Electric vehicle automatic driving path planning method and system based on high-precision positioning

Abstract

The application provides a kind of electric vehicle automatic driving path planning method and system based on high-precision positioning, which comprises: obtaining the first real-time position of electric vehicle, and obtaining the current electric quantity, battery health state information and set destination position of electric vehicle;Based on the first real-time position and high-precision map matching, determine the first lane information, determine the road information of each road between the first real-time position and the destination position based on the high-precision map;Determine the maximum driving range based on the current electric quantity and the battery health state information, determine the target charging pile based on the first real-time position, the maximum driving range and the charging pile position information and the current use state information of each charging pile in the high-precision map;Path planning based on the first real-time position, destination position, first lane information, road information and target charging pile, get the first automatic driving path.The application improves the accuracy and safety of path planning in electric vehicle automatic driving scenario.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer technology, and particularly relates to an electric vehicle automatic driving path planning method and system based on high-precision positioning. BACKGROUND

[0002] At present, the common driving path planning method is mainly based on electronic map and navigation algorithm, by obtaining the starting point and ending point position information, using road network data and traditional path search algorithm (such as Dijkstra algorithm, A* algorithm, etc.) to calculate a path from the starting point to the ending point.

[0003] However, the traditional path planning method cannot fully consider the special needs and real-time state of the electric vehicle. For example, the electric vehicle has limited endurance, and if the distribution of charging piles along the way and the real-time power of the vehicle are not considered during path planning, the vehicle may be stranded halfway due to power depletion; at the same time, the traditional path planning method lacks accurate perception of the real-time state of the road when facing complex road conditions (such as construction sections, traffic control areas), and cannot adjust the path in time, reducing the safety of driving, and it is difficult to meet the strict requirements of path planning accuracy and safety in the electric vehicle automatic driving scene. SUMMARY

[0004] The present application provides an electric vehicle automatic driving path planning method and system based on high-precision positioning, aiming to improve the accuracy and safety of path planning in the electric vehicle automatic driving scene.

[0005] In a first aspect, the present application provides an electric vehicle automatic driving path planning method based on high-precision positioning, comprising:

[0006] obtaining the first real-time position of the electric vehicle based on a high-precision positioning device, and obtaining the current power, battery health state information and set destination position of the electric vehicle;

[0007] matching the first real-time position with a pre-stored high-precision map to determine the first lane information where the electric vehicle is currently located, and determining the road information of each road between the first real-time position and the destination position based on the high-precision map;

[0008] determining the maximum driving range based on the current power and battery health state information, and determining the target charging pile based on the first real-time position, the maximum driving range, and the charging pile position information and current use state information of each charging pile in the high-precision map;

[0009] plan a path based on the first real-time position, the destination position, the first lane information, the road information, and the target charging pile, to obtain a first automatic driving path; distances between each charging pile in the target charging pile are less than or equal to a first preset distance threshold.

[0010] In a second aspect, the present application further provides an electric vehicle automatic driving path planning system based on high-precision positioning, which is applied to the electric vehicle automatic driving path planning method based on high-precision positioning as described in the first aspect; the electric vehicle automatic driving path planning system based on high-precision positioning comprises:

[0011] a data acquisition module configured to acquire a first real-time position of an electric vehicle based on a high-precision positioning device, and acquire a current power of the electric vehicle, battery health state information, and a set destination position;

[0012] a data matching module configured to match the first real-time position with a pre-stored high-precision map, determine first lane information in which the electric vehicle is currently located, and determine road information of each road between the first real-time position and the destination position based on the high-precision map;

[0013] a charging pile positioning module configured to determine a maximum driving range based on the current power and the battery health state information, and determine a target charging pile based on the first real-time position, the maximum driving range, and charging pile position information and current use state information of each charging pile in the high-precision map;

[0014] a path planning module configured to plan a path based on the first real-time position, the destination position, the first lane information, the road information, and the target charging pile, to obtain a first automatic driving path; the road information includes road speed limit, number of lanes, lane direction, and traffic signal light state; distances between each charging pile in the target charging pile are less than or equal to a first preset distance threshold.

[0015] In a third aspect, the present application further provides an electronic device, which comprises a memory configured to store a computer software program, and a processor configured to read and execute the computer software program, thereby realizing the electric vehicle automatic driving path planning method based on high-precision positioning as described in any one of the above aspects.

[0016] In a fourth aspect, the present application further provides a non-transitory computer readable storage medium, wherein the storage medium stores a computer software program, and the computer software program is executed by a processor to realize the electric vehicle automatic driving path planning method based on high-precision positioning as described in any one of the above aspects.

[0017] In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the high-precision positioning-based electric vehicle automatic driving path planning method of any one of the above aspects.

[0018] The high-precision positioning-based electric vehicle automatic driving path planning method provided by the embodiments of the present application determines the lane information in which the electric vehicle is currently located and the road information of each road in the driving process by combining the high-precision map with the real-time position and the destination position, so that the real-time state of the road can be accurately perceived. On the other hand, within the maximum driving range determined according to the current power and the battery health state information, the high-precision map position information and the current use state information of each charging pile are combined to screen a suitable target charging pile, effectively solving the range anxiety problem of the electric vehicle. Finally, the automatic driving path planned according to the real-time position, the destination position, the lane information, the road information and the target charging pile takes into account both the charging pile and the real-time state of the road, effectively avoiding the vehicle from breaking down halfway due to power problems, and at the same time, the real-time road conditions can also be perceived in complex road conditions, improving the accuracy and safety of path planning in the electric vehicle automatic driving scenario. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is one of the flowcharts of the high-precision positioning-based electric vehicle automatic driving path planning method provided by the embodiments of the present application;

[0020] Figure 2 is another flowchart of the high-precision positioning-based electric vehicle automatic driving path planning method provided by the embodiments of the present application;

[0021] Figure 3 is a structural schematic diagram of the high-precision positioning-based electric vehicle automatic driving path planning system provided by the embodiments of the present application;

[0022] Figure 4 is an embodiment diagram of the electronic device provided by the embodiments of the present application;

[0023] Figure 5 is an embodiment diagram of the computer readable storage medium provided by the embodiments of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0025] In the description of the present application, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first", "second" can be explicitly or implicitly included one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0026] In the description of the present application, the term "for example" is used to indicate "as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily interpreted as more preferred or more advantageous than other embodiments. The following description is given in order to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that those skilled in the art can recognize that the present application can be implemented without using these specific details. In other examples, well-known structures and processes will not be described in detail to avoid unnecessary details making the description of the present application obscure. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope consistent with the principles and characteristics disclosed.

[0027] Optionally, referring to Figure 1 , Figure 1 is one of the flowcharts of the high-precision positioning-based automatic driving path planning method for electric vehicles provided by the present application. In the embodiment of the present application, the execution subject of the high-precision positioning-based automatic driving path planning method for electric vehicles is the vehicle control system. Therefore, the high-precision positioning-based automatic driving path planning method for electric vehicles includes:

[0028] Step 10, obtaining the first real-time position of the electric vehicle based on the high-precision positioning device, and obtaining the current power, battery health state information of the electric vehicle and the set destination position.

[0029] Optionally, the electric vehicle in the embodiment of the present application can be an electric motorcycle. In a specific scenario, the electric motorcycle automatically drives from the current position to a relatively distant destination. The electric motorcycle is integrated with a vehicle control system, a high-precision positioning device, a battery management system, etc. The high-precision positioning device is a fusion device of a global satellite navigation system (such as GPS, Beidou, etc.) and an inertial navigation system (INS). Therefore, after the electric vehicle starts, the vehicle control system starts the high-precision positioning device to perform real-time positioning on the electric motorcycle to obtain the first real-time position. At the same time, the vehicle control system communicates with the battery management system (BMS) to obtain the current power and battery health state information of the electric motorcycle. In addition, the user inputs the set destination position on the operation interface of the vehicle control system, and the destination position is transmitted to the vehicle control system.

[0030] In an embodiment, in a certain city, an electric motorcycle is near the intersection of XX Avenue and XX Road. The vehicle control system activates the high-precision positioning device installed on the electric motorcycle, accurately calculates the first real-time position of the electric motorcycle as (longitude: 118.789012, latitude: 32.001234) by receiving Beidou satellite signals and its own inertial navigation data. Then, the vehicle control system communicates with the battery management system to obtain the current power of 60%, and the battery health status information shows that the battery capacity attenuation rate is 8%, which is in a normal state. Subsequently, the user inputs the destination as XX Science and Technology Park which is about 80 kilometers away from the current position on the touch screen of the vehicle control system.

