Method for planning driving route of automatic driving vehicle by using high-precision map

Abstract

The invention relates to the technical field of route planning, in particular to an automatic driving vehicle driving route planning method utilizing a high-precision map, and the method comprises the steps: obtaining high-precision map data at each moment and a preset number of moments before, and the high-precision map data comprises parameter data of each node and road section on the high-precision map at each moment; obtaining the obstacle influence cost of each road section according to the total number of the obstacles of each road section and the position distribution characteristics between the obstacles; obtaining the special road condition cost of each road section according to the distribution characteristics of the traffic flow data of each road section before each moment; and obtaining the passing cost of the automatic driving vehicle when the automatic driving vehicle passes through each road section, obtaining a cost function of an Astar algorithm by combining the distance from each node of the automatic driving vehicle to the vehicle driving end point, and obtaining a final path planning result by adopting the Astar algorithm. The invention aims at enabling the driving planning route of the automatic driving vehicle to reach the optimal state.

Description

Technical Field

[0001] The present application relates to the field of route planning technology, and in particular to a method for planning a route for an autonomous driving vehicle using a high-precision map. Background Art

[0002] In recent years, intelligent transportation systems (ITS), a comprehensive application system that deeply integrates cutting-edge technologies such as communication, control, and artificial intelligence, have become a focus of widespread research. Autonomous driving, as a crucial component of ITS, has become a new research hotspot. Route planning, as part of the autonomous driving decision-making module, has a significant impact on the safety and efficiency of autonomous vehicles.

[0003] The document "Chen Tao, Li Chengxin, Zhao Chongyang. Research on route planning strategies for autonomous vehicles considering safety and efficiency [J]. Transportation Science and Engineering, 2024, 40(06): 127-134" mentioned that by determining the starting point and end point to generate the target route, and introducing different cost conditions, the path search algorithm can simulate different types of routes. However, the route formulated by a single cost condition is often difficult to achieve the optimal state. Therefore, many researchers have established a path planning method based on multiple costs by analyzing factors such as the shortest route, vehicle density on the road, and the probability of failure, and then calculating the route cost.

[0004] However, using this approach to planning and adjusting autonomous vehicle routes still has shortcomings. Currently, the sensor data captured by autonomous vehicles is limited, and thus the scope for route adjustment is limited. Therefore, route planning for autonomous vehicles is primarily based on high-precision maps. However, due to the lag in HD map data, road conditions may have changed significantly after route planning is completed, making it difficult to achieve an optimal route. Summary of the Invention

[0005] In view of the above, it is necessary to provide a route planning method for autonomous driving vehicles using high-precision maps to solve the above problems.

[0006] One embodiment of the present application provides a method for planning a route for an autonomous driving vehicle using a high-precision map, the method comprising: Obtaining high-precision map data at each moment and a preset number of moments before, including parameter data for each node and road segment on the high-precision map at each moment; wherein the parameter data includes the number of obstacles, obstacle locations, and traffic flow data for each road segment; Based on the total number of obstacles on each road section and the location distribution characteristics of the obstacles when the autonomous vehicle starts to pass through each road section, the obstacle impact cost of the autonomous vehicle when passing through each road section is obtained; based on the distribution characteristics of the traffic flow data of each road section before each time, the special road condition cost of each road section is obtained; Based on the obstacle impact cost and the special road condition cost of each road section, the travel cost of the autonomous driving vehicle when passing through each road section is obtained; based on the travel cost of the road section from the starting point to each node of the autonomous driving vehicle, combined with the distance from each node to the vehicle's travel end point, the cost function of the Astar algorithm is obtained, and the Astar algorithm is used to obtain the final path planning result.

[0007] Among them, an intersection with a selectable path greater than or equal to a preset value on the high-precision map at each moment is called a node.

[0008] The only path from one node to another is considered as a road segment.

[0009] The obstacle impact cost of the autonomous driving vehicle when passing through each road section is obtained as follows: For each road segment, calculate the degree of dispersion of the distance measure between all two combinations of obstacles in each road segment; The negative correlation mapping result of the discrete degree is positively fused with the total number of obstacles in each road section to obtain the obstacle cost of the autonomous driving vehicle in each road section.

