177] In another embodiment of the present application, the vehicle controller also includes an evaluation and feedback function after the vehicle reaches the destination. Under this function, the vehicle controller can provide strong support for improving the planning efficiency and safety of the vehicle by automatically collecting and analyzing data during the vehicle's driving process.
[0178] Specifically, the vehicle controller can obtain various key data of the vehicle while it is traveling along the target driving path. Among them, the key data include but are not limited to: vehicle dynamic data such as the vehicle's driving trajectory, speed, acceleration, and passenger data such as the physiological state (such as heart rate, blood pressure), behavioral response (such as action sequence in an emergency), and reaction time of the user in the vehicle. These key data will serve as the basis for evaluating the planning effect of the vehicle's driving path.
[0179] Afterwards, the onboard controller can conduct in-depth analysis and mining of the key data collected above. For example, the onboard controller can use existing data analysis technology and algorithms to conduct multi-dimensional evaluation of the vehicle's driving process, including driving time, route selection, occupant safety, warning system performance, etc. At the same time, the onboard controller can also identify bottlenecks in the driving process, occupant reaction patterns, and potential safety hazards.
[0180] Afterwards, the on-board controller can generate a detailed and practical evaluation report based on the results of the above data analysis. The evaluation report can clearly present the various data indicators and evaluation results of the vehicle during driving, and is accompanied by corresponding charts and visual displays. In addition, the evaluation report will also put forward targeted improvement suggestions and optimization plans, including improving the driving path, optimizing the triggering conditions of the warning system, and improving the emergency response capabilities of the occupants, so that car manufacturers, rescue agencies and other relevant parties can understand the system performance and make targeted improvements.
[0181] Afterwards, the vehicle controller can convey the above evaluation report to relevant parties through a fast and effective feedback mechanism. The feedback methods include but are not limited to email notification, online report viewing, etc., to ensure that relevant parties can obtain the evaluation report in a timely manner and take corresponding actions.
[0182] Afterwards, in addition to the evaluation and feedback after a single planning, the on-board controller can also continuously accumulate experience and optimize the evaluation algorithm based on the evaluation reports of multiple plans to improve the accuracy and effectiveness of the evaluation.
[0183] It should be noted that in this embodiment, when the vehicle controller collects and analyzes key data, the vehicle controller needs to strictly abide by the principles of data security and privacy protection. For example, the vehicle controller can use encryption technology and permission control mechanism to ensure the security of the above data during transmission and storage. At the same time, the vehicle controller can also perform desensitization processing on the collected data to protect the personal privacy of the vehicle user.
[0184] In another embodiment of the present application, the vehicle controller may also include an intelligent learning and optimization function. Under this function, the vehicle controller assumes the important responsibility of continuously learning and optimizing path planning to ensure that efficient and safe decisions can be made in the face of various complex situations.
[0185] Specifically, the onboard controller can continuously collect data from the planning process, which includes multiple information such as vehicle status, user reactions in the vehicle, external environment, etc. After cleaning and integration, the above data is used to train and optimize internal models, such as the path scoring model.
[0186] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0187] Corresponding to a path determination method described in the above embodiment, Figure 4 The schematic diagram of the structure of a path determination device provided by an embodiment of the present application is shown. For the convenience of explanation, only the part related to the embodiment of the present application is shown. Figure 4 The path determination device 400 includes: a first acquisition unit 41, a first path determination unit 42, an evaluation unit 43, an input unit 44 and a second path determination unit 45. Among them:
[0188] The first acquisition unit 41 is used to acquire current environment information of the current area where the vehicle is located when a path planning instruction is detected; the path planning instruction carries the path planning information.
[0189] The first path determination unit 42 is used to determine a plurality of candidate paths based on the current environment information and the path planning information.
[0190] The evaluation unit 43 is used to obtain the path information of each candidate path, and evaluate each candidate path based on the path information to obtain evaluation results of each candidate path in multiple different dimensions.