[0031] Step 20, based on the first real-time position and the pre-stored high-precision map, determine the first lane information of the electric vehicle currently located, and based on the high-precision map, determine the road information of each road between the first real-time position and the destination position.

[0032] Further, the vehicle control system matches the obtained first real-time position with the high-precision map pre-stored in the system, determines the first lane information of the electric motorcycle currently located by comparing the position coordinates and geographic information on the map. At the same time, the vehicle control system calculates the road information of each road between the first real-time position and the destination position based on the high-precision map, such as querying the high-precision map database to obtain the speed limit of each road (such as 40 km / h, 60 km / h, etc.), the number of lanes (such as 2 lanes, 4 lanes, etc.), the direction of the lane (one-way, two-way, etc.), and the state of the traffic signal (red light, green light, yellow light, or signal failure, etc.).

[0033] Continuing the above embodiment, the vehicle control system matches the first real-time position of the electric motorcycle (longitude: 118.789012, latitude: 32.001234) with the high-precision map, and determines that the electric motorcycle is currently located in the rightmost lane of XX Avenue from south to north. Then, all possible roads from the current position to XX Science and Technology Park are found in the high-precision map. Among them, XX Avenue has a speed limit of 60 km / h and is a four-lane two-way road; a section of the elevated bridge connecting XX Avenue and the road leading to XX Science and Technology Park has a speed limit of 80 km / h and is a six-lane two-way road; some branch roads have a speed limit of 40 km / h and are two-lane two-way roads. At the same time, it is obtained that at a crossroads on the route, the current traffic signal state is green, and the remaining traffic time is 30 seconds.

[0034] Step 30, based on the current power and the battery health status information, determine the maximum driving range, and based on the first real-time position, the maximum driving range, and the charging pile position information and current use state information of each charging pile in the high-precision map, determine the target charging pile.

[0035] Further, the vehicle control system calculates the maximum driving range according to the obtained current power and battery health state information, in combination with an energy consumption model of the electric motorcycle, wherein the energy consumption model is established based on factors such as motor efficiency, vehicle weight, driving speed, etc. of the vehicle. Further, the vehicle control system searches for the location information and current use state information (idle, in use, etc.) of each charging pile in the high-precision map, takes the first real-time position as the starting point, and based on the maximum driving range, filters out the charging piles within the driving range, and determines a group of target charging piles according to the constraint condition that the distance between each charging pile in the target charging pile is less than or equal to a first preset distance threshold (such as 5 kilometers), specifically as the process of steps 301 to 304.

[0036] In an embodiment, according to the current power of 60% and the battery health state information, in combination with the energy consumption model of the electric motorcycle, the vehicle control system calculates that the maximum driving range is 100 kilometers under the current road condition and driving mode. In the high-precision map, 15 charging piles within a range of 100 kilometers from the current position are searched, of which 5 charging piles are in an idle state and 10 are in use, and there is a group of 3 idle charging piles, the distance between each of which is less than 5 kilometers, and they are respectively located in the commercial plaza parking lot at a distance of 30 kilometers, 33 kilometers and 35 kilometers from the current position, so the three charging piles are determined as the target charging piles.

[0037] Step 40, path planning based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile, to obtain the first automatic driving path.

[0038] Further, the vehicle control system performs path planning based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile, to obtain the first automatic driving path, specifically as the process of steps 401 to 404.

[0039] The embodiment of the application determines the lane information of the electric vehicle currently located and the road information of each road in the driving process by combining the real-time position and the destination position with the high-precision map, so that the real-time state of the road can be accurately perceived. On the other hand, the maximum driving range determined according to the current power and the battery health state information is combined with the charging pile location information and the current use state information of each charging pile in the high-precision map to filter out suitable target charging piles, effectively solving the range anxiety problem of the electric vehicle. Finally, the automatic driving path planned according to the real-time position, the destination position, the lane information, the road information and the target charging pile takes into account both the charging pile and the real-time state of the road, effectively avoiding the vehicle from breaking down due to power problems, and at the same time, the real-time road condition can also be perceived under complex road conditions, improving the accuracy and safety of path planning in the automatic driving scenario of the electric vehicle.

[0040] In an embodiment, the process of step 301 to step 304 comprises:

[0041] Step 301, according to the first real-time position and the charging pile position information of each charging pile, all charging piles that can be reached by the electric vehicle within the maximum driving range are screened out, and the reachable charging piles are obtained.

[0042] Optionally, the vehicle control system calculates the distance between the first real-time position and each charging pile position based on the geographic spatial distance calculation method, and screens out the reachable charging piles whose distance is less than or equal to the maximum driving range by comparing the calculated distance and the maximum driving range of the electric motorcycle.

[0043] Continuing the above embodiment, in the example that the electric motorcycle is at the intersection of XX Avenue and XX Road (longitude: 118.789012, latitude: 32.001234) and the maximum driving range is 100 kilometers, the vehicle control system traverses all charging pile position information in the high-precision map. Optionally, the Haversine formula is used to calculate the spherical distance d between the first real-time position and each charging pile position:

[0044]

[0045] wherein r is the average radius of the earth (6371 kilometers), is the latitude and longitude of the first real-time position of the electric motorcycle, is the latitude and longitude of the charging pile position. After calculation, there are 15 charging piles within 100 kilometers from the current position, and these 15 charging piles are determined as the reachable charging piles.

[0046] Step 302, for each charging pile in the reachable charging piles, a usage state history sequence within a preset time window is constructed based on its current usage state information, and a vacancy probability sequence of each charging pile is determined according to the frequency of the idle state in the usage state history sequence.

[0047] Further, for each charging pile in the reachable charging piles, the vehicle control system obtains its usage state information within a preset time window (such as the past 24 hours), and constructs a usage state history sequence, wherein the usage state history sequence is composed of a series of time stamps and usage states (0 represents idle, 1 represents occupied) at the corresponding time. Further, the vehicle control system calculates the frequency of the idle state in each window by sliding a fixed length window on the usage state history sequence through a sliding window algorithm, and obtains the vacancy probability sequence of each charging pile.

[0048] Taking one of the reachable charging piles A as an example, the vehicle control system obtains a past 24-hour use state history sequence of the charging pile A: [(t1, 0), (t2, 1), (t3, 0),..., (t n , 1)]. A sliding window length of 1 hour (including 60 time point data) is set, and the sliding window starts from the beginning of the sequence. In the first window, the idle state occurs 30 times, and the total data points are 60 times, so the idle probability of the window is P1 = 0.5. The sliding window is continuously calculated to obtain the idle probability P2 = 0.4 of the second window, and so on, to obtain the idle probability sequence [P1, P2,..., P m ] of the charging pile A, wherein m is the total number of sliding windows.

[0049] In step 303, the charging time of each charging pile is estimated based on the current use state of each charging pile and the driving trajectory of the vehicle within the preset range of each charging pile, to obtain the waiting time of the electric vehicle when reaching each charging pile.

[0050] Further, the vehicle control system estimates the charging time of each charging pile according to the current use state of each charging pile and the driving trajectory data of other vehicles within the preset range of each charging pile, to obtain the waiting time of the electric vehicle when reaching each charging pile, specifically as the process of steps 3031 to 3033.

[0051] In step 304, the target charging pile is determined based on the idle probability sequence and the waiting time of each charging pile in the reachable charging pile.

[0052] Further, the vehicle control system determines the target charging pile according to the idle probability sequence and the waiting time of each charging pile in the reachable charging pile, specifically as the process of steps 3041 to 3044.

[0053] The embodiment of the application can accurately determine the optimal target charging pile based on the real-time position and maximum driving range of the electric motorcycle, comprehensively consider the accessibility, idle probability and waiting time of the charging pile, fully consider the dynamic changes of the use state of the charging pile and the charging waiting condition of the vehicle, improve the accuracy and efficiency of finding a suitable charging pile for the electric motorcycle in the long-distance automatic driving process, reduce the risk of running out of power due to the failure to find a charging pile, and ensure the smooth completion of the long-distance journey of the electric motorcycle.

[0054] In an embodiment, the process of steps 3031 to 3033 includes:

[0055] In step 3031, if the current use state is an idle state, the waiting time is determined to be 0.

