A method and device for scheduling route conflicts of unmanned logistics vehicles under a route network structure

Abstract

The present invention belongs to the technical field related to vehicle route planning and vehicle scheduling, and specifically is a method and device for scheduling route conflicts of unmanned logistics vehicles under a route network structure. The present invention uses a V2X vehicle-mounted terminal, so that the vehicle can obtain vehicle information in real time, and achieve real-time vehicle-side scheduling. Through the road network structure, locked routes and locked routes, and the use of stations, vehicles can improve the efficiency of multi-vehicle traffic in narrow sections and single-vehicle sections, and avoid congestion. At the same time, because of the use of a road network structure, corresponding road network relationships can be made for different environments, thereby improving the usability of the program.

Description

Technical Field

[0001] The present invention mainly relates to the technical fields related to unmanned vehicle scheduling and path planning, and specifically to a method and device for scheduling route conflicts of unmanned logistics vehicles under a route network structure. Background Art

[0002] Currently, many manufacturing companies have begun to use unmanned vehicles to transport finished products from finished product production workshops to warehouse storage. In this regard, unmanned vehicles need to carry different finished products to different warehouses. In general, in order to reasonably utilize the workshop space, a single lane is generally set up for passage between finished products and each warehouse. When multiple unmanned vehicles are running at the same time, route conflicts are prone to occur. Reasonable allocation of transportation tasks can improve the overall transportation efficiency; reasonable planning of unmanned vehicle routes can avoid blockages and collisions during transportation, and ensure the fluidity and safety of the transportation process.

[0003] There are many ways to allocate tasks and plan routes for multiple unmanned vehicles. One is to build models as the main method. The main process is to build corresponding mathematical models and use algorithms to solve task allocation and route planning, and develop prototype systems for task scheduling. In order to achieve the goal of evenly allocating handling tasks to unmanned vehicles, effectively planning collision-free and shortest paths for unmanned vehicles, and improving work efficiency. However, this method requires scheduling through a prototype system, requires collecting a large amount of data for model training, and conducts multiple, long-term stability tests. The performance of the algorithm is highly dependent on the quality and quantity of the training data. If the data is biased or insufficient, the output of the algorithm may be inaccurate. One is an algorithm-based method. The main process is to select and design algorithm strategies for specific scenarios, and establish scenario models to improve the algorithm's computational efficiency and accuracy. Task allocation, path planning, and path conflicts are handled using different algorithms to achieve path planning and collaborative scheduling. However, this method has low scene adaptability, and different scenarios require corresponding optimization and adjustment of the algorithm, which has high algorithm requirements. Summary of the invention

[0004] In order to solve the deficiencies of current technologies, the present invention combines existing technologies and, starting from practical applications, provides a method and device for scheduling route conflicts of unmanned logistics vehicles under a route network structure. When the unmanned vehicle enters automatic driving, the vehicle needs to obtain the task route under the route network structure and control the vehicle traffic strategy according to the task route situation. For example: in a road network structure, only one vehicle is allowed to pass, or in narrow road conditions, two vehicles start from the same place to the same destination, and then return. The two vehicles will inevitably encounter a route conflict. According to the route conflict situation, the vehicle side obtains information about other vehicles through communication equipment, analyzes the route conflict situation and the traffic permission strategy, and performs traffic control to avoid conflicts between the two vehicles on the same route section, which leads to congestion during the automatic driving process.

[0005] The technical solution of the present invention is as follows:

[0006] According to one aspect of the present invention, a method for scheduling route conflicts of unmanned logistics vehicles under a route network structure is provided, the method comprising:

[0007] Establish the routes from the departure point to each target point and make them into a route network. In the route network, the routes from the departure point to each target point and back are marked with digital numbers. The departure point and the target point are regarded as stations. Define vehicle types, local task routes, and locked task routes. Vehicle types include return vehicles and departure vehicles. The local task route is the route ID corresponding to the current station to the next station. The locked task route is obtained by the local task route association mapping.

