A path planning method and device

Abstract

The application discloses a path planning method and device, through automatic driving vehicle identification on the shooting picture of the intersection to be monitored, and obtaining the path planning of each vehicle from the database, combining the plan of the intersection to be monitored, generating the road point planning graph of the intersection to be monitored, and presenting the road point of each automatic driving vehicle to the user. Compared with the prior art which cannot realize path planning of multiple automatic driving vehicles, the application can realize path planning of multiple vehicles, and the output road point planning graph can provide effective data support for subsequent automatic driving strategy, road dredging and planning research.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a path planning method and apparatus. Background Technology

[0002] With the development of science and communication technologies, autonomous driving is gradually entering people's lives. Current autonomous driving systems are controlled by the vehicle's built-in autonomous driving system, which is responsible for controlling the vehicle's internal hardware and software. For example, the autonomous driving system controls hardware such as the vehicle's engine, battery, and air conditioning system, or it controls in-vehicle software such as navigation software. Regarding path planning, for the sake of user experience and safety, the vehicle generally plans the route itself, with the autonomous driving system planning the driving path based on its own location, destination, and driving strategy.

[0003] However, existing path planning schemes are not suitable for certain control scenarios, such as when the same server or terminal device plans and displays paths for multiple vehicles to assess the overall rationality of the autonomous driving strategy. Current technology cannot achieve unified path planning for multiple autonomous vehicles, and there is a lack of relevant solutions in this regard. Summary of the Invention

[0004] This invention provides a path planning method and apparatus that can perform path planning for multiple vehicles, and the output waypoint planning map can provide effective data support for subsequent research.

[0005] An embodiment of the present invention provides a path planning method, comprising:

[0006] The system acquires real-time images of the first intersection to be monitored and identifies all autonomous vehicles in the images of the first intersection, obtaining the first vehicle information corresponding to each autonomous vehicle.

[0007] Based on the first vehicle information corresponding to each autonomous vehicle, the database is queried to obtain the first path plan of each autonomous vehicle when passing through the intersection to be monitored; wherein, each path plan includes the position information and speed information of each autonomous vehicle at each moment.

[0008] Based on each first path plan and the intersection plan corresponding to the intersection to be monitored, a first waypoint planning map of the intersection to be monitored is obtained; wherein, the first waypoint planning map includes the path planning trajectory of each autonomous vehicle, and each path planning trajectory is presented in the form of waypoints on the waypoint planning map.

[0009] Furthermore, after obtaining the waypoint planning map of the intersection to be monitored, the process also includes:

[0010] When the next monitoring cycle arrives, the second intersection image of the intersection to be monitored is acquired in real time, and all autonomous vehicles in the first intersection image are identified to determine whether the autonomous vehicles on the intersection to be monitored have changed.

[0011] If not, the process ends and the system waits for the next monitoring cycle.

[0012] If so, then obtain the second vehicle information corresponding to each second autonomous vehicle in the second intersection image;

[0013] Based on the information of each of the second vehicles, the database is queried to obtain the second path plan for each of the second autonomous vehicles when passing through the intersection to be monitored.

[0014] Based on each second path plan and the first waypoint plan, the second waypoint plan of the intersection to be monitored is obtained.

[0015] Furthermore, obtaining the second waypoint planning map of the intersection to be monitored based on each second path plan and the first waypoint planning map specifically involves:

[0016] Sequentially determine whether the second autonomous vehicle has already appeared in the first waypoint planning map;

[0017] If it has already occurred, compare the first path plan and the second path plan to obtain the speed comparison information of the two path plans at each time step, and generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map based on the second path plan and the speed comparison information.

[0018] If it does not appear, then according to the second path planning, the path planning trajectory of the second autonomous vehicle on the second waypoint planning map is generated.

[0019] Furthermore, the step of identifying all autonomous vehicles in the image captured at the first intersection and obtaining the first vehicle information corresponding to each autonomous vehicle specifically involves:

[0020] The relative position information of each autonomous vehicle in the image taken at the first intersection is obtained, and the initial geographical location information of each autonomous vehicle is obtained based on the actual geographical information of the intersection to be monitored and the relative position information.

[0021] Extract image information from the pictures taken by each autonomous vehicle at the first intersection, and obtain the unique vehicle code of each autonomous vehicle according to the preset recognition algorithm;

[0022] The initial geographical location information and unique vehicle code of each autonomous vehicle are used as the corresponding first vehicle information.