[0056] Optionally, if the current use state is idle, it means that the charging pile is not charging a vehicle, at this time the vehicle control system directly determines the waiting time of the charging pile corresponding to 0 minutes, which means that the electric motorcycle can start charging immediately after reaching the charging pile. Continue with the above embodiment, in the scenario of the electric motorcycle at the intersection of XX Road and XX Road, looking for a target charging pile, the vehicle control system obtains that the current use state of one of the reachable charging piles C is idle. Based on this, the vehicle control system does not need to perform additional calculations, and directly determines that the waiting time of the electric motorcycle when reaching the charging pile C is 0 minutes.

[0057] Step 3032, if the current use state is occupied, determine the total number of vehicles of each charging pile and the remaining charging time of the vehicle being charged.

[0058] Further, when the current use state of the charging pile is occupied, the vehicle control system obtains the number of vehicles being charged by the charging pile, and analyzes the number of vehicles that are expected to arrive at the charging pile before the electric motorcycle arrives at the charging pile through high-precision map and vehicle trajectory data. Add the two parts of the vehicle number to get the total number of vehicles. In addition, the vehicle control system also obtains the remaining charging time of the vehicle being charged through the communication between the charging pile and the vehicle being charged.

[0059] Continue with the above embodiment, for the reachable charging pile D, the vehicle control system judges that its current use state is occupied. After inquiry, there is currently 1 car being charged in the charging pile D, and through the analysis of the trajectory data, it is found that there are still 2 cars that will arrive at the charging pile before the electric motorcycle is expected to arrive at the charging pile D. Therefore, the total number of vehicles of the charging pile is 1+2=3. The vehicle control system obtains that the vehicle being charged still has 25 minutes of charging time through the communication between the charging pile and the vehicle being charged.

[0060] Step 3033, based on the total number of vehicles of each charging pile, the remaining charging time of the vehicle being charged, and the average charging operation time of each vehicle, the charging time is estimated, and the waiting time when reaching each charging pile is determined.

[0061] Further, the vehicle control system uses a queuing model to estimate the charging time according to the total number of vehicles of each charging pile, the remaining charging time of the vehicle being charged, and the average charging operation time of each vehicle (which can be obtained by statistical analysis of historical charging data). The G / G / 1 queuing model based on random process (general arrival time interval, general service time, single service station) is used to consider the randomness of vehicle arrival time and the uncertainty of charging time, and to calculate the waiting time of the electric motorcycle when reaching each charging pile. The specific calculation formula is as follows: where, W qis the average waiting time, ρ = λ / μ is the system utilization, λ is the average vehicle arrival rate (obtained by presetting the range of vehicle historical arrival data statistics), μ is the average service rate (μ = 1 / T avg , T avg is the average charging operation time of each vehicle), is the variance of the vehicle arrival time interval, is the variance of the charging service time.

[0062] Continuing in the example of charging pile D, the vehicle control system obtains, through historical data statistics, that the average vehicle arrival rate λ = 0.2 vehicles / minute in this area, the average charging operation time T avg of each vehicle = 30 minutes, so the average service rate μ = 1 / 30 vehicles / minute, and the system utilization ρ = 6. At the same time, through historical data calculation, the variance of the vehicle arrival time interval is 4, and the variance of the charging service time is 9. Substituting the parameters into the G / G / 1 queuing model formula, the average waiting time is obtained as: W q = (6 2 + 4 + 9) / (4 * 6 * 1 / 30) = 61.25 minutes.

[0063] The embodiment of the application can accurately estimate the waiting time of the electric motorcycle when it arrives at each charging pile for charging piles in different use states, fully considers the uncertainty of the vehicle arrival time and the fluctuation of the charging service time, provides accurate waiting time parameters for selecting target charging piles in subsequent path planning, and enables the electric motorcycle to reasonably plan the path to the charging pile based on more accurate charging waiting time information in the long-distance automatic driving process, reduces unnecessary waiting, improves charging efficiency and the reliability of the trip planning, and ensures the smooth completion of long-distance driving of the electric motorcycle.

[0064] In an embodiment, the process of steps 3041 to 30443 includes:

[0065] Step 3041, for each charging pile in the reachable charging pile, the distance loss value of each charging pile is determined according to the distance between the first real-time position and the charging pile position information of each charging pile, and the comprehensive evaluation index of each charging pile is determined based on the distance loss value of each charging pile, the idle probability sequence and the waiting time.

[0066] Optionally, for each charging pile in the reachable charging pile, the vehicle control system calculates the distance d i between the first real-time position and the charging pile position information by using a geographic spatial distance algorithm, and constructs a distance loss value L d (d i), wherein the distance loss value is used to evaluate the value of the charging pile according to the distance, the farther the distance, the greater the loss, and the specific formula is L d (d i )=1 / exp(d i -10)。

[0067] Further, the vehicle control system performs mean calculation on the idle probability sequence P ij of the charging pile (wherein i represents the charging pile number, and j represents different time points in the time sequence) to obtain the average idle probability Further, the vehicle control system determines the comprehensive evaluation index E i of each charging pile according to the distance loss value, the average idle probability, and the waiting time T i , and the specific formula is:

[0068] Continuing the above embodiment, the electric motorcycle is at the intersection of XX Avenue and XX Road (longitude: 118.789012, latitude: 32.001234), and 15 reachable charging piles are obtained in step 301. Taking the charging pile numbered 1 as an example, the vehicle control system calculates the distance d1 between it and the current position of the electric motorcycle as 20 kilometers by using the Haversine formula, and therefore the distance loss value L d (d1)≈0.000045. The idle probability sequence of this charging pile is [0.6, 0.7, 0.8], and therefore the average idle probability The waiting time T1 is 15 minutes, and therefore the comprehensive evaluation index of the charging pile numbered 1 is E1≈1.97*10 -6 . The above calculation is sequentially performed on the 15 reachable charging piles to obtain their respective comprehensive evaluation indexes.

[0069] In step 3042, the charging pile with a comprehensive evaluation index greater than or equal to a preset index threshold value is determined as a candidate charging pile, and the device health index of each charging pile is determined based on the number of faults and the cumulative downtime of each charging pile in the candidate charging pile.

[0070] Further, the vehicle control system compares the comprehensive evaluation index E i with a preset index threshold value E th , and determines the charging pile with E i ≥E th as a candidate charging pile. For each candidate charging pile, the vehicle control system obtains its historical number of faults F i and cumulative downtime H i , calculates the device health index S i by establishing a device health evaluation model, so as to measure the reliability of the charging pile device, and the specific formula is:

[0071] S i = 1 / [1 + aF i + bH i ], a and b are preset coefficients.

[0072] In an embodiment, the preset index threshold E th = 1 * 10 -6 , a = 0.1, b = 0.05, and among the 15 reachable charging piles, the comprehensive evaluation index of 5 charging piles is greater than or equal to the threshold value, and the 5 charging piles are determined as candidate charging piles. Taking one of the candidate charging piles (No. 3) as an example, the historical fault times F2 = 2 times, and the cumulative downtime H3 = 10 hours, therefore, the equipment health index S i of the candidate charging pile No. 3 is approximately 0.59, and the equipment health index of the 5 candidate charging piles is calculated respectively.

[0073] In step 3043, the candidate charging pile with the equipment health index greater than or equal to the preset health threshold and the distance from the first real-time position less than or equal to the preset charging pile is determined as the target charging pile.

[0074] Further, the vehicle control system sets a preset health threshold S th and a preset distance threshold D th , and the charging pile with the equipment health index S i ≥ S th and the distance d i from the first real-time position d th ≥ D is determined as the target charging pile. In an embodiment, the preset health threshold S th = 0.5, and the preset distance threshold D th = 30 kilometers. Among the 5 candidate charging piles, the equipment health index of 3 charging piles is greater than or equal to 0.5, and the distance of the 3 charging piles from the current position of the electric motorcycle is less than 30 kilometers, therefore, the vehicle control system determines the 3 charging piles as the target charging pile.

[0075] The embodiment of the present application can consider multiple factors such as distance, idle probability, waiting time, and equipment health from the reachable charging piles, and screen out the most suitable target charging pile, therefore, the use value and reliability of the charging pile are comprehensively considered, the charging pile with too long distance, too long waiting time or unstable equipment is avoided, the electric motorcycle can accurately plan the path to the high-reliability charging pile in the long-distance automatic driving process, the risk of trip interruption is reduced, and the safe long-distance driving of the electric motorcycle is ensured.

