Multi-vehicle collaborative path planning method based on Internet of Vehicles

Abstract

The present invention relates to the field of vehicle networking and intelligent transportation technology, and in particular to a multi-vehicle collaborative path planning method based on the vehicle networking. Aiming at the limitations of real-time path planning based on real-time traffic flow data during long-distance driving, the present invention segments all feasible driving paths, and based on the real-time vehicle data of the first vehicle, the number of second vehicles on each road section, and the time it takes for the vehicle to be processed to travel to each road section, predicts the expected second vehicle density of each road section when the vehicle to be processed travels to each road section by road section and time period. Real-time path planning is performed based on the predicted expected second vehicle density of each road section when the vehicle to be processed travels to each road section. By segmented spatiotemporal prediction and dynamic density adaptation, the time period density prediction accurately matches the arrival time of the vehicle to be processed, thereby solving the lag and blindness problems of traditional real-time path planning in long-distance scenarios.

Description

Technical Field

[0001] The present invention relates to the field of vehicle networking and intelligent transportation technology, and in particular to a multi-vehicle collaborative path planning method based on the vehicle networking. Background Art

[0002] In recent years, with the continuous improvement of people's quality of life, cars have become their primary means of transportation, resulting in an annual increase in car sales. With the development of intelligent vehicles, the Internet of Vehicles (IoV) uses wireless communication technology to connect vehicles with other vehicles, road infrastructure, pedestrians, and cloud systems in the surrounding environment into an intelligent collaborative network, enabling real-time data interaction and sharing, thereby improving traffic efficiency, safety, and intelligence. Path planning directly affects efficiency, cost, and safety, and is also a key support for the realization of autonomous driving and smart cities.

[0003] Currently, existing vehicle-based routing methods often achieve real-time routing by acquiring real-time traffic flow data and performing real-time road condition analysis. For example, Chinese Patent Publication No. CN108286981B discloses a vehicle routing method, apparatus, and computer device for the vehicle routing system. The method includes acquiring vehicle data from connected vehicle terminals in the vehicle routing system and storing the acquired vehicle data; dividing vehicle regions into corresponding road sections based on the longitude and latitude data of the vehicle data; performing road condition analysis for each road section based on speed and direction data; acquiring and storing traffic flow data for each road section; receiving a vehicle routing request from a requesting end, the vehicle routing request including the longitude and latitude parameters of the current location and destination of the vehicle to be routed; searching the stored traffic flow data for the road section corresponding to the longitude and latitude parameters based on the vehicle routing request; performing real-time traffic flow data analysis on the corresponding road section; acquiring a routing result for the vehicle to be routed; and returning the routing result to the requesting end. This method can improve the accuracy and efficiency of vehicle routing. It can be seen that the existing technology is to achieve real-time path planning by obtaining real-time traffic flow data for real-time road condition analysis, which has at least the following problems: on the one hand, for long-distance driving routes, such as driving on a highway, the real-time route planned at the time of departure is appropriate, but it is only applicable at this moment. There is a long time interval between the time the vehicle departs and the process of getting on and off the highway. As time goes by, the vehicle data and road conditions on the planned route are constantly changing. It is very likely that the route planned at the time of departure or when getting on the highway is not applicable when actually driving on the highway. There is a lack of accurate prediction of future road condition changes, and there are long-term planning limitations. On the other hand, the current path planning is not comprehensive. For example, not all vehicles have vehicle networking functions, or some people do not agree to upload data for safety and privacy reasons, which will lead to a large difference between the road condition analysis results and the actual road conditions, making the path planning effect unsatisfactory. Summary of the Invention

[0004] To this end, the present invention provides a multi-vehicle collaborative path planning method based on the Internet of Vehicles, which is used to overcome the problems in the existing technology that long-distance path planning is not applicable and path planning is not comprehensive, resulting in inconsistency between the path planning traffic flow and the actual traffic flow under the driving conditions.

[0005] To achieve the above objectives, the present invention provides a multi-vehicle collaborative path planning method based on the Internet of Vehicles, comprising:

[0006] In response to preset information to be processed, determining several feasible driving paths for the vehicle according to the starting point and the end point of the vehicle to be processed, and segmenting each feasible driving path;

[0007] Retrieving in real time all real-time vehicle path planning data of other connected vehicles in the vehicle network, obtaining vehicles corresponding to the real-time vehicle path planning associated with any road segment of each feasible driving path from all the real-time vehicle path planning data, recording them as first vehicles, and obtaining in real time vehicle data of each first vehicle, including the number of first vehicles, the position of each first vehicle, and the speed of each first vehicle;

[0008] Detect vehicles on each road section in real time, record them as second vehicles, and obtain the number of second vehicles on each road section in real time;

[0009] Obtain the real-time speed of the vehicle to be processed, and determine the time it takes for the vehicle to be processed to travel to each road section based on the real-time speed of the vehicle to be processed;

[0010] Determining an expected second vehicle density of each road section when the vehicle to be processed travels to each road section based on the vehicle data of the first vehicle, the number of second vehicles on each road section, and the time it takes for the vehicle to be processed to travel to each road section;

[0011] Performing path planning based on the expected second vehicle density of each road section when the vehicle to be processed travels to each road section, and determining a vehicle planning path for the vehicle to be processed;

[0012] Other networked vehicles that re-planned their routes between the time the networked vehicle data was acquired and the current time are recorded as third vehicles, and based on the third vehicle data, a vehicle to be processed is determined to execute the pre-planned route of the vehicle or to re-acquire the vehicle data of the first vehicle;

[0013] When executing the vehicle pre-planned path, the vehicle planned path of the vehicle to be processed is uploaded to all real-time vehicle path planning data of networked vehicles in the vehicle network.