[0023] Furthermore, the step of querying the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan for each autonomous vehicle when passing through the monitored intersection specifically involves:

[0024] Based on the vehicle's unique code, the database is queried to obtain the third path plan corresponding to each vehicle's unique code; wherein, the third path plan is uploaded and stored in the database by each autonomous vehicle.

[0025] Based on the initial geographical location information, a path planning trajectory of a preset length is extracted from the third path planning, and based on each path planning trajectory, a first path planning is generated for each autonomous vehicle when passing through the intersection to be monitored.

[0026] Accordingly, another embodiment of the present invention provides a path planning device, including: a first acquisition module, a first path planning module, and a waypoint planning map generation module;

[0027] The first acquisition module is used to acquire the first intersection image of the intersection to be monitored in real time, and to identify all autonomous vehicles in the first intersection image to acquire the first vehicle information corresponding to each autonomous vehicle.

[0028] The first path planning module is used to query the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan of each autonomous vehicle when passing through the intersection to be monitored; wherein, each path plan includes the position information and speed information of each autonomous vehicle at each moment.

[0029] The waypoint planning map generation module is used to obtain the first waypoint planning map of the intersection to be monitored based on each first path plan and the intersection plan corresponding to the intersection to be monitored; wherein, the first waypoint planning map includes the path planning trajectory of each autonomous vehicle, and each path planning trajectory is presented in the form of waypoints on the waypoint planning map.

[0030] Furthermore, the path planning device also includes: a second acquisition module, a judgment module, a waiting module, a third acquisition module, and a second path planning module;

[0031] The second acquisition module is used to acquire the second intersection image of the intersection to be monitored in real time when the next monitoring cycle arrives.

[0032] The judgment module is used to identify all autonomous vehicles in the image captured at the first intersection and to determine whether any autonomous vehicles at the monitored intersection have changed.

[0033] The waiting module is used to wait for the next monitoring cycle when the judgment module determines that the judgment result is negative;

[0034] The third acquisition module is used to acquire the second vehicle information corresponding to each second autonomous driving vehicle in the second intersection image when the judgment module determines that the judgment result is yes.

[0035] The second path planning module is used to query the database based on the information of each second vehicle to obtain the second path plan for each second autonomous vehicle when passing through the intersection to be monitored.

[0036] The waypoint planning map generation module is also used to obtain the second waypoint planning map of the intersection to be monitored based on each second path plan and the first waypoint planning map.

[0037] Furthermore, the waypoint planning map generation module includes: a judgment unit, a first planning unit, and a second planning unit;

[0038] The judgment unit is used to sequentially determine whether the second autonomous vehicle has appeared in the first waypoint planning map;

[0039] The first planning unit is used to compare the first path plan and the second path plan when the judgment unit determines that the judgment result has occurred, to obtain the speed comparison information of the two path plans at each time, and to generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map based on the second path plan and the speed comparison information.

[0040] The second planning unit is used to generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map according to the second path planning when the judgment unit determines that the judgment result is not present.

[0041] Furthermore, the first acquisition module is used to identify all autonomous vehicles in the image captured at the first intersection and obtain the first vehicle information corresponding to each autonomous vehicle, specifically:

[0042] The relative position information of each autonomous vehicle in the image taken at the first intersection is obtained, and the initial geographical location information of each autonomous vehicle is obtained based on the actual geographical information of the intersection to be monitored and the relative position information.

[0043] Extract image information from the pictures taken by each autonomous vehicle at the first intersection, and obtain the unique vehicle code of each autonomous vehicle according to the preset recognition algorithm;

[0044] The initial geographical location information and unique vehicle code of each autonomous vehicle are used as the corresponding first vehicle information.

[0045] Furthermore, the first path planning module is used to query the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan for each autonomous vehicle when passing through the monitored intersection, specifically:

[0046] Based on the vehicle's unique code, the database is queried to obtain the third path plan corresponding to each vehicle's unique code; wherein, the third path plan is uploaded and stored in the database by each autonomous vehicle.

[0047] Based on the initial geographical location information, a path planning trajectory of a preset length is extracted from the third path planning, and based on each path planning trajectory, a first path planning is generated for each autonomous vehicle when passing through the intersection to be monitored.