[0010] The special road condition cost of each road section is obtained as follows: Calculate the mean and range of all traffic flow data of each road section before each moment, and use the negative correlation mapping result after positive fusion of the mean and the range as the special road condition cost of each road section at each moment.

[0011] The mathematical relationship of the special road condition cost is: ;in, is the special road condition cost of the i-th road section, is the mean traffic flow data of the i-th road section before the Z-th moment, is the range value of all traffic flow data of the i-th road section before the Z-th moment.

[0012] Among them, the passage cost of the autonomous driving vehicle when passing through each road section is determined by the forward fusion result of the obstacle impact cost and the special road condition cost of the corresponding road section.

[0013] The formula for obtaining the cost function of the Astar algorithm is: Where, represents the cost from the starting point to the jth node; represents the actual cost from the starting point to the jth node; Represents the estimated cost from the jth node to the destination.

[0014] The actual cost is specifically the sum of the travel costs of all road sections passed by the autonomous driving vehicle from the starting point to the j-th node.

[0015] The estimated cost is specifically the Euclidean distance from the j-th node to the destination.

[0016] This application has at least the following beneficial effects: This application analyzes the number and position changes of obstacles in a single road section during the process of route planning by an autonomous vehicle using a high-precision map to obtain the obstacle impact cost of each road section, which helps to determine whether the obstacles in each road section will affect the normal operation of the autonomous vehicle; further, considering that there is a certain lag in the update of road conditions on high-precision maps, the special road condition cost of each road section is obtained according to the distribution characteristics of the traffic flow data of each road section before each moment. Through in-depth analysis of historical traffic flow data, the traffic pattern of certain road sections in a specific time period can be identified; based on the obstacle impact cost and the special road condition cost of each road section, the traffic flow cost of the autonomous vehicle when passing through each road section is obtained. The beneficial effect of the Astar algorithm is to comprehensively consider the travel costs of each road section under various circumstances. Based on the travel costs of the road sections from the starting point to each node of the autonomous driving vehicle, combined with the distance from each node to the vehicle's destination, the cost function of the Astar algorithm is obtained, so that the route planned by the autonomous driving vehicle reaches the optimal state. This solves the problem that when the autonomous driving vehicle uses high-precision maps for path planning, due to the lag of the data on the high-precision maps, the road conditions may change significantly after the route planning is completed, making it difficult for the route to reach the optimal state. This can greatly improve the adaptability of the autonomous driving vehicle in the road environment, so that the path planning can always remain in the optimal state. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of the method for planning a route for an autonomous vehicle using a high-precision map provided in this application; Figure 2 A schematic diagram of nodes and road sections provided for this application; Figure 3 This is a schematic diagram of obtaining the cost function provided by this application. DETAILED DESCRIPTION

[0018] In the description of the embodiments of this application, words such as "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "or," and "for example" is intended to present the relevant concepts in a concrete manner.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the art of this application. The terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0020] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or precedence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the methods. Without departing from the scope of protection of this application, the order of execution of multiple steps can be interchanged with each other, and some steps can also be deleted.

[0021] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0022] This application proposes a method for planning the route of an autonomous vehicle using a high-precision map, which is applied to the field of route planning technology. Figure 1 , the method comprises the following steps: S1: Obtain high-precision map data at each moment and a preset number of moments before, including parameter data of each node and road section on the high-precision map at each moment; wherein, the parameter data includes data such as the number of obstacles, obstacle locations, and traffic flow for each road section.

[0023] At time Z, the autonomous driving vehicle is ready to go to the destination. The route planning system obtains the map data of each update of the high-precision map in the past T hours including the Zth moment, and obtains the obstacle quantity data, location data, and traffic flow data of each road section, and pre-processes the data.

[0024] In this application, T is a preset number, and its value range is 24~168. In this embodiment, the value of T is 168, and the implementer can choose according to the actual situation; the data preprocessing adopts the mean filling method, and the implementer can adopt other data preprocessing algorithms under the premise of ensuring data integrity.