[0014] Furthermore, the time it takes for the vehicle to be processed to travel to each road section is determined, including:

[0015] The length of each road section and the real-time position of the vehicle to be processed are obtained, and the time it takes for the vehicle to be processed to travel to each road section is determined based on the real-time speed of the vehicle to be processed, the real-time position of the vehicle to be processed and the length of each road section.

[0016] Furthermore, the position of the first vehicle when the vehicle to be processed travels to each road section is determined based on the time when the vehicle to be processed travels to each road section, the speed data of the first vehicle and the position data of the first vehicle; and the number of first vehicles corresponding to each road section when the vehicle to be processed travels to each road section is determined based on the position of the first vehicle when the vehicle to be processed travels to each road section.

[0017] Furthermore, the real-time first vehicle density of each road section is determined according to the length of each road section and the real-time number of first vehicles on each road section, and the real-time second vehicle density of each road section is determined according to the length of each road section and the real-time number of second vehicles on each road section.

[0018] Furthermore, determining the expected second vehicle density includes:

[0019] The expected second vehicle density of each road section when the vehicle to be processed travels to each road section is determined based on the first vehicle number of each road section, the real-time first vehicle density of each road section, and the real-time second vehicle density of each road section.

[0020] Furthermore, the congestion coefficient of each road section is determined according to the expected second vehicle density of each road section when the vehicle to be processed travels to each road section, and the path planning is performed according to the congestion coefficient of each road section to determine the vehicle pre-planned path of the vehicle to be processed.

[0021] Furthermore, it also includes: determining the preference ranking of each feasible driving path according to the expected second vehicle density and preference allocation weight of each road section when the vehicle to be processed travels to each road section, and determining the vehicle pre-planned path of the vehicle to be processed according to the preference ranking of each feasible driving path.

[0022] Furthermore, segmenting each feasible driving path includes:

[0023] Count all exits on each feasible driving path of the vehicle to be processed;

[0024] Each feasible driving path is segmented according to all exits on the feasible driving path, and the area between two adjacent exits on the same feasible driving path is a road segment.

[0025] Furthermore, it also includes:

[0026] A correction coefficient is determined based on the real-time second vehicle density and expected second vehicle density of each road section on which the vehicle to be processed has traveled on the pre-planned path of the vehicle to be processed, and the expected second vehicle density of each road section when the vehicle to be processed travels to each road section is corrected based on the correction coefficient.

[0027] Furthermore, determining that the vehicle to be processed executes the pre-planned vehicle path or re-acquires the vehicle data of the first vehicle based on the third vehicle data includes:

[0028] If the number of the third vehicles is less than a preset threshold, the vehicle to be processed executes the vehicle pre-planned path;

[0029] If the number of the third vehicles is greater than or equal to a preset threshold, the vehicle data of the first vehicle is reacquired.

[0030] Compared with the prior art, the present invention has the following advantages:

[0031] The present invention provides a multi-vehicle collaborative path planning method based on the Internet of Vehicles. Aiming at the limitations of real-time path planning based on real-time traffic flow data in long-distance driving, each feasible driving path is segmented. Based on the real-time vehicle data of the first vehicle, the number of second vehicles on each road section and the time when the vehicle to be processed travels to each road section, the expected second vehicle density of each road section when the vehicle to be processed travels to each road section is predicted by section and time period. Real-time path planning is performed based on the predicted expected second vehicle density of each road section when the vehicle to be processed travels to each road section. Through segmented spatiotemporal prediction and dynamic density adaptation, the time period density prediction accurately matches the arrival time of the vehicle to be processed, thereby solving the problems of lag and blindness of traditional real-time path planning in long-distance scenarios.

[0032] Furthermore, there are a large number of vehicles connected to the Internet of Vehicles, and the amount of real-time data generated is huge. The first vehicle of the present invention is a networked vehicle in the Internet of Vehicles that is related to the feasible driving path of the vehicle to be processed. That is, the networked vehicles in the Internet of Vehicles are initially screened, and only vehicles related to the feasible driving path of the vehicle to be processed are selected. This can effectively reduce the network transmission pressure between the on-board terminal of the vehicle to be processed and the cloud platform, thereby reducing communication delays and improving the real-time nature of data.

[0033] Furthermore, the real-time vehicle planning path of any networked vehicle of the present invention is uploaded to all real-time vehicle path planning data of networked vehicles in the vehicle network in real time, that is, all real-time vehicle path planning data of networked vehicles in the vehicle network is updated in real time, that is, the first vehicle of the present application is also updated according to all real-time vehicle path planning data of networked vehicles in the vehicle network. The first vehicle includes vehicles that have not yet traveled to the feasible driving paths of the vehicle to be processed, which are vehicles that will travel to the feasible driving paths of the vehicle to be processed in the future according to the vehicle planning paths, and vehicles that are traveling on the feasible driving paths of the vehicle to be processed. When the real-time vehicle planning path of a new vehicle is related to the feasible driving path of the vehicle to be processed, it will be added to the first vehicle in a timely manner. When the first vehicle that is traveling on the feasible driving path of the vehicle to be processed leaves the feasible driving path of the vehicle to be processed, it will also be removed from the first vehicle in a timely manner to ensure the real-time and accuracy of the first vehicle data.

