Vehicle path planning method and system based on big data

Abstract

The invention provides a vehicle path planning method and system based on big data, and the method comprises the steps: generating a first planning path and a second planning path which are opposite in driving direction and are consistent in starting point, and dynamically planning the first planning path according to the different passing data of the same road section in the first planning path and the second planning path, so that the passing cost of the target vehicle from the driving starting point to the driving ending point is the lowest or the passing time is the shortest. Equivalently, road sections which may be unfavorable for vehicle driving are identified according to different traffic data in the same road section, and then the driving path of the vehicle is planned by using the big data, so that the urban traffic flow can be reasonably regulated and controlled. Therefore, existing road resources and navigation resources can be fully utilized, the road traffic burden can be relieved, and traffic jam and traffic accidents are reduced. Meanwhile, fuel consumption of the user can be reduced, and the traffic cost of the vehicle user is saved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data and path planning, and in particular to a vehicle path planning method and system based on big data. Background Art

[0002] With the rapid development of the economy, the number of motor vehicles in my country has increased year by year and the increment has been increasing; coupled with the acceleration of urbanization, the concentration of vehicles has increased, and the growing number of cars has made the burden of road traffic systems in many places increasingly heavier, while people's demand for a more convenient traffic environment has continued to increase. This contradiction has led to many traffic problems and even larger social contradictions. The increasingly serious traffic problems have attracted widespread attention from all sectors of society and have become a focus of public concern.

[0003] At present, most of my country's road systems cannot fully match urban planning, resulting in traffic congestion and inconvenience. On the other hand, some roads cannot be fully utilized, resulting in waste of resources, which in turn increases the traffic burden. It can be seen that when conducting urban traffic management, it is necessary not only to carry out reasonable road construction planning, but also to reasonably regulate urban traffic flow so that resources can fully play their due role. As for vehicle travel, the navigation system is an indispensable and important function of today's vehicles. However, the current navigation system generally does not consider the impact of traffic conditions on vehicle driving when planning the route, nor does it predict the future traffic conditions and adjust the route accordingly, which makes it easy for traffic to concentrate on traffic congestion sections or sections with possible congestion. Therefore, how to avoid too many vehicles concentrating on certain sections to cause congestion, effectively divert urban traffic flow, and reduce traffic congestion and traffic accidents is a problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a vehicle path planning method and system based on big data, so as to solve the technical problems existing in the prior art.

[0005] To achieve the above objectives and other related objectives, the present invention provides a vehicle path planning method based on big data, comprising the following steps:

[0006] Based on a road network data model established in advance or in real time, a first planned path for the target vehicle to travel from a starting point to a destination, and a second planned path for the target vehicle to travel from the destination to a starting point are generated;

[0007] During the driving process of the target vehicle, the traffic data of the first planned path and the second planned path are collected respectively, and the traffic data includes: traffic flow data, road obstacle data, traffic accident data and weather condition data;

[0008] Comparing the traffic data of the first planned path with the traffic data of the second planned path, identifying traffic data with differences in the same road section, and recording the corresponding road section as a road section to be adjusted;

[0009] According to the path planning constraints determined in advance or in real time, the first planned path is dynamically planned through the road section to be adjusted so that the target vehicle continues to drive to the destination along the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time.

[0010] Optionally, the process of dynamically planning the first planned path through the road section to be adjusted includes:

[0011] Acquire all the spare sections that reach the end point of the road section to be adjusted based on the start point of the road section to be adjusted in the direction from the driving start point to the driving end point;

[0012] Prioritize and combine the road sections according to the road section distance, congestion coefficient, road safety, road section charges, road section obstacle status, and road section road grade, and generate the driving section corresponding to each combination through the spare road sections;

[0013] Displaying the driving section corresponding to each combination through the on-board display screen of the target vehicle, and receiving the sorting combination selected by the user based on the display result, which is recorded as a standby sorting combination;

[0014] The travel section corresponding to the backup sorting combination is used to replace the section to be adjusted, and all remaining untraveled sections after dynamic planning are associated to obtain a dynamically planned path.

[0015] Optionally, if the user does not select a sorting combination on the vehicle-mounted display screen, the first planned path is used as a dynamic planned path;

[0016] Alternatively, when there is no spare road section corresponding to the road section to be adjusted in the direction from the driving starting point to the driving end point, the first planned path is used as the dynamic planned path.

[0017] Optionally, the road network model includes:

[0018] Lu = (X, Y, Z);

[0019] Where Lu represents the road network model;

[0020] X represents the intersection of the road segment or the starting and ending points of the road, where X = {1, 2, 3, ...};

[0021] Y represents the road segment connecting two nodes, where Y = {(i, j)|i, j∈A, i≠j};

[0022] Z represents the weight set of the road section, and the weight in the road network model is related to the time period of the road section, where Z = {z ij (n)|(i, j)∈B, n∈N}, N is a natural number;

[0023] i represents the predecessor node of the road segment, j represents the successor node of the road segment, and (i, j) represents the road segment;

[0024] z ij (n) represents the weight of the road segment (i, j) in time period n.

[0025] Optionally, based on a road network data model established in advance or in real time, generating a first planned path for a target vehicle to travel from a driving starting point to a driving end point, and generating a second planned path for the target vehicle to travel from the driving end point to a driving starting point includes:

[0026] Obtaining a road network in the road network model, and setting a grade attribute for distinguishing low-grade roads from high-grade roads, and adding a road area number to all roads in the road network and assigning it to be empty; wherein the high-grade roads and the low-grade roads are determined based on road section distance, congestion coefficient, road safety, road section charges, road section obstacle status, and road section road grade;

[0027] A road section node is randomly selected from all roads as a target node. When the level attribute of the target node is less than the level attribute h of the high-level road, the road area number of the target node is assigned to A, and all adjacent nodes connected to the target node and whose road area number values ​​are empty are queried and added to the preset dictionary g; and

[0028] Take a road node x in the preset dictionary g. If the level attribute of the road node x is less than the level attribute h of the high-level road, and the road area number of the road node x is empty, then set the road area number of the road node x to A, and query all adjacent road nodes connected to the road node x and whose road area number values ​​are empty, add them to the preset dictionary g, and remove the road node x from the preset dictionary g until the preset dictionary g is an empty set;

[0029] Let A=A+1, and randomly select a road node with an empty road area number from all roads and continue to execute the loop until the road area numbers of all roads are not empty, thus obtaining a road partition network;

[0030] The road partition network is used to generate a first planned path for the target vehicle to travel from a driving start point to a driving end point, and a second planned path for the target vehicle to travel from the driving end point to a driving start point is generated.