[0008] Use the V2X vehicle terminal to transmit the real-time data of the vehicle. When the vehicle-side task allocation program obtains the transportation task, the vehicle publishes its own vehicle information through the V2X vehicle terminal and monitors the information of other vehicles. The vehicle scheduling strategy is as follows:

[0009] Determine whether the vehicle has received other vehicle information. If no other vehicle information is received, the vehicle traffic strategy is: pass, drive. When other vehicle information is received, determine the vehicle type of the own vehicle. If the vehicle type is a departure vehicle, execute step S1. If the vehicle type is a return vehicle, execute step S2.

[0010] Step S1: Determine whether the next station of the vehicle is occupied, if it is occupied, execute step S11, if it is not occupied, execute step S12;

[0011] Step S11 specifically includes:

[0012] Step S111: Publish the vehicle traffic strategy: queue at the station, park;

[0013] Step S112: Determine the vehicle type of the monitored vehicle. If the monitored vehicle type is a departure vehicle, determine whether the current route ID of the monitored vehicle is greater than the route ID corresponding to the current station plus 1. If so, issue the traffic strategy of the own vehicle: follow the vehicle or drive.

[0014] If the monitored vehicle type is a return vehicle, determine whether the current route ID of the monitored vehicle is greater than the maximum value of the locked task route ID issued by the own vehicle. If so, the traffic strategy of the own vehicle is: follow the vehicle and drive;

[0015] Step S12 specifically includes:

[0016] S121, determining whether there is a return vehicle within the locked mission route of the own vehicle, if yes, issuing the own vehicle's traffic strategy: queuing at the station, parking, and waiting for the return vehicle to pass, if no, issuing the own vehicle's traffic strategy: passing, driving;

[0017] S122, determining the vehicle type of the monitored vehicle, and when the vehicle type is a departure vehicle, comparing the route ID of the own vehicle with the route ID of the monitored vehicle, and issuing a traffic strategy for the own vehicle based on the comparison result;

[0018] Step S2: Based on the vehicle information of the monitored vehicle, the locked route array is calculated. The locked route array is obtained by matching the locked task route of the monitored vehicle with the global task route of the own vehicle. Then, it is determined whether the current route ID of the vehicle is equal to the first route ID of the locked route array. If so, the own vehicle stops and waits for the monitored vehicle to arrive at the station before continuing to the next station. Otherwise, the own vehicle continues to move forward to the next station.

[0019] Further, step S2 also includes:

[0020] In addition, a judgment condition is added to determine whether the current station of the own vehicle is equal to the next station of the monitored vehicle. If so, the passage strategy of the own vehicle is: pass, drive.

[0021] Furthermore, laser radar is used to collect map data of the transportation scene and establish a high-precision point cloud map. According to the actual route from the departure point to each target location in the transportation scene, a corresponding route network is established on the high-precision point cloud map. The route network should enable the unmanned vehicle to reach any one or more target locations from the departure point and return to the departure point.

[0022] Furthermore, the specific method of marking the route ID with digital numbers is as follows: the first route ID of the departure point is marked as 1, and then each route ID is increased by 1, and the numbers are marked in sequence until the return to the departure point.

[0023] Furthermore, when dispatching vehicles, it is set that only one vehicle can enter each target location, and other vehicles need to wait outside. At the same time, the route is restricted to a single-vehicle lane. The current station is defined as the station closest to the vehicle, and the next station is defined as the target location to which the vehicle will go.

[0024] Furthermore, the vehicle type is determined by the next stop of the vehicle. When the next stop of the vehicle is the departure location, the vehicle is a departure vehicle, otherwise the vehicle is a return vehicle.

[0025] Further, in step S1, the specific method for determining whether the next station of the own vehicle is occupied is as follows: determine whether the next station of the own vehicle is the same as the current station of the listening vehicle. If they are the same, it is determined that the next station of the own vehicle is occupied; otherwise, it is determined that the next station of the own vehicle is not occupied.