[0025] It should be noted that the road segment division method in the high-precision map is as follows: an intersection with a selectable path greater than or equal to a preset value is called a node; the only directly connected path from one node to another is a road segment; in this embodiment, the preset value is 3, and the selectable path refers to the path corresponding to different travel directions that can be selected from a road intersection. The schematic diagram of the node and road segment is as follows Figure 2 As shown, Figure 2 In the figure, there are 3 and 4 selectable paths corresponding to point P and point Q respectively, so P and Q are both nodes, and the path from node P to node Q is called segment PQ.

[0026] S2: Based on the total number of obstacles in each road section and the location distribution characteristics of the obstacles when the autonomous vehicle starts to pass through each road section, the obstacle impact cost of the autonomous vehicle when passing through each road section is obtained; based on the distribution characteristics of the traffic flow data of each road section before each time, the special road condition cost of each road section is obtained.

[0027] Typically, vehicles encounter few obstacles on urban roads, and even if they do, they are fixed and immobile, so these obstacles generally have little impact on route planning. However, routes aren't always urban roads; they can also be block roads, which are often filled with obstacles like street vendors, parked non-motorized vehicles, and motor vehicles. The more obstacles there are, the greater the impact on vehicle efficiency. Consequently, traditional route planning algorithms often calculate higher costs for roads with more obstacles, and therefore avoid selecting these roads for route planning.

[0028] However, in reality, the more obstacles there are, the greater the impact on vehicle traffic efficiency is not necessarily the case. If the obstacles are located in a disorderly manner and have a low regularity, the impact on vehicle traffic efficiency will be greater. When the obstacles are located in a regular manner, such as when parked vehicles are in designated parking spaces and stalls and street debris are placed on both sides of the road, then in this case, even if there are more obstacles, the impact on the traffic efficiency of passing vehicles will actually be lower.

[0029] In summary, if the obstacle distribution on a road section is relatively chaotic and the distances between obstacles vary widely, the obstacles on that road section have a greater impact on vehicle traffic efficiency. Conversely, if the obstacle distribution is relatively regular and the distances between obstacles vary less, the impact on vehicle traffic efficiency is smaller, and the cost to autonomous vehicles of navigating that road section is lower. To reflect the impact of the aforementioned obstacle distribution on vehicle traffic efficiency, we now calculate the obstacle impact cost of each road section: for each road section, we calculate the degree of dispersion of the distance metric between all pairs of obstacles in each road section; we then perform a positive fusion of the negative correlation mapping result of this degree of dispersion with the total number of obstacles on each road section to obtain the obstacle cost of each road section.

[0030] In this embodiment, taking the i-th road section as an example, the mathematical relationship of the obstacle impact cost is: ;in, is the obstacle impact cost of the autonomous vehicle when passing through the i-th road segment, is the total number of obstacles in the i-th road section, is the variance of the Euclidean distance between obstacles in the i-th road segment, which is used to measure the degree of dispersion of the distance between obstacles; It represents a parameter preset to be greater than zero, used to prevent the denominator from being 0. In this embodiment, the value is 0.01.

[0031] Typically, when autonomous vehicles plan their routes, they consider sections with low traffic volume to be less expensive, so they choose these sections for route planning. However, in actual driving, sections with low traffic volume don't necessarily have higher traffic efficiency. This can be due to unexpected situations, such as accidents or temporary road construction, or poor road conditions, such as excessive gravel or uneven road surfaces, which can lead most vehicles to avoid such roads. However, due to the lag in HD map updates, there's no guarantee that these special road conditions will be immediately updated during route planning. Consequently, these sections may be considered less traffic-heavy, resulting in higher traffic efficiency and inclusion in the route plan. This can result in the final planned route still being suboptimal. Therefore, further analysis is needed to avoid this situation.