[0048] As can be seen from the above, this invention provides a path planning method and apparatus. It identifies autonomous vehicles by capturing images of intersections to be monitored, retrieves the path plans for each vehicle from a database, and combines this with a plan view of the intersection to generate a waypoint planning map, presenting the waypoints of each autonomous vehicle to the user. Compared to existing technologies that cannot perform path planning for multiple autonomous vehicles, this invention can perform path planning for multiple vehicles, and the output waypoint planning map can provide effective data support for subsequent research on autonomous driving strategies, road management, and planning.

[0049] Furthermore, after obtaining the first waypoint planning map, the present invention can continue to obtain the second waypoint planning map based on the real-time situation and present the speed changes of each autonomous vehicle to the user, such as whether the vehicles that have appeared have decelerated or accelerated, and the waypoint planning of vehicles that have not appeared, making the monitoring situation of the intersection to be monitored more intuitive and improving the user experience.

[0050] Furthermore, the waypoint planning map of the present invention can also be used for subsequent autonomous vehicle control. For example, based on the waypoint planning map, the path planning of each autonomous vehicle at the intersection to be monitored can be replanned, and corresponding control commands can be generated and sent to the vehicles at the intersection to achieve one-to-many remote control. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating an embodiment of the path planning method provided by the present invention;

[0052] Figure 2 This is a flowchart illustrating another embodiment of the path planning method provided by the present invention;

[0053] Figure 3-5 This is a flowchart illustrating a specific implementation of the path planning method provided by the present invention;

[0054] Figure 6 This is a schematic diagram of an embodiment of the path planning device provided by the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] See Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the path planning method provided by the present invention. Figure 1 As shown, the specific steps are as follows:

[0057] Step 101: Acquire the first intersection image of the intersection to be monitored in real time, identify all autonomous vehicles in the first intersection image, and obtain the first vehicle information corresponding to each autonomous vehicle.

[0058] In this embodiment, images of the first intersection to be monitored can be obtained, but are not limited to, through surveillance cameras or other monitoring devices.

[0059] In this embodiment, step 101 specifically involves: obtaining the relative position information of each autonomous vehicle in the image taken at the first intersection, and obtaining the initial geographical location information of each autonomous vehicle based on the actual geographical information of the intersection to be monitored and the relative position information; extracting the image information of each autonomous vehicle in the image taken at the first intersection, and obtaining the unique vehicle code of each autonomous vehicle based on a preset recognition algorithm; and using the initial geographical location information and unique vehicle code of each autonomous vehicle as the corresponding first vehicle information.

[0060] In this embodiment, the image captured at the first intersection will display the relative position information (relative position of the vehicle at the intersection) and image information of the autonomous vehicles. By combining these two pieces of information with the actual geographical location of the intersection and the preset recognition algorithm, the initial geographical location information and unique vehicle code (used to identify which autonomous vehicle) of each vehicle can be obtained at the time the image was captured.

[0061] In this embodiment, the recognition algorithm is existing technology, and how to identify vehicles through image content is also existing technology, which will not be described in detail here.

[0062] Step 102: Based on the first vehicle information corresponding to each autonomous vehicle, query the database to obtain the first path plan of each autonomous vehicle when passing through the intersection to be monitored; wherein, each path plan includes the position information and speed information of each autonomous vehicle at each moment.

[0063] In this embodiment, step 102 specifically involves: querying the database based on the vehicle's unique code to obtain the third path plan corresponding to each vehicle's unique code; wherein the third path plan is uploaded and stored in the database by each autonomous vehicle; extracting a path planning trajectory of a preset length from the third path plan based on the initial geographical location information, and generating a first path plan for each autonomous vehicle when passing through the intersection to be monitored based on each path planning trajectory.

[0064] In this embodiment, path plans uploaded by each autonomous vehicle are collected through a preset database. Since path plans are generally generated when the vehicle's destination is determined, these path plans can be generated by the execution entity of this invention or by the autonomous driving system on the vehicle.

[0065] In this embodiment, the corresponding autonomous vehicle is first located based on its unique vehicle code. Then, the database is queried to obtain the corresponding path specifications. The path plan invoked is the complete plan from the origin to the destination. However, this invention only needs the part of the path plan that passes through the intersection to be monitored. Therefore, based on the initial geographical location information, a preset length is extracted from the third path plan. The preset length can be set according to the actual situation of the intersection, such as the intersection length, intersection width, or custom settings.