[0031] Optionally, the process of generating a first planned path for the target vehicle from a driving start point to a driving end point by using the road partition network, and generating a second planned path for the target vehicle from a driving end point to a driving start point includes:

[0032] Determine a first road node and a second road node based on the driving direction of the target vehicle; wherein the first road node and the second road node of the first planned path are respectively the driving starting point and the driving end point, and the first road node and the second road node of the second planned path are respectively the driving end point and the driving starting point;

[0033] Determining whether the driving starting point and the driving end point belong to the same road area number;

[0034] If they belong to the same road area number, the minimum cost path from the first road node to the second road node is directly generated by searching the roads corresponding to the same road area number to obtain the first planned path or the second planned path; wherein the minimum cost path includes: the lowest driving path cost after combining the road section distance, congestion coefficient, road safety, road section toll, road section obstacle status and road section road grade;

[0035] If they do not belong to the same road area number, then based on the level attribute h of the high-level road, search for the optimal entry node P1 and the optimal exit node P2 between the first road node and the second road node, and based on the road area number where the first road node is located, search for the minimum cost path of the optimal entry node P1 and the optimal exit node P2 to obtain the first planned path; and, based on the road area number where the second road node is located, search for the minimum cost path of the optimal entry node P1 and the optimal exit node P2 to obtain the second planned path.

[0036] Optionally, during the driving of the target vehicle, the method further includes:

[0037] The target vehicle senses a preset distance range through a sensor pre-installed on the target vehicle to determine whether there is an obstacle within the preset distance range;

[0038] If there is an obstacle, a prompt message is issued, and based on the current driving state of the target vehicle, the motion state of the obstacle, and the real-time distance of the obstacle from the target vehicle, the path of the target vehicle is changed to obtain an obstacle avoidance driving path, and the obstacle avoidance driving path is displayed on the on-board display screen of the target vehicle;

[0039] If there is no obstacle, the target vehicle is controlled to maintain the current driving state.

[0040] Optionally, after the target vehicle reaches the driving destination according to the first planned path, the method further includes:

[0041] Based on the traffic data of the first planned path and the second planned path during the driving process of the target vehicle, the corresponding traffic data is converted into a matrix describing the change of the average traffic speed according to the time characteristics and the space characteristics, which is recorded as the target matrix;

[0042] The batch size of the neural network is set based on the target matrix, the neural network is initialized, and the learning rate and loss function are set according to the network structure, number of neurons and weight vector of the neural network; wherein one neuron corresponds to one traffic influence factor;

[0043] Calculate the similarity between each neuron and the weight vector, and take the neuron corresponding to the maximum similarity as the target neuron;

[0044] Updating the weight vector of the target neuron, as well as updating the learning rate and the loss function, and after completing the update, determining whether the learning rate is less than a preset value, or determining whether the number of updates reaches a preset number of iterations;

[0045] If the learning rate is less than a preset value, or the number of updates reaches a preset number of iterations, each neuron is marked, and the traffic influencing factor of the traffic data is found based on the marked neurons;

[0046] If the learning rate is greater than or equal to a preset value and the number of updates does not reach a preset number of iterations, the similarity between each neuron and the weight vector is recalculated, and the learning rate or the number of updates is continued to be determined until the traffic influencing factor of the traffic data is found.

[0047] Optionally, the traffic influencing factors of the traffic data include at least one of the following: traffic flow data, road obstacle data, traffic accident data, and weather condition data;

[0048] Alternatively, the traffic influencing factors of the traffic data include at least one of the following: section distance, congestion coefficient, road safety, section toll, section obstacle status, and section road grade.

[0049] The present invention also provides a vehicle path planning system based on big data, the system comprising:

[0050] An initial path module is used to generate a first planned path for a target vehicle from a driving start point to a driving end point according to a road network data model established in advance or in real time, and to generate a second planned path for the target vehicle from a driving end point to a driving start point;

[0051] A traffic data collection module, used to collect the traffic data of the first planned path and the second planned path respectively during the driving process of the target vehicle, wherein the traffic data includes: traffic flow data, road obstacle data, traffic accident data and weather condition data;

[0052] A traffic data comparison module, used to compare the traffic data of the first planned path with the traffic data of the second planned path, identify traffic data with differences in the same road section, and record the corresponding road section as a road section to be adjusted;

[0053] The path dynamic planning module is used to dynamically plan the first planned path through the to-be-adjusted road section according to the path planning constraints determined in advance or in real time, so that the target vehicle continues to drive to the destination along the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time.

[0054] As described above, the present invention provides a vehicle path planning method and system based on big data, which has the following beneficial effects: the present invention first generates a first planned path for a target vehicle to travel from a starting point to a destination based on a road network data model established in advance or in real time, and generates a second planned path for a target vehicle to travel from a destination to a starting point; then, during the driving process of the target vehicle, the traffic data of the first planned path and the traffic data of the second planned path are collected respectively, and the traffic data include: traffic flow data, road obstacle data, traffic accident data and weather condition data; then, the traffic data of the first planned path and the traffic data of the second planned path are compared, and the traffic data with differences in the same section are identified, and the corresponding section is recorded as a section to be adjusted; finally, according to the path planning constraints determined in advance or in real time, the first planned path is dynamically planned through the section to be adjusted, so that the target vehicle continues to travel to the destination according to the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time. It can be seen that the present invention generates two first and second planned paths with opposite driving directions and the same starting point, and then dynamically plans the first planned path according to the different traffic data of the same section in the first and second planned paths, so that the target vehicle can drive from the starting point to the end point with the lowest traffic cost or the shortest travel time. It is equivalent to the present invention identifying the sections that may be unfavorable for vehicle driving (such as construction area sections, accident-prone sections, traffic accident sections, traffic congestion sections, etc.) with different traffic data in the same section, and then plans the vehicle's driving path by using big data, so that the urban traffic flow can be reasonably regulated. Therefore, the present invention can not only make full use of existing road resources and navigation resources, but also reduce the burden of road traffic, reduce traffic jams and traffic accidents. At the same time, the present invention can help users reduce fuel consumption and save vehicle users' transportation costs by providing them with a driving path with the lowest travel cost or the shortest travel time. In addition, the present invention helps reduce vehicle emissions and reduce environmental pollution by optimizing the driving path and reducing congestion, and path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of a process flow of a vehicle path planning method based on big data provided by an embodiment of the present invention;