[0076] In an embodiment, the process of steps 401 to 404 includes:

[0077] Step 401, a directed graph is constructed with the target charging pile and the road intersections of each road between the first real-time position and the destination position as nodes, the road segment distance between adjacent intersections and the position distance between the intersection and the target charging pile as edges, and the road speed limit, the number of lanes, the lane direction and the traffic signal state between adjacent intersections and each intersection and the target charging pile as edge attributes.

[0078] Optionally, the vehicle control system takes the target charging pile and the road intersections of each road between the first real-time position and the destination position as nodes, and gives each node a unique identifier. The road segment distance between adjacent intersections and the position distance between the intersection and the target charging pile are calculated as edges, and the road speed limit, the number of lanes, the lane direction and the traffic signal state between adjacent intersections and each intersection and the target charging pile are given as edge attributes to obtain a directed graph.

[0079] Continuing the above embodiment, in the scenario where the electric motorcycle is located at the intersection of XX Avenue and XX Road and the destination is XX Science and Technology Park, and 3 target charging piles have been determined, 8 road intersections are identified between the current position and the destination, plus 3 target charging piles, a total of 12 nodes. The distance between each node is calculated, for example, the distance from node 1 (current position) to node 2 (adjacent intersection) is 1 km, the distance from node 2 to node 3 is 1.5 km, etc. At the same time, the edge attributes are given, such as the road segment speed limit from node 1 to node 2 is 60 km / h, 4 lanes in both directions, the lane direction is bidirectional, and the current traffic signal state is green for 40 seconds; the road segment speed limit from node 2 to one of the target charging piles is 40 km / h, 2 lanes in both directions, the lane direction is unidirectional, and there is no traffic signal, a directed graph containing nodes and edge attributes is constructed.

[0080] Step 402, based on the directed graph, the valid road segments are selected according to the road speed limit of each road segment and the safe maximum driving speed of the electric vehicle.

[0081] Further, the vehicle control system obtains the road speed limit v limit of each road segment based on the constructed directed graph, and selects in combination with the safe maximum driving speed v max of the electric vehicle. The selection condition is set, if the speed limit v limit of the road segment is greater than the safe maximum driving speed v max of the electric vehicle, and there is no special provision (such as electric vehicle exclusive lane at a specific time period) that allows the electric vehicle to drive at its safe maximum driving speed, the road segment is regarded as invalid; otherwise, it is retained as a valid road segment.

[0082] In an embodiment, the safe maximum driving speed v max= 80 km / h, in the directed graph, a certain road segment is limited to 100 km / h and has no special provisions, and the vehicle control system determines that this road segment is an invalid road segment; another road segment is limited to 70 km / h, and it is retained as a valid road segment. By performing the above judgment on all road segments in the directed graph, the valid road segments that meet the conditions are screened out, and the road segments on which the electric vehicle cannot travel at a safe speed are excluded.

[0083] In step 403, based on the directed graph, the first lane information is combined with the lane direction and lane compatibility of each lane of each road segment in the valid road segment to screen out the passable lane from the first real-time position in the valid road segment. The lane compatibility represents that the lane type and lane authority of the lane are consistent with the vehicle type of the electric vehicle.

[0084] Further, the vehicle control system screens the passable lane from the first real-time position based on the first lane information, in combination with the lane direction and lane compatibility of each lane of each road segment in the valid road segment, wherein the lane compatibility is determined by judging whether the lane type (such as a motor lane, a non-motor lane, etc.) and the lane authority (such as whether the electric vehicle is allowed to pass) of the lane are consistent with the vehicle type of the electric vehicle. Only the lane whose direction is consistent with the driving direction of the electric vehicle and the lane compatibility is screened as the passable lane from the first real-time position.

[0085] Continuing the above embodiment, it is known that the electric motorcycle is currently in the rightmost lane (first lane information) of XX Avenue in the south-north direction, and in a certain valid road segment, there are three lanes, the left lane is a left-turn-only lane, the middle lane is a straight and left-turn mixed lane, and the right lane is a straight lane and allows the electric vehicle to pass. The vehicle control system screens out the right lane of the road segment as the passable lane based on the demand of the electric motorcycle to travel straight to the destination, in combination with the lane direction and the lane compatibility. Similar screening is performed on all valid road segments to determine the complete set of passable lanes.

[0086] In step 404, based on the directed graph, the position distance between each road segment in the valid road segment and the target charging pile is calculated, and the road segment with a distance less than or equal to a threshold value is screened as a target road segment. Based on the directed graph, the passable lane, and the target road segment, path planning is performed to obtain a first automatic driving path.

[0087] Further, the vehicle control system calculates the position distance between each road segment in the valid road segment and the target charging pile based on the directed graph, and screens the road segment with a distance less than or equal to a threshold value as a target road segment.

[0088] Further, the vehicle control system performs path planning based on the directed graph, the passable lane, and the target road segment to obtain a first automatic driving path, specifically as the processes of step 4041 to step 4044.

[0089] The embodiment of the application converts the actual road network into a structured directed graph, fully considers the road speed limit, lane direction, lane compatibility and target charging pile position for path planning, ensures that the planned automatic driving path is not only reasonable in distance, but also ensures that the electric motorcycle is always in a passable lane during driving, and can smoothly reach the target charging pile for charging, improves the safety of automatic driving, and ensures that the electric motorcycle completes long-distance travel.

[0090] In an embodiment, the processes of steps 4041 to 4044 include:

[0091] Step 4041, based on the directed graph, according to the connectivity of adjacent lanes in the passable lane and the connectivity of adjacent road segments in the target road segment, an initial path is generated with the first real-time position as the starting point and the destination position as the end point. Each path includes a plurality of continuous lanes and road segments.

[0092] Optionally, the vehicle control system uses a graph search algorithm (such as an improved version of depth-first search DFS or breadth-first search BFS) to search for a path based on the directed graph, according to the connectivity of adjacent lanes in the passable lane and the connectivity of adjacent road segments in the target road segment, starting from the first real-time position and ending at the destination position. In the search process, the connectivity rules of the lanes and road segments are strictly followed to ensure that each generated path is composed of a plurality of continuous lanes and road segments, and an initial path set is obtained.

[0093] Continuing in the scenario where the electric motorcycle is located at the intersection of XX Avenue and XX Road, and the destination is XX Science and Technology Park, and the passable lane and the target road segment have been determined. A breadth-first search algorithm is used to start searching from the current position node. In the search process, it is found that the specific lane in the passable lane where the electric motorcycle is located is connected to the next road segment, and the search range is gradually expanded along the connected lanes and road segments. After searching, three initial paths are generated: path 1 passes through A lane, a road segment, B lane, b road segment, etc.; path 2 passes through C lane, c road segment, D lane, d road segment, etc.; and path 3 passes through E lane, e road segment, F lane, f road segment, etc. These paths all meet the requirement of being composed of continuous lanes and road segments.

[0094] Step 4042, for each road segment of each path in the initial path, based on the directed graph, the road speed limit and the road length of each lane are determined to determine the ideal travel time of each road segment under the influence of no traffic signal, and based on the ideal travel time, the signal light cycle, the red light duration and the current red light remaining time of each lane are combined to determine the actual travel time of each road segment.

[0095] Further, for each road segment in the initial path, the vehicle control system obtains the road speed limit v of each lane corresponding to the road segment based on the directed graph, and determines the actual travel time t of each road segment based on the road speed limit v and the road length L of each lane.i and road length l i , the ideal passing time t i is calculated under the influence of no traffic light by the formula l i / v ideal-i . Then, combined with the signal light cycle T cycle-i , red light duration T red-i and current red light remaining time t remain-i of each lane, the actual passing time t actual-i is calculated. If it is currently green and the remaining time is enough to pass the road section, that is, T remain-i ≥t ideal-i , then t actual-i =t ideal-i ; if it is currently green but the remaining time is not enough, that is, T remain-i <t ideal-i , then t actual-i =t ideal-i -T remain-i +T cycle-i ; if it is currently red, then t actual-i =T red-i -T remain-i +t ideal-i .

[0096] Continue to take the a road section in the initial path 1 as an example, its corresponding lane speed limit v a =60km / h, road length l a =2km, the ideal passing time t ideal-a =2min is calculated by the formula. The signal light cycle T cycle-a =120s of the lane, red light duration T red-a =40s, current red light remaining time T remain-a =10s, since it is currently red, the actual passing time t actual-a =40-10+120=150s. The above calculation is sequentially performed on all road sections in the initial path to obtain the actual passing time of each road section.