[0066] Step 103: Based on each first path plan and the intersection plan corresponding to the intersection to be monitored, obtain the first waypoint planning map of the intersection to be monitored; wherein, the first waypoint planning map includes the path planning trajectory of each autonomous vehicle, and each path planning trajectory is presented in the form of waypoints on the waypoint planning map.

[0067] In this embodiment, after obtaining each first path plan, a corresponding intersection plan is constructed based on the geographical information of the intersection to be monitored, to present users with more intuitive road and driving information. Based on the first path plans, the path planning trajectories of each autonomous vehicle are extracted, and the trajectories are divided into several waypoints at preset intervals. Then, based on the location information of each waypoint, corresponding speed information is configured for each waypoint, generating the waypoint planning trajectory for each autonomous vehicle. Finally, vehicle identifiers representing each autonomous vehicle are placed at the initial geographical location information corresponding to each vehicle, generating the first path plan for each autonomous vehicle when passing through the intersection to be monitored at the current monitoring time.

[0068] In this embodiment, the preset interval can be 1 second or 500 milliseconds.

[0069] As an example of this embodiment, see Figure 2 , Figure 2 This is a schematic diagram of another embodiment of the path planning method provided by the present invention. Figure 2 and Figure 1 The difference lies in the fact that, after step 103, steps 204 to 208 are also included, and the specific details of each step are as follows:

[0070] Step 204: When the next monitoring cycle arrives, acquire the real-time image of the second intersection of the monitored intersection, and identify all autonomous vehicles in the image of the first intersection to determine whether there have been any changes in the autonomous vehicles at the monitored intersection. If there have been any changes, proceed to step 205; otherwise, proceed to step 208.

[0071] In this embodiment, the monitoring period can be set to 1 second or 500 milliseconds. Whether a vehicle has changed can be determined based on vehicle data or vehicle identification, such as whether the number of vehicles has increased or decreased, whether the vehicle's unique code is the same as the original vehicle's unique code, and whether an increase or decrease has occurred. This step is mainly used for the equipment to monitor whether there has been an increase or decrease in the number of vehicles at the intersection.

[0072] Step 205: Obtain the second vehicle information corresponding to each second autonomous driving vehicle in the image captured at the second intersection.

[0073] The principle behind this step is the same as that of step 101, and will not be repeated here.

[0074] Step 206: Based on the information of each second vehicle, query the database to obtain the second path plan for each second autonomous vehicle when passing through the intersection to be monitored.

[0075] The principle behind this step is the same as that of step 102, and will not be repeated here.

[0076] Step 207: Obtain the second waypoint planning map of the intersection to be monitored based on each second path planning map and the first waypoint planning map.

[0077] In this embodiment, step 207 specifically includes:

[0078] Sequentially determine whether the second autonomous vehicle has already appeared in the first waypoint planning map;

[0079] If it has already occurred, compare the first path plan and the second path plan to obtain the speed comparison information of the two path plans at each time step, and generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map based on the second path plan and the speed comparison information.

[0080] If it does not appear, then according to the second path planning, the path planning trajectory of the second autonomous vehicle on the second waypoint planning map is generated.

[0081] In this example, the system first determines whether the second autonomous vehicle has already appeared in the first waypoint planning map. If so, the waypoint planning map needs to display the vehicle's speed comparison information. If not, the corresponding path planning trajectory is directly generated according to the second path planning. For example, if there are 4 vehicles passing through the intersection to be monitored in the first waypoint planning map, and a new vehicle is found during the second detection, the speed change information of the first 4 vehicles will be displayed in the second path planning map, such as vehicle 1 accelerating, vehicle 2 decelerating, vehicle 3 decelerating, and vehicle 4 maintaining its original speed. The newly added vehicle will only display its waypoint planning trajectory.

[0082] Step 208: End, wait for the next monitoring cycle.

[0083] In this embodiment, both the first waypoint planning map and the second waypoint planning map can be used for subsequent vehicle control. For example, by dragging an autonomous vehicle on the first waypoint planning map to move it from one waypoint to another, corresponding control commands can be generated based on speed and location information and sent to the autonomous vehicle so that it reaches the waypoint before the next monitoring cycle, achieving remote control. Simultaneously, the next monitoring cycle can also intuitively reflect the specific situation of this control, whether the vehicle actually reached the waypoint, and whether there were any deviations, providing more data support for subsequent one-to-many control.