[0056] Figure 2 A schematic diagram of the hardware structure of a vehicle path planning system based on big data provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0058] It should be noted that the illustrations provided in this embodiment are only used to illustrate the basic concept of the present invention in a schematic manner. Therefore, the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0059] See also Figure 1 As shown, this embodiment provides a vehicle path planning method based on big data, comprising the following steps:

[0060] Based on the road network data model established in advance or in real time, a first planned path for the target vehicle to travel from the starting point to the end point of travel is generated, and a second planned path for the target vehicle to travel from the end point to the starting point of travel is generated. As an example, in this embodiment or other embodiments, the target vehicle includes but is not limited to new energy vehicles used by ordinary users, fuel vehicles used by ordinary users, and emergency vehicles used by non-ordinary users. As an example, in this embodiment or other embodiments, the road network data model established in advance or in real time includes: Lu = (X, Y, Z); wherein Lu represents the road network model; X represents the intersection of the road section or the starting and ending point of the road, wherein X = {1, 2, 3, ...}; Y represents the road section connecting two nodes, wherein Y = {(i, j) | i, j∈A, i≠j}; Z represents the weight set of the road section, and the weight in the road network model is related to the time period in which the road section is located, wherein Z = {z ij (n)|(i, j)∈B, n∈N}, N is a natural number; i represents the predecessor node of the road segment, j represents the successor node of the road segment, (i, j) represents the road segment; z ij (n) represents the weight of the road segment (i, j) in time period n.

[0061] During the driving process of the target vehicle, the traffic data of the first planned path and the second planned path are collected respectively, and the traffic data includes: traffic flow data, road obstacle data, traffic accident data and weather condition data;

[0062] Comparing the traffic data of the first planned path with the traffic data of the second planned path, identifying traffic data with differences in the same road section, and recording the corresponding road section as a road section to be adjusted;

[0063] According to the path planning constraints determined in advance or in real time, the first planned path is dynamically planned through the road section to be adjusted so that the target vehicle continues to drive to the destination along the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time.

[0064] Specifically, this embodiment can accurately plan the best driving path from the starting point to the end point for the target vehicle based on the road network data model established in advance or in real time. That is, the first planned path from the starting point to the end point of the target vehicle is generated as a navigation route in the usual sense; secondly, in order to cope with possible return needs or two-way traffic flow analysis, a second planned path from the end point back to the starting point is also generated. During the actual driving process of the target vehicle, this embodiment will collect the traffic data on the first planned path and the second planned path in real time. These traffic data cover traffic flow data, which is used to analyze the busyness of the road section; road obstacle data, such as construction areas, temporary closures, etc., which affect vehicle traffic; traffic accident data, which reflects the impact of sudden road conditions on traffic; and weather condition data, such as the impact of rain, snow, haze and other weather on driving conditions and visibility. After collecting these data, this embodiment will conduct a detailed comparison and analysis of these data. By comparing the traffic data of the first planned path and the second planned path on the same road section, it is possible to accurately identify the sections with differences. These sections often have potential problems or changes and require further attention and processing. Therefore, these sections are marked as sections to be adjusted and become the focus of subsequent dynamic planning. For the sections to be adjusted, this embodiment will perform dynamic planning according to the path planning constraints determined in advance or in real time. These constraints are intended to ensure that the target vehicle can continue to travel along the optimal route to achieve the goal of minimum travel cost or shortest travel time. During the dynamic planning process, this embodiment will adjust the planned path according to real-time data, avoid congested sections, bypass obstacles, reduce the impact of traffic accidents, and consider the impact of weather conditions on driving conditions, so as to ensure that the target vehicle can reach the destination safely and quickly.

[0065] It can be seen that this embodiment generates two first and second planned paths with opposite driving directions and the same starting point, and then dynamically plans the first planned path according to the different traffic data of the same section in the first and second planned paths, so that the target vehicle can drive from the starting point to the end point with the lowest traffic cost or the shortest travel time. This is equivalent to identifying the sections that may be unfavorable for vehicle driving (such as construction area sections, accident-prone sections, traffic accident sections, traffic congestion sections, etc.) with different traffic data in the same section in this embodiment, and then plans the vehicle's driving path by using big data, so that the urban traffic flow can be reasonably regulated. Therefore, this embodiment can not only make full use of existing road resources and navigation resources, but also reduce the burden of road traffic, reduce traffic congestion and traffic accidents. At the same time, this embodiment can help users reduce fuel consumption and save vehicle users' transportation costs by providing them with a driving path with the lowest travel cost or the shortest travel time. In addition, this embodiment helps reduce vehicle emissions and environmental pollution by optimizing the driving path and reducing congestion, and path planning. Therefore, this embodiment can reflect changes in road conditions in real time, ensuring the accuracy and effectiveness of the planned path. Secondly, by comparing and analyzing the traffic data of different paths, potential problems can be discovered and adjusted in a timely manner, improving the flexibility and adaptability of path planning. Finally, the setting of path planning constraints enables this embodiment to plan according to different needs and preferences, meeting the personalized needs of different users.

[0066] According to the above record, in an exemplary embodiment, the process of dynamically planning the first planned path through the road section to be adjusted includes:

[0067] Acquire all the spare sections that reach the end point of the road section to be adjusted based on the start point of the road section to be adjusted in the direction from the driving start point to the driving end point;

[0068] Prioritize and combine the road sections according to the road section distance, congestion coefficient, road safety, road section charges, road section obstacle status, and road section road grade, and generate the driving section corresponding to each combination through the spare road sections;

[0069] Displaying the driving section corresponding to each combination through the on-board display screen of the target vehicle, and receiving the sorting combination selected by the user based on the display result, which is recorded as a standby sorting combination;

[0070] The travel section corresponding to the backup sorting combination is used to replace the section to be adjusted, and all remaining untraveled sections after dynamic planning are associated to obtain a dynamically planned path.

[0071] Among them, if the user does not select a sorting combination on the vehicle-mounted display screen, the first planned path will be used as the dynamic planned path; or, when there is no spare section corresponding to the section to be adjusted in the direction from the starting point to the end point, the first planned path will be used as the dynamic planned path.