[0097] Step 4043, based on the position distance between each road section and the target charging pile, the charging consumption time of each road section during charging is predicted, and the total road length of each road section is determined based on the road length of each lane.

[0098] Further, the vehicle control system predicts the charging consumption time of each road section during charging according to the position distance d i between each road section and the target charging pile, combined with the charging power P of the electric motorcycle, the current power Q current and the battery capacity Q total , through the formula wherein n is the number of all road segments, to predict the charging consumption time t of each road segment at the time of charging charge-i At the same time, the length of each lane corresponding to the road is directly obtained, the length of all lanes contained in each road segment is added, and the total length L of the road is determined total-i .

[0099] In an embodiment, for the b road segment in the path 1, the distance d b to the target charging pile is 3 kilometers, the charging power P of the electric motorcycle is 6 kW, the current power Q current is 60%, the battery capacity Q total is 80 Ah, and it is assumed that there are a total of 5 road segments, and the sum of the distances to the target charging pile is kilometers, so the calculated charging consumption time of the b road segment is seconds. The b road segment consists of two lanes with lengths of 1.2 kilometers and 0.8 kilometers, so the total length L total-b of the road is 1.2+0.8=2 kilometers. The above calculation is performed for all road segments to obtain the charging consumption time and the total length of the road.

[0100] Step 4044, determining a first candidate path based on the total length of the road, the charging consumption time and the actual passing time of each road segment, and determining a first automatic driving path based on the number of remaining lanes and the number of charging piles containing the target charging pile in each path of the first candidate path.

[0101] Further, the theoretical driving time of each road segment is calculated according to the total length of the road of each road segment combined with the average driving speed of the electric motorcycle on the road segment, and the theoretical driving time of each road segment is added to the charging consumption time and the actual passing time to obtain the total time consumption of each path.

[0102] Further, the vehicle control system determines the path with a total time consumption less than or equal to a preset time consumption threshold as the first candidate path, and determines the first automatic driving path according to the number of remaining lanes and the number of charging piles containing the target charging pile in each path of the first candidate path, specifically as steps 40441 to 40444.

[0103] The embodiment of the present application can consider lane connectivity, traffic lights, charging demand and other factors under complex road network conditions, filter the optimal first automatic driving path from multiple possible paths, accurately calculate the passing time and charging time under different conditions, and comprehensively weigh multiple key indicators of the path through multi-objective optimization, so as to ensure that the planned automatic driving path can not only make the electric motorcycle quickly and efficiently reach the destination, but also guarantee sufficient charging guarantee and good lane passing conditions during driving, and improve the practicality of the automatic driving path planning of the electric motorcycle.

[0104] In an embodiment, the process of step 40441 to step 40444 includes:

[0105] Step 40441, the path with the remaining lane number greater than the first preset number and the charging pile number greater than the second preset number in the first candidate path is determined as the second candidate path.

[0106] Optionally, the vehicle control system evaluates each path in the first candidate path, checks the remaining lane number N lanes and the charging pile number N charger of the target charging pile. Set the first preset number N lanes-th and the second preset number N charger-th , the path meeting N lanes >N lanes-th and N charger >N charger is determined as the second candidate path, to ensure that the remaining second candidate path has sufficient lane selection space and charging facility guarantee. Continue with the above embodiment, there are 3 paths in the first candidate path. Set the first preset number N lanes-th =1, the second preset number N charger-th =1. The remaining lane number N lanes-1 =2 of path 1, the charging pile number N charger-1 =2; the remaining lane number N lanes-2 =1 of path 2, the charging pile number N charger-2 =1; the remaining lane number N lanes-3 =3 of path 3, the charging pile number N charger-3 =2. By comparison, path 1 and path 3 meet the conditions and are determined as the second candidate path.

[0107] Step 40442, the second candidate path is screened to obtain the third candidate path, with the sum of the turning angle change rates of adjacent road segments in each path being the smallest.

[0108] Further, for each path in the second candidate path, the vehicle control system calculates the sum S θThe rate of change of steering angle is the difference in steering angle Δθ between adjacent road segments. i With road segment length l i The ratio of .

[0109] Furthermore, the vehicle control system aims to minimize... With the goal of selecting S θ The shortest path is selected as the third candidate path to ensure that the path with smooth steering and high driving comfort is chosen.

[0110] Continuing with the above embodiment, in the second candidate path, path 1 comprises 5 road segments, with turning angle differences between adjacent road segments of 15°, 10°, 5°, and 8°, corresponding to road segment lengths of 2km, 1.5km, 1km, and 2.5km, respectively. The sum of the turning angle change rates of path 1 is calculated to be S. θ-1 =7.5 + 6.67 + 5 + 3.2 = 22.37. Route 3 contains 4 road segments with turning angle differences of 8°, 6°, and 4° between adjacent segments. The corresponding segment lengths are 3km, 2km, and 1.5km. The sum of the turning angle change rates for Route 3 is calculated to be S. θ-3 =2.67 + 3 + 2.67 = 8.34. Since S θ-3 θ-1 Path 3 was identified as the third candidate path.

[0111] Step 40443: The third candidate path is filtered with the objective of minimizing the sum of the lane curvatures of adjacent lanes in each road segment of each path, to obtain the fourth candidate path.

[0112] Furthermore, for each of the third candidate paths, the vehicle control system calculates the sum of the lane curvatures S of the adjacent lanes in each road segment. curvature Wherein, the lane curvature is the reciprocal of the radius of curvature of the lane centerline, i.e., 1 / R. i , where R i Let be the radius of curvature of the i-th lane. Furthermore, the vehicle control system aims to minimize... With the goal of selecting S curvature The shortest path is selected as the fourth candidate path to ensure that the selected path has a small lane curvature, making driving safer.

[0113] Continuing with the above embodiment, path 3 comprises four road segments, with the radii of curvature of adjacent lanes in each segment being R1 = 500m, R2 = 1000m, R3 = 800m, and R4 = 1200m, respectively. Calculate the sum of the lane curvatures S. curvature-3 =0.002 + 0.001 + 0.00125 + 0.00083 ≈ 0.00508. After comparing with other possible third candidate paths, the S of path 3 is... curvature ​The minimum, so it is determined as the fourth candidate path.

[0114] Step 40444, screening the fourth candidate path with the sum of the number of adjacent lane changes of each road segment in each path as the target, to obtain the first autonomous driving path.

[0115] Further, the vehicle control system calculates the sum of the number of adjacent lane changes N of each road segment in each path in the fourth candidate path. lane-changes The target is to minimize N lane-changes , and the path with the least number of changes is screened as the first autonomous driving path, in order to reduce unnecessary lane changes and improve driving efficiency and safety.

[0116] Continue with the above example, path 3 has a sum of adjacent lane changes N lane-changes-3 = 2 times throughout the entire trip. After comparing with other possible fourth candidate paths, path 3 has the minimum N lane-changes , so the vehicle control system determines path 3 as the first autonomous driving path.

[0117] Embodiments of the present application screen candidate paths from multiple dimensions, ultimately determining the first autonomous driving path with the best overall performance. Not only does it ensure that the path has enough lane selection space and charging facility guarantee, but it also minimizes the steering angle change rate, lane curvature and lane change number, improving driving comfort, safety and efficiency, and paying more attention to the detailed features of the path and the actual driving experience. It provides a practical autonomous driving path planning for electric motorcycles, ensuring the smooth progress of long-distance travel of electric motorcycles.

[0118] Optionally, referring to Figure 2 , Figure 2 is a flowchart of the electric vehicle autonomous driving path planning method based on high-precision positioning provided by the present application, and after step 40, it further includes:

[0119] Step 50, during the driving of the electric vehicle, based on the second real-time position of the electric vehicle obtained by the high-precision positioning device in real time, determining the second lane information of the electric vehicle in the high-precision map.

[0120] Optionally, during the driving of the electric motorcycle according to the first autonomous driving path, the vehicle control system continuously starts the high-precision positioning device, which combines satellite navigation system and inertial navigation technology to obtain the second real-time position of the electric motorcycle in real time. Subsequently, the vehicle control system matches the second real-time position with the high-precision map, and determines the second lane information of the electric motorcycle in the high-precision map by comparing the position coordinates and the lane data in the map, including the lane number and the lane direction.