[0034] Furthermore, the second vehicle of the present invention is a real-time vehicle traveling on each section of all feasible driving paths of the vehicle to be processed, including the first vehicle (connected vehicle in the Internet of Vehicles) and other vehicles traveling on each section of all feasible driving paths of the vehicle to be processed. Other vehicles include vehicles that do not have the Internet of Vehicles function, vehicles that have not turned on the Internet of Vehicles function, etc. Path planning is performed based on the expected second vehicle density of each section when the vehicle to be processed travels to each section. Compared with path planning based only on the first vehicle or the first vehicle density, it takes more comprehensive considerations, based on the prediction of the full vehicle density (second vehicle density) rather than relying solely on local data uploaded by the Internet of Vehicles, reducing misjudgment of road conditions due to missing data, and avoiding deviations in road condition analysis due to some vehicles not being connected to the Internet. For example, the actual traffic condition is congested but there are only a few connected vehicles, resulting in fewer vehicles in the displayed road conditions. This improves the consistency between the road condition analysis results and the actual road condition data, and significantly improves the consistency between the path planning results and the actual road conditions.

[0035] Furthermore, the present invention predicts the total vehicle density (expected second vehicle density) when the vehicles to be processed arrive at each road section in real time, rather than relying solely on the static data of networked vehicles at the current moment, thereby avoiding planning failures caused by time differences and avoiding the problem that a certain road section is unobstructed at departure, but congestion is caused by the gathering of non-networked vehicles when the vehicles to be processed arrive.

[0036] Furthermore, the present invention determines a correction coefficient for the real-time second vehicle density and expected second vehicle density of each historical road section that has been traveled according to the vehicle planning path of the vehicle to be processed, and corrects the expected second vehicle density of each road section when the vehicle to be processed travels to each road section next according to the correction coefficient. Based on the actual second vehicle density data of the historical road sections, the inherent deviation of the prediction model is corrected. The correction coefficient represents the distribution pattern of non-networked vehicles in history, such as the gathering of commuting vehicles during holiday peak hours, the impact of the same time period and similar weather on traffic flow, etc., so that the prediction is closer to the actual full traffic flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a multi-vehicle collaborative path planning method based on the Internet of Vehicles according to an embodiment of the present invention;

[0038] Figure 2 A schematic diagram of all feasible driving routes determined according to the starting point and the end point in an embodiment of the present invention;

[0039] Figure 3 A schematic diagram of segmenting a path according to exits on a feasible driving path according to an embodiment of the present invention;

[0040] Figure 4 This is a flowchart of step S5 of an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0042] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0043] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside", and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0044] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0045] See also Figure 1 As shown, an embodiment of the present invention provides a multi-vehicle collaborative path planning method based on the Internet of Vehicles, including:

[0046] Step S1, in response to preset information to be processed, determining several feasible driving paths for the vehicle according to the starting point and end point of the vehicle to be processed, and segmenting each feasible driving path;

[0047] Step S2: Retrieving all real-time vehicle path planning data of other connected vehicles in the vehicle network in real time, obtaining vehicles corresponding to the real-time vehicle planned paths associated with any road segment of each feasible driving path from all real-time vehicle path planning data, recording them as first vehicles, and obtaining vehicle data of each first vehicle in real time, including the number of first vehicles, the position of each first vehicle, and the speed of each first vehicle;

[0048] Step S3: detecting vehicles on each road section in real time, recording them as second vehicles, and obtaining the number of second vehicles on each road section in real time;

[0049] Step S4, obtaining the real-time speed of the vehicle to be processed, and determining the time it takes for the vehicle to be processed to travel to each road section according to the real-time speed of the vehicle to be processed;

[0050] Step S5, determining the expected second vehicle density of each road section when the vehicle to be processed travels to each road section based on the vehicle data of the first vehicle, the number of second vehicles on each road section, and the time it takes for the vehicle to be processed to travel to each road section;

[0051] Step S6, performing path planning based on the expected second vehicle density of each road section when the vehicle to be processed travels to each road section, and determining a pre-planned vehicle path for the vehicle to be processed;

[0052] Other networked vehicles that re-planned their routes between the time the networked vehicle data was acquired and the current time are recorded as third vehicles, and based on the third vehicle data, a vehicle to be processed is determined to execute the pre-planned route of the vehicle or to re-acquire the vehicle data of the first vehicle;

[0053] When executing the vehicle pre-planned path, the vehicle planned path executed by the vehicle to be processed is uploaded to all real-time vehicle path planning data of networked vehicles in the vehicle network.

[0054] The present invention provides a multi-vehicle collaborative path planning method based on the Internet of Vehicles. Aiming at the limitations of real-time path planning based on real-time traffic flow data in long-distance driving, each feasible driving path is segmented. Based on the real-time vehicle data of the first vehicle, the number of second vehicles on each road section and the time when the vehicle to be processed travels to each road section, the expected second vehicle density of each road section when the vehicle to be processed travels to each road section is predicted by section and time period. Real-time path planning is performed based on the predicted expected second vehicle density of each road section when the vehicle to be processed travels to each road section. Through segmented spatiotemporal prediction and dynamic density adaptation, the time period density prediction accurately matches the arrival time of the vehicle to be processed, thereby solving the problems of lag and blindness of traditional real-time path planning in long-distance scenarios.

[0055] It can be understood that the meaning of being related to any road segment of each feasible driving path in the real-time vehicle path planning data is that the real-time vehicle planning path in the real-time vehicle path planning data includes any road segment on all feasible driving paths of the vehicle to be processed.

[0056] It can be understood that the data acquisition in step S2 and step S3 in this embodiment is performed simultaneously.

[0057] As an embodiment, when the vehicle to be processed is path-planned by the multi-vehicle collaborative path planning method based on the Internet of Vehicles of the present invention, it is executed in real time and in a continuous loop, for example, once every minute.