[0072] Specifically, the present embodiment first obtains all possible spare sections from the starting point to the end point of the section to be adjusted in the direction from the driving starting point to the driving end point. It ensures that the vehicle user has enough alternatives to deal with possible changes in road conditions, thereby improving the flexibility of path planning. Secondly, the present embodiment prioritizes and combines these spare sections. The criteria for the priority ranking combination include section distance (reflecting the total driving time), congestion coefficient (reflecting the busyness of the road), road safety (taking into account factors such as accident-prone areas), section charges (for sections that need to pay tolls), section obstacle status (such as construction, accidents, etc.) and section road grade (such as expressways, main roads, etc.). This multi-dimensional evaluation method ensures that the present embodiment can comprehensively consider various factors, so as to select the most suitable driving route for the target vehicle. Secondly, the present embodiment generates the driving section corresponding to each combination by using these spare sections, and displays this information to the driver through the on-board display screen in the target vehicle. Not only does it improve the driver's participation, but it also enables the driver to choose the most suitable driving route according to his or her preferences and needs. Then, after the driver selects a sorting combination based on the displayed results, this embodiment records it as a spare sorting combination, and uses the driving section corresponding to this combination to replace the original section to be adjusted. It reflects the real-time and dynamic nature of path planning, so that the driving route can be adjusted in real time according to real-time road condition information and driver feedback. Finally, this embodiment forms a complete dynamic planning path by associating all the remaining untraveled sections after dynamic planning. This path not only takes into account the current road condition information, but also combines the driver's preferences and needs, so it can ensure that the target vehicle can continue to travel along the optimal route. In addition, if the user does not select a sorting combination on the vehicle-mounted display screen, the first planned path is used as the dynamic planning path; or, if there is no spare section corresponding to the section to be adjusted in the direction from the starting point to the end point, the first planned path is used as the dynamic planning path.

[0073] It can be seen that the advantage of the dynamic planning process provided by the present embodiment is its comprehensiveness and flexibility. Through multi-dimensional evaluation criteria and real-time road condition information, the present embodiment can select the most suitable route for the target vehicle to ensure efficient and safe driving. At the same time, through the participation and feedback of the driver, the present embodiment can better meet the personalized needs of the driver and improve the user experience of the present embodiment. In addition, this dynamic planning process can also be adjusted in real time according to changes in road conditions to ensure the real-time and accuracy of the route.

[0074] According to the above records, in an exemplary embodiment, based on a road network data model established in advance or in real time, a first planned path for a target vehicle to travel from a driving start point to a driving end point is generated, and a process of generating a second planned path formed by the target vehicle traveling from the driving end point to the driving start point includes: obtaining a road network in the road network model, and setting a grade attribute used to distinguish between low-grade roads and high-grade roads, and adding a road area number to all roads in the road network and assigning it to empty; wherein the high-grade roads and the low-grade roads are determined based on road section distance, congestion coefficient, road safety, road section charges, road section obstacle status and road section grade; randomly selecting a road section node from all roads as a target node, and when the grade attribute of the target node is less than the grade attribute h of the high-grade road, assigning the road area number of the target node to A, and querying all neighboring nodes connected to the target node and whose road area number values ​​are empty. Connect nodes and add them to the preset dictionary g; and, take a road node x in the preset dictionary g, if the level attribute of the road node x is less than the level attribute h of the high-level road, and the road area number of the road node x is assigned to be empty, then set the road area number of the road node x to be A, and query all adjacent road nodes connected to the road node x and with empty road area number values, add them to the preset dictionary g, and remove the road node x from the preset dictionary g until the preset dictionary g is an empty set; set A=A+1, and randomly select a road node with an empty road area number from all roads to continue the loop execution until the road area numbers of all roads are not empty, and terminate to obtain a road partition network; use the road partition network to generate a first planned path for the target vehicle to drive from the driving starting point to the driving end point, and generate a second planned path for the target vehicle to drive from the driving end point to the driving starting point.

[0075] Specifically, the present embodiment first obtains the road network in the road network model, which includes all roads in the road network model and basic information of each road, such as length, width, geographical location, etc. In order to plan the path more accurately, the present embodiment distinguishes low-level roads from high-level roads by setting a grade attribute. The determination of this grade attribute is usually based on multiple factors, such as section distance, congestion coefficient, road safety, section charges, section obstacle status, and section road grade. High-level roads often have shorter driving distances, lower congestion, higher road safety, fewer charges and obstacles, and higher road grades. Secondly, the present embodiment adds a road area number to all roads in the road network and initializes it to a null value. The introduction of this road area number is to divide the road network into different areas to facilitate subsequent path planning. Then, the present embodiment starts to construct a road partition network by randomly selecting a section node as the starting target node. In the process of constructing the road partition network, the present embodiment checks the grade attribute of the target node. If the grade attribute of the target node is less than the preset high-level road grade attribute h, then the node will be classified as a road area, and the road area number will be assigned to A. Subsequently, this embodiment queries all adjacent nodes connected to the target node and whose road area numbers have not been assigned, and adds these nodes to a preset dictionary g. Next, this embodiment takes out a road node x from the dictionary g and checks its grade attribute again. If the grade attribute of the road node x is also less than the grade attribute h of the higher-grade road, and its road area number has not been assigned, then the node will also be classified as the same road area and assigned the same road area number A. At the same time, this embodiment continues to query all adjacent road nodes connected to the road node x and whose road area numbers have not been assigned, and adds them to the dictionary g. After completing these operations, the road node x will be removed from the dictionary g. This process will be repeated until the dictionary g is an empty set, that is, all connected road nodes are classified into the corresponding road area, so that all road nodes with road area number A can be obtained. Then, this embodiment increases the road area number A by 1, which is equivalent to entering the road area number assignment process of the next road. A road node whose road area number has not been assigned is randomly selected from all roads, and the above node classification and dictionary update process are repeated. This process will continue until the road area numbers of all roads are assigned, that is, the entire road network is completely divided into different road areas. Finally, using this divided road partition network, a first planned path for the target vehicle from the driving start point to the driving end point and a second planned path from the driving end point to the driving start point are generated. Since the roads have been divided into different areas, this embodiment can give priority to high-level roads and roads in the same area, thereby generating a more reasonable and efficient driving path.

[0076] It can be seen that this embodiment can more accurately evaluate the capacity and driving efficiency of different roads by distinguishing between low-level roads and high-level roads, so as to select a more suitable driving route. Secondly, through the division of road areas, this embodiment can divide the entire road network into several relatively independent areas, which is convenient for subsequent path search and planning. This not only reduces the search range, but also improves planning efficiency. Finally, this method combines the basic information of the road with the real-time traffic conditions, and can generate a driving path that is more in line with the actual situation, thereby improving the accuracy and practicality of path planning.