[0121] Continuing the above embodiment, the electric motorcycle travels along the first autonomous driving path planned on XX Avenue, and the initial first lane information is that the rightmost lane of XX Avenue from south to north. The vehicle control system obtains the second real-time position of the electric motorcycle as (longitude: 118.791234, latitude: 32.003456) through the high-precision positioning device in real time. After matching the position with the high-precision map, it is determined that the electric motorcycle is in the middle lane of XX Avenue from south to north at this time, that is, the second lane information is obtained.

[0122] Step 60, based on the second lane information and the first autonomous driving path, the lateral position deviation, the heading angle deviation and the deviation change rate of the electric vehicle in the driving process are determined.

[0123] Further, the vehicle control system calculates the lateral position deviation D lateral , the heading angle deviation D heading and the deviation change rate R deviation of the electric motorcycle in the driving process based on the obtained second lane information and the pre-planned first autonomous driving path. The lateral position deviation is obtained by calculating the distance between the current position of the electric motorcycle and the center line of the first autonomous driving path in the vertical direction; the heading angle deviation is the difference between the current heading angle of the electric motorcycle and the ideal heading angle of the first autonomous driving path at the corresponding position; and the deviation change rate is obtained by differentiating the lateral position deviation and the heading angle deviation within a certain time interval.

[0124] The calculation formula of the lateral position deviation is as follows:

[0125] Where (x0, y0) is the second real-time position coordinate, (x path , y path ) is the coordinate of the corresponding position of the first autonomous driving path.

[0126] The calculation formula of the heading angle deviation is as follows:

[0127] D heading = θ - θ path , where θ is the current heading angle of the electric motorcycle, and θ path is the ideal heading angle of the first autonomous driving path at the corresponding position.

[0128] The calculation formula of the deviation change rate is as follows:

[0129] R deviation = ΔD lateral / Δt (lateral position deviation change rate, Δt is the time interval), R deviation-heading = ΔD heading / Δt (heading angle deviation change rate).

[0130] Step 70, based on the lateral position deviation and the heading angle deviation and their corresponding deviation threshold values, determine a lane departure risk level. The lane departure risk level includes a first risk level and a second risk level, the second risk level being higher than the first risk level.

[0131] Further, the vehicle control system sets a lateral position deviation threshold value D lateral-th and a heading angle deviation threshold value D heading-th . The calculated lateral position deviation D lateral and the heading angle deviation D heading are compared with the corresponding threshold values. If D lateral < D lateral-th and D heading < D heading-th , the lane departure risk level is the first risk level; if D lateral ≥ D lateral-th or / and D heading ≥ D heading-th , the lane departure risk level is the second risk level, the second risk level being higher than the first risk level.

[0132] In an embodiment, the lateral position deviation threshold value D lateral-th = 0.002 km, and the heading angle deviation threshold value D heading-th = 3°. In the above calculation, D lateral = 0.0013 km < D lateral-th , and D heading = 1° < D heading-th , so the lane departure risk level is the first risk level. If in the subsequent calculation, the lateral position deviation becomes 0.003 km, since 0.003 > 0.002, the lane departure risk level becomes the second risk level.

[0133] Step 80, if the lane departure risk level is the first risk level, the second real-time position is determined as the path correction starting point. If the lane departure risk level is the second risk level, a prediction interval time is determined based on the deviation change rate, and the path correction starting point is determined based on the second real-time position, the prediction interval time, and the current driving speed and the current heading angle of the electric vehicle.

[0134] Further, if the lane departure risk level is the first risk level, the vehicle control system directly determines the second real-time position as the path correction starting point, and prepares to perform a small-amplitude path adjustment. If the lane departure risk level is the second risk level, the vehicle control system determines a prediction interval time T predect based on the deviation change rate R deviation , and the specific formula is T predect = k / |R deviation|, (wherein k is a constant, such as k=0.1). Further, the vehicle control system determines the path correction starting point based on the second real-time position, the predicted interval time T predect , the current driving speed v c of the electric vehicle, and the current heading angle θ c , by the formulas x1=x0+v c *T predect *sin(θ c +D heading ), y1=y0+v c *T predect *cos(θ c +D heading ).

[0135] In an embodiment, when the lane deviation risk level is the first risk level, the second real-time position (longitude: 118.791234, latitude: 32.003456) is directly determined as the path correction starting point. When the lane deviation risk level becomes the second risk level, the lateral position deviation change rate R deviation =0.0002 km / s, and D heading is 1°, the predicted interval time T predect =0.1 / 0.0002=500 s is calculated. Given that the current driving speed v c =20 m / s, the current heading angle θ c =5°, and the second real-time position coordinates are (x0=118.791234, y0=32.003456), the path correction starting point is calculated as x1=118.791234+20*500*sin(6°) / 1000=118.791234+0.10453≈118.89586, y1=32.003456+20*500*cos(6°) / 1000=32.003456+0.994521≈32.997977.

[0136] Step 90, the first autonomous driving path is corrected based on the path correction starting point to obtain a second autonomous driving path.

[0137] Further, the vehicle control system corrects the first autonomous driving path according to the path correction starting point to obtain a second autonomous driving path, specifically as the process of steps 901 to 903.

[0138] The embodiment of the present application can monitor the deviation of the vehicle position from the planned path in real time during the driving of the electric motorcycle, accurately determine the lane deviation risk level, and reasonably determine the path correction starting point according to different risk levels. Not only the deviation of the lateral position and the heading angle is considered, but also the deviation change rate is considered to dynamically adjust the path correction strategy. The deviation during the driving of the vehicle can be responded more timely and accurately, the safety risk caused by lane deviation is effectively reduced, the electric motorcycle is ensured to always drive along a safe and reasonable path during automatic driving, and the reliability and safety of automatic driving are improved.

[0139] In an embodiment, the processes of steps 901 to 903 include:

[0140] Step 901, a plurality of local candidate paths are generated by searching in the high-precision map with the path correction starting point as the center, combining the preset search range and the second lane information. Each local candidate path meets the drivable conditions of the vehicle.

[0141] Optionally, the vehicle control system sets a preset search range (for example, a circular area with a radius of 500 meters) with the path correction starting point as the center, and searches in the high-precision map in combination with the second lane information. During the searching process, the turning radius, acceleration limit and other drivable conditions of the vehicle are considered to generate a plurality of local candidate paths starting from the path correction starting point. Each local candidate path needs to meet the vehicle dynamics constraints to ensure that the electric motorcycle can actually drive. In an embodiment, the path correction starting point is located near the intersection of XX Avenue and XX Road, and the second lane information shows that the vehicle is currently located in the middle lane of XX Avenue from south to north. The vehicle control system generates the following local candidate paths within a radius of 500 meters with the point as the center, combining the road information in the high-precision map and the drivable conditions of the vehicle.

[0142] Path A: continue straight along XX Avenue, turn right into XX Road after 3 intersections;

[0143] Path B: continue straight along XX Avenue, turn left into XX Street after 2 intersections;

[0144] Path C: immediately turn right into XX Branch, and then merge into XX Avenue after 2 intersections.

[0145] Step 902, the deviation index of each local candidate path is determined based on the lateral position deviation and the heading angle deviation of each point in each local candidate path.

[0146] Further, for each point in each local candidate path, the vehicle control system calculates its lateral position deviation D lateral and the heading angle deviation D headingand based on the lateral position deviation and the heading angle deviation of each point, determine a deviation index of each local candidate path, in which the deviation values and their change rates of the points are considered, and the deviation index of each local candidate path is obtained by weighted summation, thus the calculation formula of the deviation index of each local candidate path is as follows:

[0147] wherein n is the number of points on the path, w1, w2, w3, w4 are weight coefficients (here w1=0.4, w2=0.3, w3=0.2, w4=0.1), and are the change rates of the lateral position deviation and the heading angle deviation respectively.

[0148] Continuing to take the local candidate path A as an example, there are 50 points on the path. The lateral position deviation and the heading angle deviation of each point, as well as their change rates, are calculated. It is assumed that after calculation, the deviation index I deviation-A of path A is 12.5. Similarly, the deviation index I deviation-B of path B is 8.3, and the deviation index I deviation-C of path C is 15.7.

[0149] Step 903, determine the path corresponding to the smallest deviation index in the local candidate path as the optimal local path, and fuse the optimal local path with the driving path of the first autonomous driving path after the correction starting point to obtain the second autonomous driving path.