[0058] The first vehicle of the present invention is a vehicle whose real-time vehicle planning path of the networked vehicles in the vehicle network is related to each feasible driving path of the vehicle to be processed, and the second vehicle is a real-time vehicle on each road section on each feasible driving path of the vehicle to be processed.

[0059] There are many vehicles connected to the Internet of Vehicles, and the amount of real-time data generated is huge. The first vehicle of the present invention is a networked vehicle in the Internet of Vehicles that is related to the feasible driving path of the vehicle to be processed. That is, the networked vehicles in the Internet of Vehicles are initially screened, and only the vehicles related to the feasible driving path of the vehicle to be processed are selected. This can effectively reduce the network transmission pressure between the on-board terminal of the vehicle to be processed and the cloud platform, thereby reducing communication delays and improving the real-time nature of data.

[0060] The real-time vehicle planning path of any networked vehicle of the present invention is uploaded to all real-time vehicle path planning data of networked vehicles in the vehicle network in real time, that is, all real-time vehicle path planning data of networked vehicles in the vehicle network is updated in real time, that is, the first vehicle of the present application is also updated according to all real-time vehicle path planning data of networked vehicles in the vehicle network. The first vehicle includes vehicles that have not yet traveled to the feasible driving paths of the vehicle to be processed, which are vehicles that will travel to the feasible driving paths of the vehicle to be processed in the future according to the vehicle planning paths, and vehicles that are traveling on the feasible driving paths of the vehicle to be processed. When the real-time vehicle planning path of a new vehicle is related to the feasible driving path of the vehicle to be processed, it will be added to the first vehicle in time. When the first vehicle that is traveling on the feasible driving path of the vehicle to be processed leaves the feasible driving path of the vehicle to be processed, it will also be removed from the first vehicle in time to ensure the real-time and accuracy of the first vehicle data.

[0061] The second vehicle of the present invention is a real-time vehicle traveling on each section of all feasible driving paths of the vehicle to be processed, that is, it includes the first vehicle (connected vehicle in the vehicle network) and other vehicles traveling on each section of all feasible driving paths of the vehicle to be processed. The other vehicles include vehicles that do not have the vehicle network function, vehicles that have not turned on the vehicle network function, etc. The path planning is performed based on the expected second vehicle density of each section when the vehicle to be processed travels to each section. Compared with path planning based only on the first vehicle or the first vehicle density, it takes more comprehensive considerations, based on the prediction of the full vehicle density (second vehicle density) rather than relying solely on local data uploaded by the vehicle network, reducing road condition misjudgment caused by data missing, and avoiding road condition analysis deviation caused by some vehicles not being connected to the network. For example, the actual traffic condition is congested but there are few connected vehicles, resulting in few vehicles displayed in the red road condition. This improves the consistency between the road condition analysis results and the actual road condition data, and significantly improves the consistency between the path planning results and the actual road conditions.

[0062] The present invention predicts the total vehicle density (expected second vehicle density) when the vehicles to be processed arrive at each road section in real time, rather than relying solely on the static data of networked vehicles at the current moment. This avoids planning failures caused by time differences and avoids the problem that a certain road section is unobstructed at departure, but congestion occurs due to the gathering of non-networked vehicles when the vehicles to be processed arrive.

[0063] In scenarios where Internet of Vehicles coverage is insufficient (such as extreme weather or signal blind spots), this invention supplements data from non-connected vehicles by installing vehicle detection sensors at the entrances and exits of each road section. This technology can generate reliable routes even in weak or no network conditions, avoiding incorrect route planning due to reliance on cloud data.

[0064] In one embodiment, see Figure 2Point A is the starting point of the vehicle to be processed, and point G is the end point of the vehicle to be processed. According to points A and G, all feasible driving paths of the vehicle to be processed are ABEFG, ABCEFG, ABC-D1-F1-G, ABC-D2-F2-G, AB-D1-F1-G, and AB-D2-F2-G, where D→F has two paths, namely the straight path D1-F1 on the left (red) and the curved path D2-F2 on the right (green).

[0065] As an embodiment, in step S1, the preset information to be processed includes all exit information on the path and the guide line starting point information of all exits on the path. This information can be obtained in advance based on the traffic signs, guide lines, signs, etc. of each road section, so it will not be repeated here.

[0066] Specifically, in step S1, segmenting each feasible driving path includes:

[0067] Count all exits on each feasible driving path of the vehicle to be processed;

[0068] Each feasible driving path is segmented based on all exits on it. A segment is defined as the distance between two adjacent exits on the same feasible driving path. Exits are recorded as intersections where one can exit the driving path.

[0069] As another embodiment, in step S1, segmenting each feasible driving path includes:

[0070] Count all exits on each feasible driving path of the vehicle to be processed;

[0071] Each feasible driving path is segmented according to the guide lines of all exits on each feasible driving path, and the area between the starting points of the guide lines at two adjacent exits on the same feasible driving path is a road segment.

[0072] In one embodiment, see Figure 3 Taking the path ABCEFG as an example, there are exits B, a, c, C and f on the path ABCEFG. The path ABEFG is divided into: segment AB, segment Ba, segment ac, segment cC, segment Cf and segment fG.

[0073] Specifically, in step S4, determining the time it takes for the vehicle to be processed to travel to each road section includes:

[0074] The length of each road section and the real-time position of the vehicle to be processed are obtained, and the time it takes for the vehicle to be processed to travel to each road section is determined based on the real-time speed of the vehicle to be processed, the real-time position of the vehicle to be processed and the length of each road section.