[0077] According to the above record, in an exemplary embodiment, the process of using the road partition network to generate a first planned path for the target vehicle from the driving start point to the driving end point, and generating a second planned path for the target vehicle from the driving end point to the driving start point includes:

[0078] Determine a first road node and a second road node based on the driving direction of the target vehicle; wherein the first road node and the second road node of the first planned path are respectively the driving starting point and the driving end point, and the first road node and the second road node of the second planned path are respectively the driving end point and the driving starting point;

[0079] Determining whether the driving starting point and the driving end point belong to the same road area number;

[0080] If they belong to the same road area number, the minimum cost path from the first road node to the second road node is directly generated by searching the roads corresponding to the same road area number to obtain the first planned path or the second planned path; wherein the minimum cost path includes: the lowest driving path cost after combining the road section distance, congestion coefficient, road safety, road section toll, road section obstacle status and road section road grade;

[0081] If they do not belong to the same road area number, then based on the level attribute h of the high-level road, search for the optimal entry node P1 and the optimal exit node P2 between the first road node and the second road node, and based on the road area number where the first road node is located, search for the minimum cost path of the optimal entry node P1 and the optimal exit node P2 to obtain the first planned path; and, based on the road area number where the second road node is located, search for the minimum cost path of the optimal entry node P1 and the optimal exit node P2 to obtain the second planned path.

[0082] Specifically, the present embodiment first determines the first road node and the second road node according to the driving direction of the target vehicle. In the first planned path, the first road node is the driving start point and the second road node is the driving end point; while in the second planned path, the roles of the two nodes are opposite, thereby ensuring that the starting point and the end point of the path planning always match the driving needs of the vehicle. Next, by judging whether the driving start point and the driving end point belong to the same road area number, if the starting point and the end point are in the same area, the vehicle can reach the destination more directly through the road network in the area without crossing other areas. If they belong to the same road area number, the minimum cost path from the first road node to the second road node will be directly searched in the area. The "minimum cost" here can be a comprehensive indicator, for example, multiple factors such as road section distance, congestion coefficient, road safety, road section toll, road section obstacle status, and road section road grade can be considered to ensure that the generated path is short, safe, and economical. If the driving start point and the driving end point do not belong to the same road area number, a more complex path search strategy can be adopted. For example, based on the grade attribute h of the high-grade road, the present embodiment searches for the optimal entry node P1 and the optimal exit node P2 between the two areas. These two nodes are usually key nodes on high-level roads connecting two areas, and they can ensure that vehicles cross the regional boundaries at a higher speed and with lower risk. After determining the optimal entry node P1 and the optimal exit node P2, the minimum cost path of P1 can be searched in the road area number where the first road node is located, and the minimum cost path of P2 can be searched in the road area number where the second road node is located. These two paths will be connected to P1 and P2 respectively to form a complete first planned path and a second planned path.

[0083] It can be seen that this embodiment can search and plan paths more efficiently by pre-dividing road areas, reducing unnecessary calculations. Secondly, using high-grade roads as bridges connecting different areas can ensure that vehicles can enjoy better road conditions and higher driving efficiency when crossing regional boundaries. Finally, by comprehensively considering multiple factors to determine the minimum cost path, it is possible to generate a high-quality path that meets both vehicle driving needs and the actual road conditions, thereby improving the accuracy and practicality of path planning.

[0084] In an exemplary embodiment, during the driving process of the target vehicle, this embodiment may also include: sensing a preset distance range through a sensor pre-installed on the target vehicle to determine whether there is an obstacle within the preset distance range; if there is an obstacle, issuing a prompt message, and based on the current driving state of the target vehicle, the motion state of the obstacle, and the real-time distance of the obstacle from the target vehicle, changing the path of the target vehicle to obtain an obstacle avoidance driving path, and displaying the obstacle avoidance driving path through an on-board display screen in the target vehicle; if there is no obstacle, controlling the target vehicle to maintain the current driving state.

[0085] Specifically, during the driving process of the target vehicle, in order to ensure driving safety and improve the driving experience, this embodiment can monitor the preset driving distance range in real time through sensors pre-installed on the target vehicle, such as laser radar, camera and ultrasonic sensor. These sensors can not only accurately capture road information, but also detect slight changes in the surrounding environment. When the sensor senses that there is an obstacle within the preset distance range, the vehicle control system will immediately enter a high alert state. At the same time, the vehicle control system will send clear prompt information to the driver, which can be conveyed in various forms such as sound, light or vibration to ensure that the driver can quickly perceive potential dangers. The vehicle control system will then calculate based on the current driving state of the target vehicle (such as speed, acceleration, steering angle, etc.), the movement state of the obstacle (such as speed, direction, acceleration, etc.), and the real-time distance of the obstacle from the target vehicle. This calculation process fully considers various possible driving scenarios and quickly generates an optimal obstacle avoidance driving path. This path can not only ensure that the vehicle safely avoids obstacles, but also minimize time loss and unnecessary vehicle operations during driving. Once the obstacle avoidance driving path is determined, it can be displayed in real time through the on-board display screen in the target vehicle. This display screen is usually located within the driver's line of sight and can clearly display the vehicle's driving trajectory, obstacle locations, obstacle avoidance paths and other information. Through this display, the driver can intuitively understand the vehicle's driving status and make corresponding operations as needed. If the sensor does not detect any obstacles within the preset distance range, it will determine that the current driving environment is safe and control the target vehicle to maintain the current driving status. This intelligent judgment and processing method can not only reduce the driver's driving pressure, but also improve the vehicle's driving efficiency and safety.

[0086] It can be seen that this embodiment can sense and process various situations in the driving process in real time by using sensors pre-installed on the vehicle and advanced algorithms, thereby ensuring the driving safety and driving experience of the vehicle. In addition, this embodiment can also make adaptive adjustments based on the driver's driving habits and driving environment to provide more personalized services.