[0084] See Figure 3-5 , Figure 3-5 This is a flowchart illustrating a specific implementation of the path planning method provided by the present invention. The specific implementation process is as follows: Figure 3 As shown, at a certain time t, there are four cars at the intersection. By searching the database, the path planning for the four cars to pass through the intersection is directly provided, displayed in the form of waypoints (waypoints refer to: at what time a vehicle arrives at which location); for example... Figure 4 As shown, after, say, 1 second, at time t+1 second, the existing four vehicles have traveled a certain distance along the planned path. At this moment, the intersection camera detects another new vehicle entering its field of view. Figure 5As shown, based on the positions of the five vehicles at time t+1s, the database is searched again, and another collective planning is performed. The new planning result may differ from the first planning, but this difference is reflected in the speed, not the route. The difference in speed will cause changes in the waypoints of the four vehicles—vehicle 1 accelerates, vehicle 2 decelerates, vehicle 3 decelerates, and vehicle 4 maintains its original speed. The intersection "monitoring-database query-planning distribution" process can be set to occur once per second. The three figures above describe a complete cycle and are representative.

[0085] Accordingly, see Figure 6 , Figure 6 This is a schematic diagram of one embodiment of the path planning device provided by the present invention, as shown below. Figure 6 As shown, the device includes: a first acquisition module 301, a first path planning module 302, and a waypoint planning map generation module 303.

[0086] The first acquisition module 301 is used to acquire the first intersection image of the intersection to be monitored in real time, and to identify all autonomous vehicles in the first intersection image to acquire the first vehicle information corresponding to each autonomous vehicle.

[0087] The first path planning module 302 is used to query the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan of each autonomous vehicle when passing through the intersection to be monitored; wherein, each path plan includes the position information and speed information of each autonomous vehicle at each moment.

[0088] The waypoint planning map generation module 303 is used to obtain the first waypoint planning map of the intersection to be monitored based on each first path plan and the intersection plan corresponding to the intersection to be monitored; wherein, the first waypoint planning map includes the path planning trajectory of each autonomous vehicle, and each path planning trajectory is presented in the form of waypoints on the waypoint planning map.

[0089] As an example of this embodiment, the path planning device further includes: a second acquisition module, a judgment module, a waiting module, a third acquisition module, and a second path planning module;

[0090] The second acquisition module is used to acquire the second intersection image of the intersection to be monitored in real time when the next monitoring cycle arrives.

[0091] The judgment module is used to identify all autonomous vehicles in the image captured at the first intersection and to determine whether any autonomous vehicles at the monitored intersection have changed.

[0092] The waiting module is used to wait for the next monitoring cycle when the judgment module determines that the judgment result is negative;

[0093] The third acquisition module is used to acquire the second vehicle information corresponding to each second autonomous driving vehicle in the second intersection image when the judgment module determines that the judgment result is yes.

[0094] The second path planning module is used to query the database based on the information of each second vehicle to obtain the second path plan for each second autonomous vehicle when passing through the intersection to be monitored.

[0095] The waypoint planning map generation module is also used to obtain the second waypoint planning map of the intersection to be monitored based on each second path plan and the first waypoint planning map.

[0096] In this embodiment, the waypoint planning map generation module includes: a judgment unit, a first planning unit, and a second planning unit;

[0097] The judgment unit is used to sequentially determine whether the second autonomous vehicle has appeared in the first waypoint planning map;

[0098] The first planning unit is used to compare the first path plan and the second path plan when the judgment unit determines that the judgment result has occurred, to obtain the speed comparison information of the two path plans at each time, and to generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map based on the second path plan and the speed comparison information.

[0099] The second planning unit is used to generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map according to the second path planning when the judgment unit determines that the judgment result is not present.

[0100] In this embodiment, the first acquisition module is used to identify all autonomous vehicles in the image captured at the first intersection and obtain the first vehicle information corresponding to each autonomous vehicle, specifically:

[0101] The relative position information of each autonomous vehicle in the image taken at the first intersection is obtained, and the initial geographical location information of each autonomous vehicle is obtained based on the actual geographical information of the intersection to be monitored and the relative position information.