[0191] The input unit 44 is used to input the current environment information and each path information into a trained path scoring model for processing to obtain a path score corresponding to each of the alternative paths; the path scoring model is trained based on a pre-built deep learning model.
[0192] The second path determination unit 45 is used to screen each of the candidate paths based on the evaluation results of each of the candidate paths in multiple different dimensions and the path scores of each of the candidate paths to obtain a target driving path.
[0193] In one embodiment of the present application, the current environment information includes the road image, obstacle information and weather conditions of the current area; the first path determination unit 42 specifically includes: a first information determination unit, a second information determination unit and a third path determination unit. Among them:
[0194] The first information determination unit is used to perform image recognition processing on the road image to determine lane lines and traffic signs of the road in the current area.
[0195] The second information determination unit is used to determine the road friction coefficient of the road and the visibility of the current area based on the weather condition.
[0196] The third path determination unit is used to perform path planning for the current area based on the lane line, the traffic sign, the obstacle information, the road friction coefficient, the visibility and the path planning information to obtain the multiple alternative paths.
[0197] In one embodiment of the present application, the path planning information includes the destination, the planning reason of the vehicle and the urgency of the planning; the third path determination unit specifically includes: a fourth path determination unit and a fifth path determination unit. Among them:
[0198] The fourth path determination unit is used to input the lane line, the traffic sign, the obstacle information and the destination into a trained path planning model for processing to obtain multiple feasible paths; the path planning model is trained by a pre-constructed neural network model.
[0199] The fifth path determination unit is used to score each of the feasible paths based on the urgency, the planning reason, the road friction coefficient and the visibility, and select multiple alternative paths from the multiple feasible paths based on the scoring results of each feasible path.
[0200] In one embodiment of the present application, the path information includes: the road condition information of each of the alternative paths, the straight-line distance and actual travel distance between the starting point and the end point of each of the alternative paths, the predicted travel time of each of the alternative paths, and the road safety level of each of the alternative paths; the evaluation unit 43 specifically includes: a first result determination unit, a second result determination unit, a third result determination unit, and a fourth result determination unit. Among them:
[0201] The first result determination unit is used to evaluate each of the road condition information for any candidate path, and obtain a first result of each of the candidate paths in the road condition dimension.
[0202] The second result determination unit is used to evaluate each of the straight-line distances and each of the actual driving distances to obtain a second result of each of the alternative paths in the distance dimension.
[0203] The third result determination unit is used to evaluate each of the predicted travel times to obtain a third result of each of the alternative paths in the time dimension.
[0204] The fourth result determination unit is used to evaluate the safety level of each of the roads to obtain a fourth result of each of the alternative paths in terms of road safety.
[0205] In one embodiment of the present application, the second path determination unit 45 specifically includes: a second acquisition unit and a sixth path determination unit.
[0206] The second acquisition unit is used to acquire planning demand information of the user in the vehicle.
[0207] The sixth path determination unit is used to search for a first path that matches the planning requirement information from the multiple alternative paths based on the evaluation results of each of the alternative paths in multiple different dimensions and the scores of each path, and determine the first path as the target driving path.
[0208] In one embodiment of the present application, the second path determination unit 45 specifically includes: an input unit and a seventh path determination unit.
[0209] The input unit is used to input the evaluation results of each of the alternative paths in multiple different dimensions into the trained path scoring model for processing in sequence to obtain a path probability matrix; the path scoring model is trained based on a pre-built deep learning model; the value of each element in the path probability matrix is used to represent the probability value of the alternative path corresponding to each element.
[0210] The seventh path determination unit is used to determine the alternative path corresponding to the element with the largest probability value in the path probability matrix as the target driving path.
[0211] In one embodiment of the present application, the path determination device 400 further includes: a first control unit, a third acquisition unit, a road condition determination unit, a state determination unit, an adjustment unit, and a second control unit.
[0212] The first control unit is used to control the vehicle to travel based on the target driving path.
[0213] The third acquisition unit is used to acquire target environment information and vehicle driving information of the target area where the vehicle is located.