[0087] In an exemplary embodiment, after the target vehicle reaches the driving destination according to the first planned path, the embodiment may further include:

[0088] Based on the traffic data of the first planned path and the second planned path during the driving process of the target vehicle, the corresponding traffic data is converted into a matrix describing the change of the average traffic speed according to the time characteristics and the space characteristics, which is recorded as the target matrix;

[0089] The batch size of the neural network is set based on the target matrix, the neural network is initialized, and the learning rate and loss function are set according to the network structure, number of neurons and weight vector of the neural network; wherein one neuron corresponds to one traffic influence factor;

[0090] Calculate the similarity between each neuron and the weight vector, and take the neuron corresponding to the maximum similarity as the target neuron;

[0091] Updating the weight vector of the target neuron, as well as updating the learning rate and the loss function, and after completing the update, determining whether the learning rate is less than a preset value, or determining whether the number of updates reaches a preset number of iterations;

[0092] If the learning rate is less than a preset value, or the number of updates reaches a preset number of iterations, each neuron is marked, and the traffic influencing factor of the traffic data is found based on the marked neurons;

[0093] If the learning rate is greater than or equal to the preset value, and the number of updates does not reach the preset number of iterations, the similarity between each neuron and the weight vector is recalculated, and the learning rate or the number of updates is continuously determined until the traffic influencing factor of the traffic data is found. The traffic influencing factor of the traffic data includes at least one of the following: traffic flow data, road obstacle data, traffic accident data, and weather condition data; or, the traffic influencing factor of the traffic data includes at least one of the following: road section distance, congestion coefficient, road safety, road section toll, road section obstacle status, and road section road grade.

[0094] It can be seen that this embodiment can deeply understand the changing patterns of traffic flow by collecting and analyzing a large amount of traffic data, so as to make more accurate predictions and decisions. Moreover, this embodiment can automatically learn and adjust parameters by using neural networks for data analysis to improve the accuracy and efficiency of predictions. Moreover, this embodiment can dynamically adjust the driving path and parameter settings according to real-time traffic data to achieve real-time traffic optimization. In addition, this embodiment also takes into account a variety of traffic influencing factors, including traffic flow, road obstacles, traffic accidents and weather conditions, so as to comprehensively evaluate traffic conditions and make reasonable decisions. Therefore, this embodiment can easily add new data sources and influencing factors to meet the needs of different scenarios.

[0095] In summary, the present invention provides a vehicle path planning method based on big data. First, based on a road network data model established in advance or in real time, a first planned path for a target vehicle to travel from a starting point to a destination is generated, and a second planned path for a target vehicle to travel from the destination to a starting point is generated; then, during the driving process of the target vehicle, the traffic data of the first planned path and the traffic data of the second planned path are collected respectively, and the traffic data include: traffic flow data, road obstacle data, traffic accident data and weather condition data; then, the traffic data of the first planned path and the traffic data of the second planned path are compared, and the traffic data with differences in the same road section are identified, and the corresponding road section is recorded as a road section to be adjusted; finally, according to the path planning constraints determined in advance or in real time, the first planned path is dynamically planned through the road section to be adjusted, so that the target vehicle continues to travel to the destination according to the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time. It can be seen that the method generates two first and second planning paths with opposite driving directions and the same starting point, and then dynamically plans the first planning path according to the different traffic data of the same section in the first and second planning paths, so that the target vehicle can drive from the starting point to the end point with the lowest travel cost or the shortest travel time. This is equivalent to the method identifying the sections that may be unfavorable for vehicle driving (such as construction area sections, accident-prone sections, traffic accident sections, traffic congestion sections, etc.) with different traffic data in the same section, and then plans the vehicle's driving path by using big data, so that the urban traffic flow can be reasonably regulated. Therefore, the method can not only make full use of existing road resources and navigation resources, but also reduce the burden of road traffic, reduce traffic jams and traffic accidents. At the same time, the method can help users reduce fuel consumption and save the transportation costs of vehicle users by providing them with the lowest travel cost or the shortest travel time. In addition, the method helps reduce vehicle emissions and environmental pollution by optimizing the driving path and reducing congestion, and the path planning helps reduce vehicle emissions and environmental pollution.

[0096] like Figure 2 As shown, in another exemplary embodiment of the present invention, the embodiment also provides a vehicle path planning system based on big data, including:

[0097] The initial path module is used to generate a first planned path for the target vehicle from the driving start point to the driving end point according to a road network data model established in advance or in real time, and to generate a second planned path formed by the target vehicle driving from the driving end point to the driving start point. As an example, in this embodiment or other embodiments, the target vehicle includes but is not limited to new energy vehicles used by ordinary users, fuel vehicles used by ordinary users, and emergency vehicles used by non-ordinary users. As an example, in this embodiment or other embodiments, the road network data model established in advance or in real time includes: Lu = (X, Y, Z); wherein Lu represents the road network model; X represents the intersection of the road section or the starting and ending point of the road, wherein X = {1, 2, 3, ...}; Y represents the road section connecting two nodes, wherein Y = {(i, j) | i, j∈A, i≠j}; Z represents the weight set of the road section, and the weight in the road network model is related to the time period in which the road section is located, wherein Z = {z ij (n)|(i, j)∈B, n∈N}, N is a natural number; i represents the predecessor node of the road segment, j represents the successor node of the road segment, (i, j) represents the road segment; z ij (n) represents the weight of the road segment (i, j) in time period n.

[0098] A traffic data collection module, used to collect the traffic data of the first planned path and the second planned path respectively during the driving process of the target vehicle, wherein the traffic data includes: traffic flow data, road obstacle data, traffic accident data and weather condition data;

[0099] A traffic data comparison module, used to compare the traffic data of the first planned path with the traffic data of the second planned path, identify traffic data with differences in the same road section, and record the corresponding road section as a road section to be adjusted;

[0100] The path dynamic planning module is used to dynamically plan the first planned path through the to-be-adjusted road section according to the path planning constraints determined in advance or in real time, so that the target vehicle continues to drive to the destination along the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time.

[0101] Specifically, this embodiment can accurately plan the best driving path from the starting point to the end point for the target vehicle based on the road network data model established in advance or in real time. That is, the first planned path from the starting point to the end point of the target vehicle is generated as a navigation route in the usual sense; secondly, in order to cope with possible return needs or two-way traffic flow analysis, a second planned path from the end point back to the starting point is also generated. During the actual driving process of the target vehicle, this embodiment will collect the traffic data on the first planned path and the second planned path in real time. These traffic data cover traffic flow data, which is used to analyze the busyness of the road section; road obstacle data, such as construction areas, temporary closures, etc., which affect vehicle traffic; traffic accident data, which reflects the impact of sudden road conditions on traffic; and weather condition data, such as the impact of rain, snow, haze and other weather on driving conditions and visibility. After collecting these data, this embodiment will conduct a detailed comparison and analysis of these data. By comparing the traffic data of the first planned path and the second planned path on the same road section, it is possible to accurately identify the sections with differences. These sections often have potential problems or changes and require further attention and processing. Therefore, these sections are marked as sections to be adjusted and become the focus of subsequent dynamic planning. For the sections to be adjusted, this embodiment will perform dynamic planning according to the path planning constraints determined in advance or in real time. These constraints are intended to ensure that the target vehicle can continue to travel along the optimal route to achieve the goal of minimum travel cost or shortest travel time. During the dynamic planning process, this embodiment will adjust the planned path according to real-time data, avoid congested sections, bypass obstacles, reduce the impact of traffic accidents, and consider the impact of weather conditions on driving conditions, so as to ensure that the target vehicle can reach the destination safely and quickly.