[0032] For a normal road, the volume of traffic follows a certain regularity, such as during rush hours and peak travel times during holidays. Therefore, the volume of traffic on a normal road only lasts for a short period before changing again, resulting in significant fluctuations in the overall traffic volume data. However, for roads with special road conditions, vehicles cannot pass through or pass through with difficulty, resulting in extremely low traffic volume on these roads for a long period of time with minimal fluctuations, resulting in relatively small changes in the overall traffic volume data. Based on this, the special road condition cost of each road section is determined based on the distribution characteristics of the traffic flow data before each time point: the mean and range of all traffic flow data for each road section before each time point are calculated, and the negative correlation mapping result after positive fusion of the mean and range values ​​is used as the special road condition cost of each road section at each time point.

[0033] In this embodiment, taking the i-th road section as an example, the mathematical relationship of its special road condition cost is: ;in, is the special road condition cost of the i-th road section, is the mean traffic flow data of the i-th road section before the Z-th moment, is the range value of all traffic flow data of the i-th road section before the Z-th moment.

[0034] S3: Based on the obstacle impact cost and the special road condition cost of each road section, the travel cost of the autonomous driving vehicle when passing through each road section is obtained; based on the travel cost of the road section from the starting point to each node of the autonomous driving vehicle, combined with the distance from each node to the end point of the vehicle, the cost function of the Astar algorithm is obtained, and the Astar algorithm is used to obtain the final path planning result.

[0035] The cost of an autonomous vehicle traveling on each road section can be determined by forward fusion of the obstacle impact cost and special road condition cost for the corresponding road section. In this embodiment, the forward fusion of multiple variables is calculated using a summation method.

[0036] It should be understood that when an autonomous vehicle is planning a route, if the number of obstacles on the road section is greater, the distribution is more chaotic, or the traffic volume on the road section is extremely low for a long time, it means that the vehicle's traffic efficiency on the road section is lower, the cost of traffic is greater, and the road section should not be planned into the driving route.

[0037] In the route planning system of an autonomous vehicle, the road sections in the HD map are first divided according to the starting and ending points determined by the user. The travel cost of each road section is then calculated and used to improve the cost function of the Astar algorithm. The improved cost function is: Where, represents the cost from the vehicle's starting point to the jth node; It represents the actual cost from the vehicle's starting point to the jth node. Its value is specifically the sum of the travel costs of all road sections that the autonomous vehicle passes through from the starting point to the jth node. represents the estimated cost from the jth node to the vehicle's destination, specifically the Euclidean distance from the jth node to the destination. Finally, the Astar algorithm is used for route planning, outputting the optimal route including road segment information. The specific steps of the Astar algorithm are well-known and will not be detailed in this application.

[0038] The schematic diagram of obtaining the cost function is as follows: Figure 3 shown.

[0039] This application uses SUMO software to build a simulated road network based on a high-precision map of a certain urban area, with a total of 50 test nodes and 62 test sections. The road network includes information on the number, location, and traffic flow of obstacles on each road. Under the same road network environment, simulation tests were conducted using a traditional route planning strategy (set as Route 1 in this solution, prioritizing traffic efficiency) and a route planning strategy that considers the influence of obstacle location patterns and the cost of special road conditions (set as Route 2 in this solution, also considering the influence of obstacle location patterns and special road conditions). The costs used in the route planning simulation results are shown in the following table: Table 1: Actual cost of the route In the above table, the travel time cost is specifically the ratio of the length of each road section to the maximum speed limit of its corresponding road section; the comprehensive travel cost is specifically the average of the obstacle impact cost, special road condition cost and travel time cost.

[0040] The route planning simulation results show that there is a significant difference in route costs between the departure and destination points. Compared to traditional routes, the route with the lowest comprehensive travel cost reduces time costs by 18.49%. The simulation results further demonstrate the effectiveness of the route planning strategy proposed in this application. It not only avoids the lag associated with route planning using high-precision maps, but also plans a low-cost, high-efficiency route.