[0102] Extract image information from the pictures taken by each autonomous vehicle at the first intersection, and obtain the unique vehicle code of each autonomous vehicle according to the preset recognition algorithm;

[0103] The initial geographical location information and unique vehicle code of each autonomous vehicle are used as the corresponding first vehicle information.

[0104] In this embodiment, the first path planning module is used to query the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan for each autonomous vehicle when passing through the intersection to be monitored, specifically:

[0105] Based on the vehicle's unique code, the database is queried to obtain the third path plan corresponding to each vehicle's unique code; wherein, the third path plan is uploaded and stored in the database by each autonomous vehicle.

[0106] Based on the initial geographical location information, a path planning trajectory of a preset length is extracted from the third path planning, and based on each path planning trajectory, a first path planning is generated for each autonomous vehicle when passing through the intersection to be monitored.

[0107] As can be seen from the above, this invention provides a path planning method and apparatus. It identifies autonomous vehicles by capturing images of intersections to be monitored, retrieves the path plans for each vehicle from a database, and combines this with a plan view of the intersection to generate a waypoint planning map, presenting the waypoints of each autonomous vehicle to the user. Compared to existing technologies that cannot perform path planning for multiple autonomous vehicles, this invention can perform path planning for multiple vehicles, and the output waypoint planning map can provide effective data support for subsequent research on autonomous driving strategies, road management, and planning.

[0108] Furthermore, after obtaining the first waypoint planning map, the present invention can continue to obtain the second waypoint planning map based on the real-time situation and present the speed changes of each autonomous vehicle to the user, such as whether the vehicles that have appeared have decelerated or accelerated, and the waypoint planning of vehicles that have not appeared, making the monitoring situation of the intersection to be monitored more intuitive and improving the user experience.

[0109] Furthermore, the waypoint planning map of the present invention can also be used for subsequent autonomous vehicle control. For example, based on the waypoint planning map, the path planning of each autonomous vehicle at the intersection to be monitored can be replanned, and corresponding control commands can be generated and sent to the vehicles at the intersection to achieve one-to-many remote control.

[0110] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0111] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A path planning method, characterized in that, include: The system acquires real-time images of the first intersection to be monitored and identifies all autonomous vehicles in the images of the first intersection, obtaining the first vehicle information corresponding to each autonomous vehicle. Based on the first vehicle information corresponding to each autonomous vehicle, the database is queried to obtain the first path plan of each autonomous vehicle when passing through the intersection to be monitored; wherein, each path plan includes the position information and speed information of each autonomous vehicle at each moment. Based on each first path plan and the intersection plan corresponding to the intersection to be monitored, a first waypoint planning map of the intersection to be monitored is obtained; wherein, the first waypoint planning map includes the path planning trajectory of each autonomous vehicle, and each path planning trajectory is presented in the form of waypoints on the first waypoint planning map; After obtaining the first waypoint planning map of the intersection to be monitored, the method further includes: When the next monitoring cycle arrives, the second intersection image of the intersection to be monitored is acquired in real time, and all autonomous vehicles in the first intersection image are identified to determine whether the autonomous vehicles on the intersection to be monitored have changed. If not, the process ends and the system waits for the next monitoring cycle. If so, then obtain the second vehicle information corresponding to each second autonomous vehicle in the second intersection image; Based on the information of each second vehicle, the database is queried to obtain the second path plan of each second autonomous vehicle when passing through the intersection to be monitored. Based on each second path plan and the first waypoint plan map, the second waypoint plan map of the intersection to be monitored is obtained; The step of obtaining the second waypoint planning map of the intersection to be monitored based on each second path plan and the first waypoint planning map specifically involves: Sequentially determine whether the second autonomous vehicle has already appeared in the first waypoint planning map; If it has already occurred, compare the first path plan and the second path plan to obtain the speed comparison information of the two path plans at each time step, and generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map based on the second path plan and the speed comparison information. If it does not appear, then according to the second path planning, the path planning trajectory of the second autonomous vehicle on the second waypoint planning map is generated.

2. The path planning method according to claim 1, characterized in that, The step of identifying all autonomous vehicles in the image captured at the first intersection and obtaining the first vehicle information corresponding to each autonomous vehicle specifically involves: The relative position information of each autonomous vehicle in the image taken at the first intersection is obtained, and the initial geographical location information of each autonomous vehicle is obtained based on the actual geographical information of the intersection to be monitored and the relative position information. Extract image information from the pictures taken by each autonomous vehicle at the first intersection, and obtain the unique vehicle code of each autonomous vehicle according to the preset recognition algorithm; The initial geographical location information and unique vehicle code of each autonomous vehicle are used as the corresponding first vehicle information.