[0102] It should be noted that the vehicle path planning system based on big data provided in the above embodiment and the vehicle path planning method based on big data provided in the above embodiment belong to the same concept, wherein the specific manner in which the vehicle path planning method based on big data performs operations has been described in detail in the above method embodiment, and will not be repeated here. In actual applications, the vehicle path planning system based on big data provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the system into different functional modules, so as to complete all or part of the functions described above through the above method embodiment, which is not limited here.

[0103] In summary, the present invention provides a vehicle path planning system based on big data. First, based on a road network data model established in advance or in real time, a first planned path for a target vehicle to travel from a starting point to a destination is generated, and a second planned path for a target vehicle to travel from the destination to a starting point is generated; then, during the driving process of the target vehicle, the traffic data of the first planned path and the traffic data of the second planned path are collected respectively, and the traffic data include: traffic flow data, road obstacle data, traffic accident data and weather condition data; then, the traffic data of the first planned path and the traffic data of the second planned path are compared, and the traffic data with differences in the same road section are identified, and the corresponding road section is recorded as a road section to be adjusted; finally, according to the path planning constraints determined in advance or in real time, the first planned path is dynamically planned through the road section to be adjusted, so that the target vehicle continues to travel to the destination according to the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time. It can be seen that the system generates two first and second planning paths with opposite driving directions and the same starting point, and then dynamically plans the first planning path according to the different traffic data of the same section in the first and second planning paths, so that the target vehicle can drive from the starting point to the end point with the lowest travel cost or the shortest travel time. This is equivalent to the system identifying the sections that may be unfavorable for vehicle driving (such as construction area sections, accident-prone sections, traffic accident sections, traffic congestion sections, etc.) with different traffic data in the same section, and then plans the vehicle's driving path by using big data, so as to reasonably regulate the urban traffic flow. Therefore, the system can not only make full use of existing road resources and navigation resources, but also reduce the burden of road traffic, reduce traffic jams and traffic accidents. At the same time, the system can help users reduce fuel consumption and save the transportation costs of vehicle users by providing them with the lowest travel cost or the shortest travel time. In addition, the system helps reduce vehicle emissions and environmental pollution by optimizing the driving path and reducing congestion, and the path planning helps reduce vehicle emissions and environmental pollution.

[0104] It should be noted that the above embodiments collect, store, use, process, transmit, provide, disclose, delete, etc. related data (such as road network data, vehicle data, etc.) with or with the consent of the user. For example, the road network data and vehicle data are authorized with the knowledge and consent of the user; or the user actively provides them after reading the relevant instructions, or the user actively authorizes / provides / uploads them when using some or all of the functions described in the above embodiments, or other methods / channels are obtained with or with the consent of the user.

[0105] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

[0106] The structures, proportions, sizes, etc. illustrated in the drawings in this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention, so they have no substantial technical significance. Any modification of the structure, change of the proportion relationship or adjustment of the size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical content disclosed by the present invention. At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" quoted in this specification are only for the convenience of description, and are not used to limit the scope of the implementation of the present invention. The change or adjustment of their relative relationship should also be regarded as the scope of the implementation of the present invention without substantially changing the technical content.

[0107] It should be understood that, although the terms first, second, etc. may be used to describe the planned paths, etc. in the embodiments of the present invention, these terms are only used to distinguish the planned paths from each other. For example, without departing from the scope of the embodiments of the present invention, the first planned path may also be referred to as the second planned path, and similarly, the second planned path may also be referred to as the first planned path.

Claims

1. A vehicle path planning method based on big data, characterized in that: The method comprises the following steps: Based on a road network data model established in advance or in real time, a first planned path for the target vehicle to travel from a starting point to a destination, and a second planned path for the target vehicle to travel from the destination to a starting point are generated; During the driving process of the target vehicle, the traffic data of the first planned path and the second planned path are collected respectively, and the traffic data includes: traffic flow data, road obstacle data, traffic accident data and weather condition data; Comparing the traffic data of the first planned path with the traffic data of the second planned path, identifying traffic data with differences in the same road section, and recording the corresponding road section as a road section to be adjusted; According to the path planning constraints determined in advance or in real time, the first planned path is dynamically planned through the road section to be adjusted so that the target vehicle continues to drive to the destination along the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time.

2. The vehicle path planning method based on big data according to claim 1 is characterized in that: The process of dynamically planning the first planned path through the road section to be adjusted includes: Acquire all the spare sections that reach the end point of the road section to be adjusted based on the start point of the road section to be adjusted in the direction from the driving start point to the driving end point; Prioritize and combine the road sections according to the road section distance, congestion coefficient, road safety, road section charges, road section obstacle status, and road section road grade, and generate the driving section corresponding to each combination through the spare road sections; Displaying the driving section corresponding to each combination through the on-board display screen of the target vehicle, and receiving the sorting combination selected by the user based on the display result, which is recorded as a standby sorting combination; The travel section corresponding to the backup sorting combination is used to replace the section to be adjusted, and all remaining untraveled sections after dynamic planning are associated to obtain a dynamically planned path.

3. The vehicle path planning method based on big data according to claim 2 is characterized in that: If the user does not select a sorting combination on the vehicle-mounted display screen, the first planned path is used as a dynamic planned path; Alternatively, when there is no spare road section corresponding to the road section to be adjusted in the direction from the driving starting point to the driving end point, the first planned path is used as the dynamic planned path.