[0026] Further, in step S122, based on the comparison result, the traffic strategy of the own vehicle is issued, specifically:

[0027] When the route ID of the own vehicle is greater than the route ID of the monitoring vehicle, the traffic strategy of the own vehicle is: pass, drive;

[0028] When the route ID of the own vehicle is equal to the route ID of the monitored vehicle, the vehicle priority is assigned according to the vehicle serial number, so that the vehicle with high priority can depart and the vehicle with low priority can wait;

[0029] When the route ID of the own vehicle is smaller than the route ID of the monitoring vehicle, the own vehicle first waits for the vehicle in front to drive away, and then determines the route ID of the monitoring vehicle. When the route ID of the monitoring vehicle minus 1 is greater than the route ID of the own vehicle, the traffic strategy of the own vehicle is: follow the vehicle and drive.

[0030] According to another aspect of the present invention, there is provided a device for implementing the route conflict scheduling method for unmanned logistics vehicles under the above-mentioned route network structure, comprising a memory, a V2X vehicle-mounted terminal, and a processor, wherein the memory is used to store the route network, the V2X vehicle-mounted terminal is used to transmit real-time data of the vehicle, and the processor is used to execute the vehicle scheduling strategy.

[0031] Beneficial effects of the present invention:

[0032] 1. The present invention uses V2X vehicle-mounted terminals, so that vehicles can instantly obtain vehicle information and achieve real-time vehicle-side scheduling. Through the unique road network structure design, the locked routes and the use of stations, vehicles can communicate with each other in narrow sections and single-vehicle sections, improve the efficiency of multi-vehicle traffic and avoid congestion. At the same time, because of the use of road network structure, corresponding road network relationships can be made for different environments, thereby improving the usability of the program.

[0033] 2. The method of the present invention is simple to implement. Through reasonable logical operations, the vehicle dispatching algorithm has low requirements, fast operation and extremely high output accuracy.

[0034] 3. The method of the present invention comprehensively considers various problems that may arise during the dispatching process of unmanned vehicles, ensures the traffic efficiency of vehicles through reasonable response strategies, and also avoids the locking of vehicles when changing vehicle types. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Attached Figure 1 It is a flow chart of the present invention.

[0036] Attached Figure 2 This is a simplified schematic diagram from the finished product production workshop to the warehouse of the present invention.

[0037] Attached Figure 3 This is a schematic diagram of the route corresponding to step 7 in Example 1.

[0038] Attached Figure 4 The route corresponding to step 8 in Example 1 is shown as follows Figure 1 .

[0039] Attached Figure 5 The route corresponding to step 8 in Example 1 is shown as follows Figure 2 .

[0040] Attached Figure 6 This is a schematic diagram of the route corresponding to step 9 in Example 1.

[0041] Attached Figure 7 The route corresponding to step 10 in Example 1 is shown as follows Figure 1 .

[0042] Attached Figure 8 The route corresponding to step 10 in Example 1 is shown as follows Figure 2 .

[0043] Attached Fig. 9 The route corresponding to step 10 in Example 1 is shown as follows Figure 3 . DETAILED DESCRIPTION

[0044] The present invention will be further described with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art may make various changes or modifications to the present invention, and these equivalent forms also fall within the scope limited by the application.

[0045] Example 1

[0046] This embodiment provides a method for scheduling route conflicts of unmanned logistics vehicles under a route network structure. In this embodiment, the scheduling method is described in detail with the departure location being the finished product workshop and the destination location being the warehouse. It should be understood that the unmanned logistics vehicle transporting goods between the workshop and the warehouse is an extremely common scenario, but this method is also applicable to similar or identical scenarios.

[0047] The specific method for scheduling route conflicts of unmanned logistics vehicles under the route network structure is as follows.

[0048] Step 1: Use LiDAR to collect map data of transportation scenarios and create a high-precision point cloud map.

[0049] Step 2. According to the actual routes from the finished product production workshop to each warehouse in the transportation scenario, establish corresponding routes on the high-precision point cloud map, and make them into a route network, so that the unmanned vehicle can reach any one or more warehouses from the finished product production workshop and return to the finished product production workshop. The route network produced is marked with the route ID. Specifically, taking the finished product workshop as the starting point, the next section of the route is marked as 1, and the next section of the route after number 1 is marked as 2. The ID increases by 1 in sequence. The standard order is to reach each warehouse in sequence from the finished product workshop and then return to the finished product workshop. It should be understood that the distance from the finished product workshop to the warehouse or from the previous warehouse to the next warehouse may be composed of multiple road network IDs. When marking the route ID, the specific ID from the main route to the warehouse is also marked. For reference Figure 2 As shown, this is a simplified schematic diagram from the finished product production workshop to the warehouse. The solid arrow in the figure is the forward route, and the dotted line is the return route.