[0075] In one embodiment, taking the path ABCEFG as an example, the vehicle to be processed is traveling on section AB. The distance from the vehicle to be processed to exit B can be determined based on the real-time position of the vehicle to be processed (time t0). Based on the distance from the vehicle to be processed to exit B and the real-time speed of the vehicle to be processed, the time t1 required for the vehicle to be processed to reach section Ba (at intersection B) can be determined. Then, the time t2 required for the vehicle to be processed to reach section ac (at intersection a), the time t3 required for the vehicle to be processed to reach section cC (at intersection c), the time t4 required for the vehicle to be processed to reach section Cf (at intersection C), and the time t5 required for the vehicle to be processed to reach section fG (at intersection f) can be determined in sequence.

[0076] See also Figure 4 , step S5 includes:

[0077] Step S51, determining the position of the first vehicle when the vehicle to be processed travels to each road section based on the time when the vehicle to be processed travels to each road section, the speed data of the first vehicle and the position data of the first vehicle, and determining the number of first vehicles corresponding to each road section when the vehicle to be processed travels to each road section based on the position of the first vehicle when the vehicle to be processed travels to each road section.

[0078] In a specific embodiment, for any first vehicle at a given moment, the position of the first vehicle when the vehicle to be processed reaches each road section = the position of the first vehicle + the distance traveled by the first vehicle when the vehicle to be processed reaches each road section. The distance traveled by the first vehicle when the vehicle to be processed reaches each road section = the speed of the first vehicle × the time required for the vehicle to be processed to reach each road section. The position of the first vehicle is obtained in real time, and the speed of the first vehicle can be the real-time speed of the first vehicle or the average speed of the first vehicle over a period of time.

[0079] In another specific embodiment, the position of the first vehicle when the vehicle travels to each road section can also be obtained based on a deep learning model or other models, and then the number of first vehicles on each road section when the vehicle travels to each road section is determined based on the position of the first vehicle when the vehicle travels to each road section.

[0080] It is understood that for any first vehicle at a certain moment, its vehicle planning path is uploaded in real time to all real-time vehicle path planning data of networked vehicles in the Internet of Vehicles. For first vehicles that have not yet traveled on all feasible driving paths of the vehicle to be processed, when the position of the first vehicle when the vehicle to be processed travels to each road section is obtained, it can be determined based on the vehicle planning path whether it has traveled on the feasible driving path of the vehicle to be processed. For first vehicles that are currently on the feasible driving path of the vehicle to be processed, it can be determined based on the vehicle planning path whether it has left the current road section, and whether it will enter another road section or leave the feasible driving path of the vehicle to be processed after leaving the current road section. In this way, the number of first vehicles on each road section when the vehicle to be processed travels to each road section can be obtained.

[0081] In one embodiment, when respectively determining the time t1 required for the vehicle to be processed to travel to the road section Ba (at the intersection B), the time t2 required for the vehicle to be processed to travel to the road section ac (at the intersection a), the time t3 required for the vehicle to be processed to travel to the road section cC (at the intersection c), the time t4 required for the vehicle to be processed to travel to the road section Cf (at the intersection C), and the time t5 required for the vehicle to be processed to travel to the road section fG (at the intersection f), only the first number of vehicles on the road section Ba is determined according to the time t1. For the road section B, The sections ac, cC, Cf and fG after a are not considered. According to time t2, only the first number of vehicles on section ac is determined. The sections cC, Cf and fG after section ac are not considered. According to time t3, only the first number of vehicles on section cC is determined. The sections Cf and fG after section cC are not considered. According to time t4, only the first number of vehicles on section Cf is determined. The section fG after section Cf is not considered. According to time t4, the first number of vehicles on section fG is determined.

[0082] Specifically, step S5 includes:

[0083] Step S52, determining the real-time first vehicle density of each road section according to the length of each road section and the real-time number of first vehicles on each road section, and determining the real-time second vehicle density of each road section according to the length of each road section and the real-time number of second vehicles on each road section.

[0084] As an implementation method, the first vehicle density = the first number of vehicles / the length of the road section, and the second vehicle density = the second number of vehicles / the length of the road section.

[0085] Specifically, step S5 includes:

[0086] Step S53, determining the expected second vehicle density, includes:

[0087] The expected second vehicle density of each road section when the vehicle to be processed travels to each road section is determined based on the first vehicle number of each road section, the real-time first vehicle density of each road section, and the real-time second vehicle density of each road section.

[0088] During implementation, the first vehicle density of each road section when the vehicles to be processed travel to the road section is determined based on the first vehicle number of each road section when the vehicles to be processed travel to the road section. The first vehicle density of each road section when the vehicles to be processed travel to the road section = the first vehicle number of each road section when the vehicles to be processed travel to the road section / the length of the road section.

[0089] Thus, the expected second vehicle density of each road section when the vehicle to be processed travels to each road section is determined based on the first vehicle density of each road section when the vehicle to be processed travels to each road section, the real-time first vehicle density of each road section, and the real-time second vehicle density of each road section.

[0090] In a specific embodiment, the expected second vehicle density of each road section when the vehicle to be processed travels to each road section = the real-time second vehicle density of each road section × the first vehicle density of each road section when the vehicle to be processed travels to each road section / the real-time first vehicle density of each road section.