[0150] Further, the vehicle control system compares the deviation indexes of the local candidate paths, and determines the path corresponding to the smallest deviation index as the optimal local path. Further, the vehicle control system fuses the optimal local path with the driving path of the first autonomous driving path after the correction starting point. In the fusion process, the continuity and differentiability of the path at the correction starting point are ensured, and the seamless connection of the two paths is realized through a smooth transition function, and finally the second autonomous driving path is obtained.

[0151] Continuing in the above example, the deviation index of path B is the smallest, and thus path B is determined as the optimal local path. It is assumed that the driving path of the first autonomous driving path after the correction starting point is to continue straight along XX Avenue to XX Science and Technology Park, and path B (continue straight along XX Avenue, turn left into XX Street after passing through 2 intersections) is fused with the original path, a quintic polynomial smooth transition function is adopted to ensure the continuity of the position, direction and curvature of the path at the correction starting point. The second autonomous driving path after fusion is: from the correction starting point, drive along path B to XX Street, and then adjust the route according to the actual situation to go to XX Science and Technology Park.

[0152] The embodiment of the present application can generate multiple candidate paths, evaluate deviation indexes and fuse optimal paths when the electric motorcycle appears a lane deviation risk, thereby not only ensuring that the corrected path meets the vehicle dynamics constraints, but also minimizing the deviation from the original planned path, realizing smooth transition of the path, enabling better response to complex road conditions and unexpected situations, improving the safety and reliability of the automatic driving of the electric motorcycle, and ensuring that the vehicle can efficiently and stably reach the destination according to the corrected path.

[0153] Further, the electric vehicle automatic driving path planning system based on high-precision positioning provided by the present application is described below, and the electric vehicle automatic driving path planning system based on high-precision positioning described below can be mutually corresponding with the electric vehicle automatic driving path planning method based on high-precision positioning described above.

[0154] Reference Figure 3 , Figure 3 is a structural schematic diagram of the electric vehicle automatic driving path planning system based on high-precision positioning provided by the present application, and the electric vehicle automatic driving path planning system based on high-precision positioning comprises:

[0155] The data acquisition module 310 is configured to acquire the first real-time position of the electric vehicle based on the high-precision positioning device, and acquire the current power, the battery health state information and the set destination position of the electric vehicle.

[0156] The data matching module 320 is configured to match the first real-time position with the pre-stored high-precision map, determine the first lane information where the electric vehicle is currently located, and determine the road information of each road between the first real-time position and the destination position based on the high-precision map.

[0157] The charging pile positioning module 330 is configured to determine the maximum driving range based on the current power and the battery health state information, and determine the target charging pile based on the first real-time position, the maximum driving range, and the charging pile position information and the current use state information of each charging pile in the high-precision map.

[0158] The path planning module 340 is configured to plan a path based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile, and obtain a first automatic driving path; the road information includes road speed limit, lane number, lane direction and traffic signal lamp state; the distance between each charging pile in the target charging pile is less than or equal to the first preset distance threshold.

[0159] The embodiment of the present application determines the lane information where the electric vehicle is currently located and the road information of each road in the driving process by combining the real-time position and the destination position with the high-precision map, so that the real-time state of the road can be accurately perceived. On the other hand, within the maximum driving range determined according to the current electric quantity and the battery health state information, the charging pile position information and the current use state information of each charging pile in the high-precision map are combined to screen the appropriate target charging pile, effectively solving the range anxiety problem of the electric vehicle. Finally, the automatic driving path planned according to the real-time position, the destination position, the lane information, the road information and the target charging pile not only considers the charging pile but also takes into account the real-time state of the road, effectively avoiding the vehicle stranded halfway due to the electric quantity problem, and at the same time, the real-time road condition can also be perceived under complex road conditions, improving the accuracy and safety of path planning in the electric vehicle automatic driving scene.

[0160] Referring to Figure 4 , Figure 4 An embodiment of an electronic device provided by the present application is shown in the figure. As shown in Figure 4 , the present application provides an electronic device 400, which includes a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, the following steps are implemented:

[0161] obtain a first real-time position of the electric vehicle based on a high-precision positioning device, and obtain the current electric quantity, the battery health state information and the set destination position of the electric vehicle;

[0162] match the first real-time position with a pre-stored high-precision map to determine the first lane information where the electric vehicle is currently located, and determine the road information of each road between the first real-time position and the destination position based on the high-precision map;

[0163] determine the maximum driving range based on the current electric quantity and the battery health state information, and determine the target charging pile based on the first real-time position, the maximum driving range, and the charging pile position information and the current use state information of each charging pile in the high-precision map;

[0164] plan a path based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile to obtain a first automatic driving path; the distance between each charging pile in the target charging pile is less than or equal to a first preset distance threshold.

[0165] Referring to Figure 5 , Figure 5 An embodiment of a computer-readable storage medium provided by the present application is shown in the figure. As shown in Figure 5As shown, the embodiment provides a computer readable storage medium 500, which stores a computer program 411, and the computer program 411 is executed by a processor to implement the following steps:

[0166] obtain a first real-time position of the electric vehicle based on a high-precision positioning device, and obtain a current power, a battery health state information and a set destination position of the electric vehicle;

[0167] match the first real-time position with a pre-stored high-precision map, determine first lane information of the electric vehicle currently located, and determine road information of each road between the first real-time position and the destination position based on the high-precision map;

[0168] determine a maximum driving range based on the current power and the battery health state information, and determine a target charging pile based on the first real-time position, the maximum driving range, and charging pile position information and current use state information of each charging pile in the high-precision map;

[0169] plan a path based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile, to obtain a first automatic driving path; distances between each charging pile in the target charging pile are less than or equal to a first preset distance threshold.

[0170] On the other hand, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program is executed by a processor, and the computer can execute the high-precision positioning based electric vehicle automatic driving path planning method provided by each method, which comprises:

[0171] obtain a first real-time position of the electric vehicle based on a high-precision positioning device, and obtain a current power, a battery health state information and a set destination position of the electric vehicle;

[0172] match the first real-time position with a pre-stored high-precision map, determine first lane information of the electric vehicle currently located, and determine road information of each road between the first real-time position and the destination position based on the high-precision map;

[0173] determine a maximum driving range based on the current power and the battery health state information, and determine a target charging pile based on the first real-time position, the maximum driving range, and charging pile position information and current use state information of each charging pile in the high-precision map;

[0174] plan a path based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile, to obtain a first automatic driving path; distances between each charging pile in the target charging pile are less than or equal to a first preset distance threshold.