[0050] Step 3: Figure 2 Taking the route shown as an example, it is set that the vehicle needs to start from the finished product production workshop and enter warehouse 1, warehouse 2, warehouse 3 and then return. Only one vehicle can enter each warehouse, and the other vehicles need to wait outside. At the same time, the route is restricted to a single-vehicle lane, and when there is a car in warehouse 2, the vehicle in warehouse 1 cannot go to warehouse 2, that is, the next warehouse is not empty and cannot leave the current warehouse. The workshop and warehouse can be regarded as sites.

[0051] Step 4: Use the V2X vehicle terminal to transmit real-time data of vehicles, so that vehicles can read each other's information.

[0052] Step 5. When a vehicle's task allocation program obtains a transportation task, the vehicle publishes its own vehicle information (self_car) through the V2X vehicle-mounted terminal device and listens to other vehicle information (recv_car). V2X means vehicle to everything, that is, the vehicle exchanges information with the outside world. The V2X vehicle-mounted terminal device is a device for exchanging information between the vehicle and the outside world. Because the task has been obtained, the vehicle can obtain the current site (Scene_index, the site closest to the vehicle. If the vehicle is in the finished product production workshop, the finished product production workshop is the current site) and the next site (nextScene_index, the warehouse or finished product workshop to which the vehicle will go), the ID of the global task route (line_ids, the global task route is the route from the finished product production workshop to the warehouse and then back.) Figure 2 As shown in the diagram, assuming that the task goes from the finished product production workshop to warehouse 3 and then returns, the global task route ID should be 1 to 28.

[0053] With the global task route and site information, the locked task route (lock_line_ids, the vehicle itself is locked according to two sites, such as from the finished product production workshop to warehouse 1, the local task route is the route ID corresponding to the current site to the route ID corresponding to the next site, that is, (1, 2, 3, 4, 5). The locked task route is obtained by the local task route association mapping and should be: (24, 25, 26, 27, 28).) The vehicle type (car_type) can be determined by the vehicle's next site (vehicle type definition: when the route ID corresponding to the vehicle's next site is the finished product production workshop (such as Figure 2 As shown in the figure, the route ID corresponding to the site of the finished product production workshop is 28, which can be obtained by associating and mapping route 1 in the route network structure) for the return vehicle (car_type=2), otherwise it is the departure vehicle (car_type=1)) The above relationship can be expressed by the following expression:

[0054] self_car.nextScene_index ==28; self_car.car_type=2;

[0055] self_car.nextScene_index≠28; self_car.car_type=1.

[0056] Step 6: According to the type of vehicle, determine whether other vehicle information has been received. If no other vehicle information is received, the vehicle passage strategy (car_passage_action: 0 default value, 1 pass, 2 follow the vehicle, 3 stop and yield, 4 queue at the station) and vehicle driving state (car_driver_state: 0 default value, 1 drive, 2 stop) are both 1, which can be expressed by the following formula:

[0057] self_car.car_passage_action=1;

[0058] self_car.car_driver_state=1;

[0059] The above formula means that if the vehicle does not receive information from other vehicles, the vehicle's traffic strategy is: pass and drive.

[0060] Step 7: When receiving other vehicle information, and the vehicle type of the vehicle itself (self_car.car_type=1), first determine whether the next station of the vehicle itself is the same as the current station of the received vehicle, that is:

[0061] self_car.nextScene_index = recv_car.Scene_index, the next site of the departure vehicle is occupied,

[0062] self_car.nextScene_index recv_car.Scene_index, the next station of the departing vehicle is not occupied;

[0063] If the next station of the vehicle is occupied, the vehicle traffic strategy is issued: station queuing, parking, the expression is as follows:

[0064] self_car.car_passage_action=4;

[0065] self_car.car_driver_state=2.