[0091] It can be understood that the first vehicle is a connected vehicle in the IoV network and associated with the possible path of the vehicle to be processed. Its location and path information is available in real time and is a known, observable sample. The second vehicle includes all vehicles traveling on the possible path of the vehicle to be processed (including the first vehicle and other vehicles on the path). However, direct access to real-time data for other vehicles (such as those not connected to the network) is difficult, making them a partially unknown sample. The overall trend of the second vehicle can be indirectly inferred by the changing trend of the first vehicle (the first vehicle density at each road section when the vehicle to be processed arrives at each road section / the real-time first vehicle density). The first vehicle at each road section when the vehicle to be processed arrives at each road section is a subset of the second vehicle at that time, and both vehicles share the same traffic environment along the same route. The real-time second vehicle density is dynamically scaled by the density change ratio (predicted value / real-time value) of the connected vehicles (the first vehicle) to predict the overall vehicle density (the second vehicle density) at the time of the vehicle to be processed's arrival. The essence is to assume that the density change trend of non-networked vehicles (other vehicles) is consistent with that of networked vehicles (the first vehicle). Therefore, the density of the entire traffic flow (the second vehicle) can be approximately estimated by the density change ratio of the first vehicle.

[0092] Specifically, step S6 includes:

[0093] Step S61 , determining the congestion coefficient of each road section according to the expected second vehicle density of each road section when the vehicle to be processed travels to each road section, performing path planning according to the congestion coefficient of each road section, and determining the vehicle pre-planned path of the vehicle to be processed.

[0094] Specifically, the congestion coefficient for each road section is calculated by dividing the expected second vehicle density for each road section when the target vehicle reaches that road section by the critical density. The critical density is half the congestion density, which is the maximum density when all vehicles on the current road section are completely stopped. This density can be set based on actual scenarios or derived from historical data analysis. In this embodiment, the optimal congestion density for a single lane is 150 vehicles / km (assuming an average distance between the rear ends of two stopped vehicles is 6.67 meters), and the critical density is 75 vehicles / km. For dual and multi-lane roads, corresponding multiples can be applied.

[0095] After obtaining the congestion coefficient of each road section, the percentage of congested sections in each feasible driving path is determined according to the congestion coefficient of each road section. A congested section is defined as a section whose congestion coefficient is greater than the coefficient threshold. The congestion coefficient of each road section is compared with the coefficient threshold to determine whether each road section is a congested section and the percentage of the congested section length in the total length of the feasible driving path. The feasible driving path with the smallest percentage of the congested section length in the feasible driving path length is selected as the vehicle pre-planned path for the vehicle to be processed.

[0096] As an implementation method, if there is more than one feasible driving path with the smallest percentage of the congested road section length to the feasible driving path length, the feasible driving path with the shortest path is selected as the vehicle pre-planned path for the vehicle to be processed.

[0097] As another implementation, if there is more than one feasible driving path with the smallest percentage of the length of the congested road section to the length of the feasible driving path, the average congestion coefficient of the feasible driving paths is calculated, and the feasible driving path with the smallest average congestion coefficient is selected as the vehicle pre-planned path for the vehicle to be processed.

[0098] Specifically, step S61 can also be: determining the preference ranking of each feasible driving path according to the expected second vehicle density and preference allocation weight of each road section when the vehicle to be processed travels to each road section, and determining the vehicle pre-planned path of the vehicle to be processed according to the preference ranking of each feasible driving path.

[0099] In this embodiment, a comprehensive path scoring scheme based on multi-indicator normalization and weight allocation aims to convert indicators such as vehicle density, distance, time, number of traffic lights, toll station fees, road grade, and user preferred driving behavior (e.g., speed range) into a unified scoring system, and calculate the comprehensive score through weighted calculation, thereby selecting the optimal path as the vehicle pre-planned path for the vehicle to be processed.

[0100] The vehicle density, distance, time, number of traffic lights, toll booth fees, road grade and other indicators are converted into a score of 0 to 100. The lower the score, the better the indicator. The details are as follows: the scores are divided according to vehicle density and vehicle density interval. The vehicle density interval is [0-congestion density]. If the vehicle density is at 1 / 4 of the length of the vehicle density interval, the score is 25 points. The scores of each road section are counted and the average score is calculated as the vehicle density conversion score; the conversion score is determined according to the length / distance of the feasible driving path. The distance conversion score = (current path distance - shortest path distance) / (longest path distance - shortest path distance) × 100; the conversion score is determined according to the travel time of the feasible path. The time conversion score = (current path estimated travel time - shortest path estimated travel time) The system uses the following formulas: estimated driving time (estimated driving time of the longest route - estimated driving time of the shortest route) × 100; the system uses segmented scoring based on the number of traffic lights: 0 (0 points), 1-3 (20 points), 4-6 (40 points), 7-10 (60 points), 10-15 (80 points), and more than 15 (100 points); the system uses linear scoring based on toll station fees: fee conversion score = current route fee / highest fee among all routes × 100; the system uses road grade scoring: road grades include expressways, national highways, provincial highways, and small roads: road grade conversion score = expressway length / total driving route length × 20 + national highway length / total driving route length × 40 + provincial highway length / total driving route length × 60 + small road length / total current route length × 100.

[0101] After the indicators are converted into scores, the weights are assigned according to the current preferences defined by the user. In one embodiment, the weight of the vehicle density indicator is 30%, the weight of the time indicator is 25%, the weight of the distance indicator is 15%, the weight of the number of traffic lights indicator is 10%, the weight of the toll station fee indicator is 10%, and the weight of the road grade indicator is 10%. .

[0102] It is understandable that the preference allocation weight is set by the user according to his or her own behavioral preferences, and will not be elaborated here.