[0175] The system embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0176] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0177] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A high-precision positioning-based automatic driving path planning method for electric vehicles, characterized in that, The method comprises the following steps: obtaining the first real-time position of the electric vehicle based on a high-precision positioning device, and obtaining the current power, battery health state information and set destination position of the electric vehicle; matching the first real-time position with a pre-stored high-precision map to determine the first lane information of the electric vehicle, and determining the road information of each road between the first real-time position and the destination position based on the high-precision map; determining the maximum driving range based on the current power and battery health state information, and determining the target charging pile based on the first real-time position, the maximum driving range, and the charging pile position information and current use state information of each charging pile in the high-precision map; path planning based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile to obtain a first automatic driving path; the distance between each charging pile in the target charging pile is less than or equal to a first preset distance threshold; wherein, after obtaining the first automatic driving path, the method further comprises the following steps: determining the second lane information of the electric vehicle in the high-precision map based on the second real-time position of the electric vehicle obtained by the high-precision positioning device in real time during the driving of the electric vehicle; determining the lateral position deviation, heading angle deviation and deviation change rate of the electric vehicle during driving based on the second lane information and the first automatic driving path; determining the lane deviation risk level based on the lateral position deviation and the heading angle deviation and their corresponding deviation threshold; the lane deviation risk level comprises a first risk level and a second risk level, and the second risk level is higher than the first risk level; if the lane deviation risk level is the first risk level, the second real-time position is determined as the path correction starting point; if the lane deviation risk level is the second risk level, a prediction interval time is determined based on the deviation change rate, and a path correction starting point is determined based on the second real-time position, the prediction interval time, and the current driving speed and current heading angle of the electric vehicle; correcting the first automatic driving path based on the path correction starting point to obtain a second automatic driving path; the step of obtaining the second automatic driving path comprises the following steps: generating a plurality of local candidate paths by taking the path correction starting point as the center, combining a preset search range and the second lane information in the high-precision map; each local candidate path meets the vehicle drivable condition; determining the deviation index of each local candidate path based on the lateral position deviation and the heading angle deviation of each point in each local candidate path; determining the optimal local path corresponding to the smallest deviation index in the local candidate path, and fusing the optimal local path with the driving path after the correction starting point of the first automatic driving path to obtain the second automatic driving path. 2.The high-precision positioning-based automatic driving path planning method for electric vehicles according to claim 1, characterized in that, The road information includes a road speed limit, a number of lanes, a lane direction, and a traffic signal state; the path planning based on the first real-time position, the destination position, the first lane information, the road information, and the target charging pile obtains a first autonomous driving path, including: a node is constructed by taking the target charging pile and road intersections of each road between the first real-time position and the destination position as nodes, a road segment between adjacent intersections and a position distance between each intersection and the target charging pile as edges, and a road speed limit, a number of lanes, a lane direction, and a traffic signal state between adjacent intersections and between each intersection and the target charging pile as edge attributes; a directed graph is constructed; effective road segments are screened out based on the directed graph and a safe maximum driving speed of the electric vehicle according to a road speed limit of each road segment; a passable lane that can be traveled from the first real-time position in the effective road segments is screened out based on the first lane information and a lane direction and lane compatibility of each lane of each road segment in the effective road segments according to the directed graph; the lane compatibility represents that a lane type and a lane authority of the lane are consistent with a vehicle type of the electric vehicle; a target road segment is screened out based on the directed graph and a position distance between each road segment and the target charging pile, and the first autonomous driving path is obtained by path planning based on the directed graph, the passable lane, and the target road segment. 3.The high-precision positioning-based automatic driving path planning method for electric vehicles according to claim 2, characterized in that, The path planning based on the directed graph, the passable lane, and the target road segment to obtain the first autonomous driving path includes: an initial path is generated based on the directed graph and connectivity of adjacent lanes in the passable lane and connectivity of adjacent road segments in the target road segment, with the first real-time position as a starting point and the destination position as an end point; each path includes a plurality of continuous lanes and road segments; for each road segment of each path in the initial path, an ideal travel time of each road segment under the influence of no traffic signal is determined based on the directed graph and a road speed limit and a road length of each lane, and an actual travel time of each road segment is determined based on the ideal travel time and a signal light period, a red light duration, and a current red light remaining time of each lane; a charging consumption time of each road segment when charging is predicted based on a position distance between each road segment and the target charging pile, and a road total length of each road segment is determined based on a road length of each lane; a first candidate path is determined based on the road total length, the charging consumption time, and the actual travel time of each road segment, and the first autonomous driving path is determined based on a remaining number of lanes and a number of charging piles containing the target charging pile of each path in the first candidate path. 4.The high-precision positioning-based automatic driving path planning method for electric vehicles according to claim 3, characterized in that, The determination of the first autonomous driving path based on the remaining number of lanes and the number of charging piles containing the target charging pile of each path in the first candidate path includes: a path in which the remaining number of lanes is greater than a first preset number and the number of charging piles is greater than a second preset number in the first candidate path is determined as a second candidate path; Screening the second candidate path by taking the sum of the change rates of the turning angles of the adjacent road segments in each path as the target to obtain a third candidate path; Screening the third candidate path by taking the sum of the lane curvatures of the adjacent lanes of each road segment in each path as the target to obtain a fourth candidate path; Screening the fourth candidate path by taking the sum of the transformation times of the adjacent lanes of each road segment in each path as the target to obtain the first automatic driving path. 5.The high-precision positioning-based automatic driving path planning method for electric vehicles according to claim 1, characterized in that, The target charging pile is determined based on the first real-time position, the maximum driving range, and charging pile position information and current use state information of each charging pile in the high-precision map, and includes: All charging piles that can be reached by the electric vehicle within the maximum driving range are screened out according to the first real-time position and the charging pile position information of each charging pile to obtain reachable charging piles; For each charging pile in the reachable charging piles, a use state history sequence within a preset time window is constructed based on the current use state information thereof, and a probability sequence of each charging pile being idle is determined according to the frequency of occurrence of an idle state in the use state history sequence; The waiting time of the electric vehicle when reaching each charging pile is estimated based on the current use state of each charging pile and the driving trajectory of a vehicle within a preset range thereof to obtain the waiting time. The target charging pile is determined based on the probability sequence of each charging pile being idle and the waiting time. 6.The high-precision positioning-based automatic driving path planning method for electric vehicles according to claim 5, characterized in that, The target charging pile is determined based on the probability sequence of each charging pile being idle and the waiting time, and includes: For each charging pile in the reachable charging piles, a distance loss value of each charging pile is determined according to the distance between the first real-time position and the charging pile position information of each charging pile, and a comprehensive evaluation index of each charging pile is determined based on the distance loss value, the probability sequence of each charging pile being idle, and the waiting time; A charging pile with a comprehensive evaluation index greater than or equal to a preset index threshold in the reachable charging piles is determined as a candidate charging pile, and a device health index of each charging pile is determined based on the number of faults and the cumulative downtime of each charging pile in the candidate charging piles; A charging pile with a device health index greater than or equal to a preset health threshold and a distance from the first real-time position less than or equal to a preset charging pile is determined as the target charging pile. 7.The high-precision positioning-based automatic driving path planning method for electric vehicles according to claim 5, characterized in that, The waiting time of the electric vehicle when reaching each charging pile is estimated based on the current use state of each charging pile and the driving trajectory of a vehicle within a preset range thereof to obtain the waiting time, and includes: If the current use state is an idle state, the waiting time is determined to be 0; If the current use state is an occupied state, the total number of vehicles of each charging pile, the remaining charging time of the charging vehicle, and the vehicle that is expected to arrive first are determined; The charging time is estimated based on the total number of vehicles, the remaining charging time, and the average charging operation time of each vehicle of each charging pile to determine the waiting time when reaching each charging pile.

8. A high-precision positioning-based automatic driving path planning system for electric vehicles, characterized in that, The application is applied to the high-precision positioning-based electric vehicle automatic driving path planning method according to any one of claims 1 to 7; the high-precision positioning-based electric vehicle automatic driving path planning system comprises: a data acquisition module configured to acquire a first real-time position of an electric vehicle based on a high-precision positioning device, and acquire current power, battery health state information and a set destination position of the electric vehicle; a data matching module configured to match the first real-time position with a pre-stored high-precision map, determine first lane information of the electric vehicle, and determine road information of each road between the first real-time position and the destination position based on the high-precision map; a charging pile positioning module configured to determine a maximum driving range based on the current power and the battery health state information, and determine a target charging pile based on the first real-time position, the maximum driving range, and charging pile position information and current use state information of each charging pile in the high-precision map; a path planning module configured to plan a first automatic driving path based on the first real-time position, the destination position, the first lane information, the road information and the target charging pile; the road information comprises road speed limit, lane number, lane direction and traffic signal state; distances between each charging pile in the target charging pile are less than or equal to a first preset distance threshold; wherein, after the first automatic driving path is obtained, the method further comprises: during driving of the electric vehicle, determining second lane information of the electric vehicle in the high-precision map based on a second real-time position of the electric vehicle acquired by the high-precision positioning device in real time; determining a lateral position deviation, a heading angle deviation and a deviation change rate of the electric vehicle during driving based on the second lane information and the first automatic driving path; determining a lane deviation risk level based on the lateral position deviation, the heading angle deviation and corresponding deviation threshold values; the lane deviation risk level comprises a first risk level and a second risk level, and the second risk level is higher than the first risk level; if the lane deviation risk level is the first risk level, determining the second real-time position as a path correction starting point; if the lane deviation risk level is the second risk level, determining a prediction interval time based on the deviation change rate, and determining a path correction starting point based on the second real-time position, the prediction interval time and a current driving speed and a current heading angle of the electric vehicle; correcting the first automatic driving path based on the path correction starting point to obtain a second automatic driving path; the step of obtaining the second automatic driving path comprises: generating a plurality of local candidate paths in the high-precision map based on the path correction starting point, a preset search range and the second lane information; each local candidate path satisfies a vehicle drivable condition; determining a deviation index of each local candidate path based on a lateral position deviation and a heading angle deviation of each point in each local candidate path. The path corresponding to the minimum deviation index in the local candidate paths is determined as an optimal local path, and the optimal local path is fused with the driving path after the correction starting point of the first automatic driving path to obtain the second automatic driving path.

Images