[0041] The present application provides a method for planning a route for an autonomous vehicle using a high-precision map, the method comprising: analyzing the number and position changes of obstacles in a single road section during route planning by the autonomous vehicle using a high-precision map, obtaining an obstacle impact cost for each road section, which helps to determine whether obstacles in each road section will affect the normal operation of the autonomous vehicle; further, considering that there is a certain lag in the update of road conditions on the high-precision map, the special road condition cost of each road section is obtained based on the distribution characteristics of the traffic flow data of each road section before each moment, and through in-depth analysis of historical traffic flow data, the traffic pattern of certain road sections in a specific time period can be identified; based on the obstacle impact cost and the special road condition cost of each road section, the special road condition cost is obtained. The beneficial effect of the Astar algorithm is to calculate the travel cost of the autonomous driving vehicle when passing through each road section. The cost of each road section under various circumstances is integrated. Based on the travel cost of the autonomous driving vehicle from the starting point to each node, combined with the distance from each node to the vehicle's end point, the cost function of the Astar algorithm is obtained, so that the route planned by the autonomous driving vehicle reaches the optimal state. This solves the problem that when the autonomous driving vehicle uses high-precision maps for path planning, due to the lag of the data on the high-precision maps, the road conditions may change significantly after the route planning is completed, making it difficult for the route to reach the optimal state. This can greatly improve the adaptability of the autonomous driving vehicle in the road environment, so that the path planning can always remain in the optimal state.

[0042] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the part of the module, program segment or code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in an order different from that disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified function or action, or may be implemented by a combination of dedicated hardware and computer instructions.

[0043] The above embodiments 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, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for planning a driving route for an autonomous driving vehicle using a high-precision map, characterized in that: The method comprises the following steps: Obtaining high-precision map data at each time and a preset number of times before, including parameter data of each node and road section on the high-precision map at each time; wherein the parameter data includes the number of obstacles, obstacle locations and traffic flow data of each road section; According to the total number of obstacles in each road section and the position distribution characteristics of obstacles when the autonomous vehicle starts to pass through each road section, the obstacle impact cost of the autonomous vehicle when passing through each road section is obtained; according to the distribution characteristics of the traffic flow data of each road section before each time, the special road condition cost of each road section is obtained; Based on the obstacle impact cost and the special road condition cost of each road section, the travel cost of the autonomous driving vehicle when passing through each road section is obtained; based on the travel cost of the road section from the starting point of the autonomous driving vehicle to each node, combined with the distance from each node to the end point of the vehicle, the cost function of the Astar algorithm is obtained, and the Astar algorithm is used to obtain the final path planning result.

2. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 1, wherein: An intersection on the high-precision map at each moment where the selectable path at a single intersection is greater than or equal to a preset value is called a node.

3. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 1, wherein: The only path from one node to another is called a road segment.

4. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 1, wherein: The obstacle impact cost of the autonomous driving vehicle when passing through each road section is obtained as follows: For each road segment, the discreteness of the distance measure between all two combinations of obstacles in each road segment is calculated; The negative correlation mapping result of the discrete degree is positively fused with the total number of obstacles in each road section to obtain the obstacle cost of the autonomous driving vehicle in each road section.

5. The method for autonomous driving vehicle route planning using a high-precision map according to claim 1, wherein: The special road condition cost of each road section is obtained as follows: The mean and range of all traffic flow data of each road section before each moment are calculated, and the negative correlation mapping result after positive fusion of the mean and the range is used as the special road condition cost of each road section at each moment.

6. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 1, wherein: The mathematical relationship of the special road condition cost is: ;in, is the special road condition cost of the i-th road section, is the mean value of traffic flow data of the ith road section before the Zth moment, is the extreme value of all traffic flow data of the ith road section before the Zth moment.

7. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 1, wherein: The travel cost of the autonomous driving vehicle when passing through each road section is determined by the forward fusion result of the obstacle impact cost and the special road condition cost of the corresponding road section.

8. The method for autonomous driving vehicle route planning using a high-precision map according to claim 1, wherein: The formula for obtaining the cost function of the Astar algorithm is: ; In the formula, represents the cost from the starting point to the jth node; represents the actual cost from the starting point to the jth node; Represents the estimated cost from the jth node to the destination.

9. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 8, wherein: The actual cost is specifically the sum of the travel costs of all road sections that the autonomous driving vehicle passes through from the starting point to the jth node.

10. The method for planning a driving route for an autonomous driving vehicle using a high-precision map according to claim 8, wherein: The estimated cost is specifically the Euclidean distance from the jth node to the travel end point.

Images