[0103] Specifically, in step S6, determining that the vehicle to be processed executes the pre-planned vehicle path or re-acquires the vehicle data of the first vehicle based on the third vehicle data includes:

[0104] If the number of the third vehicles is less than a preset threshold, the vehicle to be processed executes the vehicle pre-planned path;

[0105] If the number of the third vehicles is greater than or equal to a preset threshold, the vehicle data of the first vehicle is reacquired.

[0106] It is understandable that during the period between when the vehicle to be processed acquires IoV data and when its pre-planned route is determined, several other networked vehicles (i.e., third vehicles) may also be performing route planning and completing the pre-planned route. If the number of third vehicles is large, this may lead to the Brace Paradox (when more than a certain number of vehicles simultaneously select a single "optimal route," local optimization may occur, leading to increased global congestion). Therefore, the present invention acquires other networked vehicles that re-planned routes between the time the IoV data was acquired and the current time, denoted as third vehicles. The number of third vehicles is compared with a preset threshold. If the number of third vehicles is less than the preset threshold, indicating a low likelihood of the Brace Paradox, the vehicle to be processed executes the pre-planned route. If the number of third vehicles is greater than or equal to the preset threshold, indicating a high likelihood of the Brace Paradox, the vehicle data of the first vehicle is re-acquired and route planning is re-performed.

[0107] As an embodiment, the preset threshold can be a certain value, for example, the preset threshold is 100. When the number of third vehicles is less than 100, the vehicle to be processed executes the vehicle pre-planned path. When the number of third vehicles is greater than or equal to 100, the vehicle data of the first vehicle is re-acquired.

[0108] Specifically, step S6 includes:

[0109] Step S62, when executing the vehicle pre-planned path, the vehicle planned path executed by the vehicle to be processed is uploaded to all real-time vehicle path planning data of the networked vehicles in the vehicle network.

[0110] Specifically, it also includes: Step S54: determining a correction coefficient based on the real-time second vehicle density and the expected second vehicle density of each road section on which the vehicle pre-planned path of the vehicle to be processed has traveled, and correcting the expected second vehicle density of each road section when the vehicle to be processed travels to each road section based on the correction coefficient.

[0111] Specifically, the real-time detection of vehicles on each road section includes: setting detection sensors for detecting vehicles at the exit and entrance of each road section, and determining the number of second vehicles on each road section based on data detected by the detection sensors at the exit and entrance of each road section.

[0112] See also Figure 3Taking path ABCEFG as an example, there are exits B, a, c, C, and f on path ABCEFG, as well as entrances b, d, and e. Vehicle detection sensors are installed at exits B, a, c, C, f, b, d, and e. Whenever a vehicle passes, enters, or exits, it can be detected, identified, and counted by the vehicle detection sensors, and the detection signal or counting data can be uploaded to the cloud platform in real time for the vehicle terminal to obtain.

[0113] Specifically, the correction coefficient = (the real-time second vehicle density of the historical vehicles to be processed traveling to a certain road section - the average value of the expected second vehicle density when the vehicles to be processed travel to a certain road section) / the average value of the expected second vehicle density when the vehicles to be processed travel to a certain road section.

[0114] The corrected expected second vehicle density of each road section = the expected second vehicle density of each road section × (1-correction coefficient).

[0115] In one embodiment, see Figure 3 Taking path ABCEFG as an example, the vehicle to be processed is traveling on road section ac. The expected second vehicle density of road section cC when the vehicle to be processed arrives at road section cC is determined based on the first number of vehicles on road section cC, the real-time first vehicle density of road section cC, and the real-time second vehicle density of road section cC. A correction coefficient is determined based on the historical real-time second vehicle density of road section Ba, which the vehicle to be processed has already traveled on its planned vehicle path, and the expected second vehicle density when the vehicle to be processed arrives at road section Ba. That is, correction coefficient = (( ...

[0116] The present invention determines a correction coefficient for the real-time second vehicle density and expected second vehicle density of each historical road section that has been traveled according to the vehicle planning path of the vehicle to be processed, and corrects the expected second vehicle density of each road section when the vehicle to be processed travels to each road section next according to the correction coefficient. Based on the actual second vehicle density data of the historical road sections, the inherent deviation of the prediction model is corrected. The correction coefficient represents the distribution pattern of non-networked vehicles in history, such as the gathering of commuting vehicles during peak holiday periods, the impact of the same time period and similar weather on traffic flow, etc., so that the prediction is closer to the actual full traffic flow.

[0117] The path planning method of the present invention is real-time, and takes into account the real-time vehicle planning paths of networked vehicles in the Internet of Vehicles, the vehicles related to all feasible driving paths of the vehicle to be processed (first vehicles), and the real-time vehicles on each section of the all feasible driving paths of the vehicle to be processed (second vehicles). As the vehicle data related to the first vehicle and the second vehicle are collected, changed, and predicted in real time, the vehicle planning path of the vehicle to be processed is also dynamically adjusted. The subsequent path segments can be replanned in combination with the latest vehicle density and accident data (if an accident occurs in the road segment ahead, the first vehicle data and the second vehicle data will change). At the segmentation node, the optimal path of the subsequent sub-segment is recalculated and planned based on the latest data. If a sub-segment is predicted to be congested, an alternative path is switched in advance (such as detouring around a parallel highway or national highway).

[0118] The path planning of the vehicle to be processed of the present invention takes into account the vehicles corresponding to the real-time vehicle path planning related to all feasible driving paths of the networked vehicles and the vehicle to be processed in all the vehicle networks (that is, the first vehicle includes vehicles that have not yet traveled to all feasible driving paths of the vehicle to be processed and vehicles that are traveling on the feasible driving paths of the vehicle to be processed). This can avoid the Brays paradox caused by the existing technology that only performs real-time traffic flow data on feasible paths to perform real-time road condition analysis and then realizes real-time path planning (when more than a certain number of vehicles simultaneously select a certain "optimal path", local optimization will lead to increased global congestion).