3. The path planning method according to claim 2, characterized in that, The step of querying the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan for each autonomous vehicle when passing through the monitored intersection is specifically as follows: Based on the vehicle's unique code, the database is queried to obtain the third path plan corresponding to each vehicle's unique code; wherein, the third path plan is uploaded and stored in the database by each autonomous vehicle. Based on the initial geographical location information, a path planning trajectory of a preset length is extracted from the third path planning, and based on each path planning trajectory, a first path planning is generated for each autonomous vehicle when passing through the intersection to be monitored.

4. A path planning device, characterized in that, include: The module consists of a first acquisition module, a first path planning module, and a waypoint planning map generation module. The first acquisition module is used to acquire the first intersection image of the intersection to be monitored in real time, and to identify all autonomous vehicles in the first intersection image to acquire the first vehicle information corresponding to each autonomous vehicle. The first path planning module is used to query the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan of each autonomous vehicle when passing through the intersection to be monitored; wherein, each path plan includes the position information and speed information of each autonomous vehicle at each moment. The waypoint planning map generation module is used to obtain the first waypoint planning map of the intersection to be monitored based on each first path plan and the intersection plan corresponding to the intersection to be monitored; wherein, the first waypoint planning map includes the path planning trajectory of each autonomous vehicle, and each path planning trajectory is presented in the form of waypoints on the first waypoint planning map; The module consists of a second acquisition module, a judgment module, a waiting module, a third acquisition module, and a second path planning module. The second acquisition module is used to acquire the second intersection image of the intersection to be monitored in real time when the next monitoring cycle arrives. The judgment module is used to identify all autonomous vehicles in the image captured at the first intersection and to determine whether any autonomous vehicles at the monitored intersection have changed. The waiting module is used to wait for the next monitoring cycle when the judgment module determines that the judgment result is negative; The third acquisition module is used to acquire the second vehicle information corresponding to each second autonomous driving vehicle in the second intersection image when the judgment module determines that the judgment result is yes. The second path planning module is used to query the database based on the information of each second vehicle to obtain the second path plan for each second autonomous vehicle when passing through the intersection to be monitored. The waypoint planning map generation module is also used to obtain the second waypoint planning map of the intersection to be monitored based on each second path plan and the first waypoint planning map; The waypoint planning map generation module includes: a judgment unit, a first planning unit, and a second planning unit; The judgment unit is used to sequentially determine whether the second autonomous vehicle has appeared in the first waypoint planning map; The first planning unit is used to compare the first path plan and the second path plan when the judgment unit determines that the judgment result has occurred, to obtain the speed comparison information of the two path plans at each time, and to generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map based on the second path plan and the speed comparison information. The second planning unit is used to generate the path planning trajectory of the second autonomous vehicle on the second waypoint planning map according to the second path planning when the judgment unit determines that the judgment result is not present.

5. The path planning device according to claim 4, characterized in that, The first acquisition module is used to identify all autonomous vehicles in the image captured at the first intersection and obtain the first vehicle information corresponding to each autonomous vehicle, specifically: The relative position information of each autonomous vehicle in the image taken at the first intersection is obtained, and the initial geographical location information of each autonomous vehicle is obtained based on the actual geographical information of the intersection to be monitored and the relative position information. Extract image information from the pictures taken by each autonomous vehicle at the first intersection, and obtain the unique vehicle code of each autonomous vehicle according to the preset recognition algorithm; The initial geographical location information and unique vehicle code of each autonomous vehicle are used as the corresponding first vehicle information.

6. The path planning device according to claim 5, characterized in that, The first path planning module is used to query the database based on the first vehicle information corresponding to each autonomous vehicle to obtain the first path plan for each autonomous vehicle when passing through the monitored intersection, specifically: Based on the vehicle's unique code, the database is queried to obtain the third path plan corresponding to each vehicle's unique code; wherein, the third path plan is uploaded and stored in the database by each autonomous vehicle. Based on the initial geographical location information, a path planning trajectory of a preset length is extracted from the third path planning, and based on each path planning trajectory, a first path planning is generated for each autonomous vehicle when passing through the intersection to be monitored.

Images