[0214] The road condition determination unit is used to analyze the road condition of the target area according to the target environment information to obtain the target road condition of the target area.
[0215] The state determination unit is used to analyze the state of the vehicle in the target area according to the vehicle driving information to obtain the target state of the vehicle.
[0216] The adjustment unit is used to adjust the target driving path according to the target road condition and / or the target state.
[0217] The second control unit is used to control the vehicle to travel based on the adjusted target driving path.
[0218] In one embodiment of the present application, the first acquisition unit 41 specifically includes: a first detection unit, a second detection unit and a fourth acquisition unit. Among them:
[0219] The first detection unit is used to detect the integrity and format of the path planning instruction when detecting the path planning instruction to obtain a first detection result.
[0220] The second detection unit is used to obtain the identity information of the sender of the path planning instruction, and perform a legitimacy detection on the identity information to obtain a second detection result.
[0221] The fourth acquisition unit is used to acquire the current environment information if the first detection result is verified to be passed and the second detection result is verified to be passed.
[0222] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.
[0223] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0224] Figure 5 This is a schematic diagram of the structure of a vehicle controller provided in one embodiment of the present application. Figure 5 As shown, the vehicle controller 5 of this embodiment includes: at least one processor 50 ( Figure 5 Only one is shown in the figure) a processor, a memory 51, and a computer program 52 stored in the memory 51 and executable on the at least one processor 50, and when the processor 50 executes the computer program 52, the steps in any of the above-mentioned path determination method embodiments are implemented.
[0225] The vehicle-mounted controller may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art will appreciate that Figure 5 It is only an example of the vehicle-mounted controller 5 and does not constitute a limitation of the vehicle-mounted controller 5. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.
[0226] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0227] In some embodiments, the memory 51 may be an internal storage unit of the vehicle controller 5, such as the memory of the vehicle controller 5. In other embodiments, the memory 51 may also be an external storage device of the vehicle controller 5, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the vehicle controller 5. Further, the memory 51 may also include both the internal storage unit of the vehicle controller 5 and an external storage device. The memory 51 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 51 may also be used to temporarily store data that has been output or is to be output.
[0228] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.
[0229] An embodiment of the present application provides a computer program product. When the computer program product runs on a vehicle-mounted controller, the vehicle-mounted controller can implement the steps in the above-mentioned method embodiments when executing the computer program product.
[0230] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A path determination method, It is characterized in that include: When a path planning instruction is detected, current environment information of the current area where the vehicle is located is obtained; The path planning instruction carries path planning information; Determining a plurality of candidate paths based on the current environment information and the path planning information; Obtaining path information of each candidate path, and evaluating each candidate path based on the path information to obtain evaluation results of each candidate path in multiple different dimensions; Inputting the current environment information and each path information into a trained path scoring model for processing to obtain a path score corresponding to each candidate path; The path scoring model is trained based on a pre-built deep learning model; Based on the evaluation results of each candidate path in multiple different dimensions and the path score of each candidate path, each candidate path is screened to obtain a target driving path.
2. The path determination method according to claim 1, It is characterized in that The current environment information includes a road image, obstacle information, and weather conditions in the current area; and determining a plurality of alternative paths based on the current environment information and the path planning information includes: Performing image recognition processing on the road image to determine lane lines and traffic signs of the road in the current area; Based on the weather condition, determining a road friction coefficient of the road and visibility of the current area; Based on the lane lines, the traffic signs, the obstacle information, the road friction coefficient, the visibility and the path planning information, path planning is performed on the current area to obtain a plurality of the alternative paths.
3. The path determination method according to claim 2, It is characterized in that The path planning information includes a destination, a planning reason for the vehicle, and a planning urgency; the path planning is performed on the current area based on the lane line, the traffic sign, the obstacle information, the road friction coefficient, the visibility, and the path planning information to obtain a plurality of the alternative paths, including: Inputting the lane line, the traffic sign, the obstacle information and the destination into a trained path planning model for processing to obtain multiple feasible paths; the path planning model is trained by a pre-built neural network model; Each of the feasible paths is scored based on the urgency, the planning reason, the road friction coefficient, and the visibility, and a plurality of alternative paths are selected from the plurality of feasible paths based on the scoring results of each feasible path.