[0066] like Figure 3 As described above, vehicle 1 is in the finished product production workshop, vehicle 2 is in warehouse 1, and the vehicles will go to warehouse 2. Vehicle 2 can go, but vehicle 1 needs to wait.

[0067] Step 8: Based on the above step 7, determine the vehicle type of the received vehicle information. If the received vehicle type is the departure vehicle (recv_car.car_type=1), determine whether the received vehicle's current route ID (recv_car.line_id) is greater than the current station plus 1:

[0068] recv_car.line_id >(recv_car.Scene_index+1);

[0069] If true, the traffic strategy of the vehicle is published: follow the vehicle, drive, the expression is as follows:

[0070] self_car.car_passage_action=2;

[0071] self_car.car_driver_state=1.

[0072] like Figure 4 As shown, when car 2 arrives at route ID 7, car 1 can depart.

[0073] If the received vehicle type is a return vehicle (recv_car.car_type=2), determine whether the route ID of the received vehicle information is greater than the last route ID of the locked task route issued by the own vehicle, that is:

[0074] When recv_car.line_id > self_car.lock_line_ids[last], the vehicle traffic strategy is published: pass, drive, the expression is as follows:

[0075] self_car.car_passage_action=1;

[0076] self_car.car_driver_state=1.

[0077] like Figure 5 As described above, the route locked by car 1 is (17, 18, 19, 20, 21). When car 2 is on route 22, car 1 can go to warehouse 2.

[0078] Step 9: If the station in step 7 is not occupied by other vehicles, first determine whether there is a return vehicle within the locked mission route of the vehicle. If there is, it is necessary to wait for the return vehicle to pass. If not, the vehicle can move forward normally. The expression is as follows: Σ i=0 (self_car.lock_line_ids[i]==recv_car.line_id).

[0079] When the result of the above formula is not 0, it means that the vehicle needs to wait.

[0080] like Figure 6 As shown, vehicle 1 is the departure vehicle, and vehicle 2 is the return vehicle. The locked routes of vehicle 1 are (20, 21, 22, 23, 24), and vehicle 2 is on route 23 at this time. Vehicle 1 needs to wait for vehicle 2 to arrive at route 25 before it can come out of warehouse 1.

[0081] At the same time, it is necessary to make another judgment. When the received vehicle information and the own vehicle are both departure vehicles, it is necessary to determine the route ID of the vehicle and make further vehicle traffic strategy judgments based on the size of the route ID. When the route ID of the own vehicle is greater than the route ID of the received vehicle information:

[0082] self_car.line_id>recv_car.line_id;

[0083] This means that the vehicle is in front of other vehicles, and the traffic strategy of the vehicle itself is: pass, drive.

[0084] self_car.car_passage_action=1;

[0085] self_car.car_driver_state=1;

[0086] When the route ID of the own vehicle is equal to the route ID of the received vehicle information:

[0087] self_car.line_id == recv_car.line_id;

[0088] Add judgment conditions and assign vehicle priority according to the vehicle number (car_level, 1 to N, 1 is the highest priority).

[0089] The vehicle's traffic strategy is determined based on priority. If car_level=1, the vehicle can depart, and if car_level=2, it needs to wait.

[0090] When the route ID of the own vehicle is less than the route ID of the received vehicle information:

[0091] self_car.line_id < recv_car.line_id;

[0092] The vehicle first waits for the vehicle in front to drive away, and then determines the route ID of the received vehicle information. When the received vehicle route ID-1 is greater than the vehicle route ID:

[0093] (recv_car.line_id - 1) > self_car.line;

[0094] Then the vehicle can start and keep following the vehicle in front.

[0095] Step 10. When receiving other vehicle information, and the vehicle type of the vehicle itself (self_car.car_type=2), that is, the vehicle itself is a return vehicle, then according to the received vehicle information, the locked route array (passive_lock_line_ids, obtained by matching the received locked route with its own global task route) can be calculated:

[0096] passive_lock_line_ids=line_ids∩ lock_line_ids;

[0097] Then determine whether the vehicle's current route ID (line_id) is in the first element of the locked route array:

[0098] self_car.line_id == passive_lock_line_ids[0];

[0099] If the formula is true, the vehicle stops and waits for another vehicle to arrive at the station before continuing to the station.