[0119] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

[0120] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that the present invention is susceptible to various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A multi-vehicle collaborative path planning method based on the Internet of Vehicles, characterized by: include: In response to preset information to be processed, determining several feasible driving paths for the vehicle according to the starting point and the end point of the vehicle to be processed, and segmenting each feasible driving path; Retrieving all real-time vehicle path planning data of other connected vehicles in the vehicle network in real time, obtaining vehicles corresponding to the real-time vehicle path planning associated with any road segment of each feasible driving path from all real-time vehicle path planning data, denoting them as first vehicles, and obtaining vehicle data of each first vehicle in real time, including the number of first vehicles, the position of each first vehicle, and the speed of each first vehicle, wherein the first vehicles include vehicles on each feasible driving path that have not yet traveled to the vehicle to be processed; Detection sensors for detecting vehicles are set up at the exit and entrance of each road section, and vehicles on each road section are detected in real time, recorded as second vehicles, and the number of second vehicles on each road section is obtained in real time, where the second vehicles include first vehicles traveling on each road section on all feasible driving paths of the vehicles to be processed and vehicles that are not connected to the vehicle network; Obtain the real-time speed of the vehicle to be processed, and determine the time it takes for the vehicle to be processed to travel to each road section based on the real-time speed of the vehicle to be processed; Determine the position of the first vehicle when the vehicle to be processed travels to each road section based on the time when the vehicle to be processed travels to each road section, the speed data of the first vehicle, and the position data of the first vehicle; and determine the number of first vehicles corresponding to each road section when the vehicle to be processed travels to each road section based on the position of the first vehicle when the vehicle to be processed travels to each road section; Determine the real-time first vehicle density of each road section according to the length of each road section and the real-time number of first vehicles on each road section, and determine the real-time second vehicle density of each road section according to the length of each road section and the real-time number of second vehicles on each road section; Determine the expected second vehicle density of each road section when the vehicle to be processed travels to each road section based on the first vehicle quantity of each road section, the real-time first vehicle density of each road section, and the real-time second vehicle density of each road section. The expected second vehicle density is a real-time prediction of the total vehicle density of the vehicle to be processed when it arrives at each road section. Performing path planning based on the expected second vehicle density of each road section when the vehicle to be processed travels to each road section, and determining a pre-planned vehicle path for the vehicle to be processed; Other networked vehicles that re-planned their routes between the time when the networked vehicle data was acquired and the current time are recorded as third vehicles. Based on the comparison of the third vehicle data with a preset threshold, it is determined whether the vehicle to be processed executes the vehicle pre-planned route or re-acquires the vehicle data of the first vehicle. The third vehicle is a vehicle that, during the period between the time when the vehicle to be processed acquires the networked vehicle data and the time when the vehicle pre-planned route is determined, has several other networked vehicles that perform route planning and complete the vehicle pre-planned route. When executing the vehicle pre-planned path, the vehicle planned path executed by the vehicle to be processed is uploaded to all real-time vehicle path planning data of networked vehicles in the vehicle network.

2. The multi-vehicle collaborative path planning method based on the Internet of Vehicles according to claim 1 is characterized in that: Determine the time it takes for vehicles to be processed to travel to each road section, including: The length of each road section and the real-time position of the vehicle to be processed are obtained, and the time it takes for the vehicle to be processed to travel to each road section is determined based on the real-time speed of the vehicle to be processed, the real-time position of the vehicle to be processed and the length of each road section.

3. The multi-vehicle collaborative path planning method based on the Internet of Vehicles according to claim 1 is characterized in that: The congestion coefficient of each feasible driving path is determined according to the expected second vehicle density of each road section when the vehicle to be processed travels to each road section, and the path planning is performed according to the congestion coefficient of each feasible driving path to determine the vehicle pre-planned path of the vehicle to be processed.

4. The multi-vehicle collaborative path planning method based on the Internet of Vehicles according to claim 1 is characterized in that: Also includes: The preference ranking of each feasible driving path is determined according to the expected second vehicle density of each road section when the vehicle to be processed travels to each road section and the preference allocation weight, and the vehicle pre-planned path of the vehicle to be processed is determined according to the preference ranking of each feasible driving path.

5. The multi-vehicle collaborative path planning method based on the Internet of Vehicles according to claim 4 is characterized in that: The segmenting of each feasible driving path includes: Count all exits on each feasible driving path of the vehicle to be processed; Each feasible driving path is segmented according to all exits on the feasible driving path, and the area between two adjacent exits on the same feasible driving path is a road segment.

6. The multi-vehicle collaborative path planning method based on the Internet of Vehicles according to claim 5 is characterized in that: Also includes: A correction coefficient is determined based on the real-time second vehicle density and expected second vehicle density of each road section on which the vehicle to be processed has traveled on the pre-planned path of the vehicle to be processed, and the expected second vehicle density of each road section when the vehicle to be processed travels to each road section is corrected based on the correction coefficient.

7. The multi-vehicle collaborative path planning method based on the Internet of Vehicles according to claim 6 is characterized in that: The determining, based on the third vehicle data, that the vehicle to be processed executes the pre-planned vehicle path or reacquiring the vehicle data of the first vehicle includes: If the number of the third vehicles is less than a preset threshold, the vehicle to be processed executes the vehicle pre-planned path; If the number of the third vehicles is greater than or equal to a preset threshold, the vehicle data of the first vehicle is reacquired.

Images