4. The vehicle path planning method based on big data according to claim 1 is characterized in that: The road network model includes: Lu=(X,Y,Z); Where Lu represents the road network model; X represents the intersection of a road section or the starting and ending points of a road, where X = {1, 2, 3, ...}; Y represents the road segment connecting two nodes, where Y = {(i, j)|i, j∈A, i≠j}; Z represents the weight set of the road section, and the weight in the road network model is related to the time period of the road section, where Z = {z ij (n)|(i, j)∈B, n∈N}, N is a natural number; i represents the predecessor node of the road segment, j represents the successor node of the road segment, and (i, j) represents the road segment; z ij (n) represents the weight of the road segment (i, j) in time period n.

5. The vehicle path planning method based on big data according to claim 1 or 4, characterized in that: The process of generating a first planned path for a target vehicle from a driving start point to a driving end point based on a road network data model established in advance or in real time, and generating a second planned path for the target vehicle from the driving end point to the driving start point includes: Obtaining a road network in the road network model, and setting a grade attribute for distinguishing low-grade roads from high-grade roads, and adding a road area number to all roads in the road network and assigning it to be empty; wherein the high-grade roads and the low-grade roads are determined based on road section distance, congestion coefficient, road safety, road section charges, road section obstacle status, and road section road grade; A road section node is randomly selected from all roads as a target node. When the level attribute of the target node is less than the level attribute h of the high-level road, the road area number of the target node is assigned to A, and all adjacent nodes connected to the target node and whose road area number values ​​are empty are queried and added to the preset dictionary g; and Take a road node x in the preset dictionary g. If the level attribute of the road node x is less than the level attribute h of the high-level road, and the road area number of the road node x is empty, then set the road area number of the road node x to A, and query all adjacent road nodes connected to the road node x and with empty road area number values, add them to the preset dictionary g, and remove the road node x from the preset dictionary g until the preset dictionary g is an empty set; Let A=A+1, and randomly select a road node with an empty road area number from all roads and continue to execute the loop until the road area numbers of all roads are not empty, thus obtaining a road partition network; The road partition network is used to generate a first planned path for the target vehicle to travel from a driving start point to a driving end point, and a second planned path for the target vehicle to travel from the driving end point to a driving start point is generated.

6. The vehicle path planning method based on big data according to claim 5 is characterized in that: The process of generating a first planned path for a target vehicle from a driving start point to a driving end point by using the road partition network, and generating a second planned path for the target vehicle from a driving end point to a driving start point includes: Determine a first road node and a second road node based on the driving direction of the target vehicle; wherein the first road node and the second road node of the first planned path are respectively the driving starting point and the driving end point, and the first road node and the second road node of the second planned path are respectively the driving end point and the driving starting point; Determining whether the driving starting point and the driving end point belong to the same road area number; If they belong to the same road area number, the minimum cost path from the first road node to the second road node is directly generated by searching the roads corresponding to the same road area number to obtain the first planned path or the second planned path; wherein the minimum cost path includes: the lowest driving path cost after combining the road section distance, congestion coefficient, road safety, road section toll, road section obstacle status and road section road grade; If they do not belong to the same road area number, then based on the level attribute h of the high-level road, search for the optimal entry node P1 and the optimal exit node P2 between the first road node and the second road node, and based on the road area number where the first road node is located, search for the minimum cost path of the optimal entry node P1 and the optimal exit node P2 to obtain the first planned path; and, based on the road area number where the second road node is located, search for the minimum cost path of the optimal entry node P1 and the optimal exit node P2 to obtain the second planned path.

7. The vehicle path planning method based on big data according to claim 1 is characterized in that: During the driving of the target vehicle, the method further includes: The target vehicle senses a preset distance range through a sensor pre-installed on the target vehicle to determine whether there is an obstacle within the preset distance range; If there is an obstacle, a prompt message is issued, and based on the current driving state of the target vehicle, the motion state of the obstacle, and the real-time distance of the obstacle from the target vehicle, the path of the target vehicle is changed to obtain an obstacle avoidance driving path, and the obstacle avoidance driving path is displayed on the on-board display screen of the target vehicle; If there is no obstacle, the target vehicle is controlled to maintain the current driving state.

8. The vehicle path planning method based on big data according to claim 1, characterized in that: After the target vehicle reaches the driving destination according to the first planned path, the method further includes: Based on the traffic data of the first planned path and the second planned path during the driving process of the target vehicle, the corresponding traffic data is converted into a matrix describing the change of the average traffic speed according to the time characteristics and the space characteristics, which is recorded as the target matrix; The batch size of the neural network is set based on the target matrix, the neural network is initialized, and the learning rate and loss function are set according to the network structure, number of neurons and weight vector of the neural network; wherein one neuron corresponds to one traffic influence factor; Calculate the similarity between each neuron and the weight vector, and take the neuron corresponding to the maximum similarity as the target neuron; Updating the weight vector of the target neuron, as well as updating the learning rate and the loss function, and after completing the update, determining whether the learning rate is less than a preset value, or determining whether the number of updates reaches a preset number of iterations; If the learning rate is less than a preset value, or the number of updates reaches a preset number of iterations, each neuron is marked, and the traffic influencing factor of the traffic data is found based on the marked neurons; If the learning rate is greater than or equal to a preset value and the number of updates does not reach a preset number of iterations, the similarity between each neuron and the weight vector is recalculated, and the learning rate or the number of updates is continued to be determined until the traffic influencing factor of the traffic data is found.

9. The vehicle path planning method based on big data according to claim 8 is characterized in that: The traffic influencing factors of the traffic data include at least one of the following: traffic flow data, road obstacle data, traffic accident data, and weather condition data; Alternatively, the traffic influencing factors of the traffic data include at least one of the following: section distance, congestion coefficient, road safety, section toll, section obstacle status, and section road grade.

10. A vehicle path planning system based on big data, characterized in that: The system comprises: An initial path module is used to generate a first planned path for a target vehicle from a driving start point to a driving end point according to a road network data model established in advance or in real time, and to generate a second planned path for the target vehicle from a driving end point to a driving start point; A traffic data collection module, used to collect the traffic data of the first planned path and the second planned path respectively during the driving process of the target vehicle, wherein the traffic data includes: traffic flow data, road obstacle data, traffic accident data and weather condition data; A traffic data comparison module, used to compare the traffic data of the first planned path with the traffic data of the second planned path, identify traffic data with differences in the same road section, and record the corresponding road section as a road section to be adjusted; The path dynamic planning module is used to dynamically plan the first planned path through the to-be-adjusted road section according to the path planning constraints determined in advance or in real time, so that the target vehicle continues to drive to the destination along the dynamically planned path; wherein the path planning constraints include: the lowest travel cost or the shortest travel time.

Images