4. The path determination method according to claim 1, It is characterized in that The path information includes: road condition information of each of the alternative paths, the straight-line distance and actual travel distance between the starting point and the end point of each of the alternative paths, the predicted travel time of each of the alternative paths, and the road safety level of each of the alternative paths; the evaluation of each of the alternative paths based on each of the path information is performed to obtain evaluation results of each of the alternative paths in multiple different dimensions, including: Evaluate each of the road condition information to obtain a first result of each of the candidate paths in terms of road condition; Evaluate each of the straight-line distances and each of the actual driving distances to obtain a second result of each of the alternative paths in the distance dimension; Evaluate each of the predicted travel times to obtain a third result of each of the alternative routes in the time dimension; The safety level of each road is evaluated to obtain a fourth result of each alternative path in the road safety dimension.
5. The path determination method according to claim 1, It is characterized in that The step of screening each of the candidate paths based on the evaluation results of each of the candidate paths in multiple different dimensions and the path score of each of the candidate paths to obtain a target driving path includes: Acquiring planning demand information of users in the vehicle; Based on the evaluation results of each of the candidate paths in multiple different dimensions and the scores of each of the paths, a first path that matches the planning requirement information is searched from the multiple candidate paths, and the first path is determined as the target driving path.
6. The path determination method according to any one of claims 1 to 5, It is characterized in that After screening each of the candidate paths based on the evaluation results of each of the candidate paths in multiple different dimensions and the path score of each of the candidate paths to obtain the target driving path, the method further includes: Controlling the vehicle to travel based on the target driving path; Accordingly, after controlling the vehicle to travel based on the target driving path, the method further includes: Acquiring target environment information and vehicle driving information of a target area where the vehicle is located; Analyzing the road conditions of the target area according to the target environment information to obtain the target road conditions of the target area; Analyzing the state of the vehicle in the target area according to the vehicle driving information to obtain the target state of the vehicle; Adjusting the target driving path according to the target road condition and / or the target state; The vehicle is controlled to travel based on the adjusted target travel path.
7. The path determination method according to any one of claims 1 to 5, It is characterized in that When the path planning instruction is detected, the current environment information of the current area where the vehicle is located is obtained, including: When the path planning instruction is detected, the integrity and format of the path planning instruction are detected to obtain a first detection result; Acquiring identity information of a sender of the path planning instruction, and performing a legitimacy check on the identity information to obtain a second test result; If the first detection result is that the verification is passed, and the second detection result is that the verification is passed, the current environment information is obtained.
8. A path determination device, It is characterized in that include: A first acquisition unit, configured to acquire current environment information of a current area where the vehicle is located when a path planning instruction is detected; The path planning instruction carries path planning information; A first path determination unit, configured to determine a plurality of candidate paths based on the current environment information and the path planning information; An evaluation unit, configured to obtain path information of each candidate path, and evaluate each candidate path based on the path information to obtain evaluation results of each candidate path in multiple different dimensions; An input unit, used for inputting the current environment information and each path information into a trained path scoring model for processing, so as to obtain a path score corresponding to each candidate path; The path scoring model is trained based on a pre-built deep learning model; The second path determination unit is used to screen each of the candidate paths based on the evaluation results of each of the candidate paths in multiple different dimensions and the path scores of each of the candidate paths to obtain a target driving path.
9. A vehicle-mounted controller comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the path determination method according to any one of claims 1 to 7 is implemented.
10. A vehicle, It is characterized in that Comprising the vehicle-mounted controller as claimed in claim 9.
Citation Information
Cited By
Path planning method and device based on adaptive heuristic
CN119671007A
Robot control method and system based on department demands
CN119704209A
Intelligent vehicle path planning method and device, intelligent vehicle and program product
CN121855568A