[0100] like Figure 7 As shown, car 1 goes to warehouse 2 and sends out a locked route (20, 21, 22, 23, 24). Car 2 receives the locked route sent by car 1 and matches the locked route (20, 21, 22, 23, 24). When car 2 is on route 20, it needs to wait for car 1 to enter warehouse 2 before it can continue to move forward.

[0101] If the formula is not true, the vehicle continues to move forward, such as Figure 8 As shown, vehicle 2 is not on the locked route of vehicle 1 or on the locked route, and vehicle 2 continues to move forward.

[0102] In order to avoid the vehicle from being locked when changing vehicle types, when the vehicle is returning, it is necessary to add an additional judgment: whether the current station of the vehicle is equal to the next station of the received vehicle.

[0103] When self_car.Scene_index == recv_car.nextScene_index, the vehicle's traffic strategy is: pass or drive.

[0104] self_car.car_passage_action=1;

[0105] self_car.car_driver_state=1;

[0106] like Fig. 9 As shown: when car 1 is in warehouse 2, car 2 is in warehouse 3. Because car 2 occupies warehouse 3, car 1 enters step 8. Car 2 cannot move forward because of the locked route. The two cars are stuck in a traffic jam. Therefore, it is necessary to add: when the current station of the own vehicle is equal to the next station of the other vehicle, the own vehicle can drive normally.

[0107] In this embodiment, only one vehicle is allowed to pass through the road network structure from the finished product workshop to the warehouse. When multiple vehicles pass through, based on the route conflict situation, the vehicle side obtains the information of other vehicles through the communication equipment, analyzes the route conflict situation and the access permission strategy, and performs access control to avoid conflicts between two vehicles on the same route, which may cause congestion during the automatic driving process.

[0108] Example 2

[0109] This embodiment also provides a route conflict scheduling device for unmanned logistics vehicles under a route network structure, which mainly includes a memory, a V2X vehicle-mounted terminal, and a processor, wherein the memory is used to store route networks, scheduling programs, etc., the V2X vehicle-mounted terminal is used to transmit real-time data of the vehicle, including the vehicle's location, the vehicle's operating status, and monitored vehicle information, etc., and the processor is used to execute the vehicle scheduling strategy in Example 1, thereby controlling the orderly passage of unmanned vehicles.

Claims

1. A method for scheduling route conflicts of unmanned logistics vehicles under a route network structure, characterized in that: include: Establish the routes from the departure point to each target point and make them into a route network. In the route network, the routes from the departure point to each target point and back are marked with digital numbers. The departure point and the target point are regarded as stations. Define vehicle types, local task routes, and locked task routes. Vehicle types include return vehicles and departure vehicles. The local task route is the route ID corresponding to the current station to the next station. The locked task route is obtained by the local task route association mapping. Use the V2X vehicle terminal to transmit the real-time data of the vehicle. When the vehicle-side task allocation program obtains the transportation task, the vehicle publishes its own vehicle information through the V2X vehicle terminal and monitors the information of other vehicles. The vehicle scheduling strategy is as follows: Determine whether the vehicle has received other vehicle information. If no other vehicle information is received, the vehicle traffic strategy is: pass, drive. When other vehicle information is received, determine the vehicle type of the own vehicle. If the vehicle type is a departure vehicle, execute step S1. If the vehicle type is a return vehicle, execute step S2. Step S1: Determine whether the next station of the vehicle is occupied, if it is occupied, execute step S11, if it is not occupied, execute step S12; Step S11 specifically includes: Step S111: Publish the vehicle traffic strategy: queue at the station, park; Step S112: Determine the vehicle type of the monitored vehicle. If the monitored vehicle type is a departure vehicle, determine whether the current route ID of the monitored vehicle is greater than the route ID corresponding to the current station plus 1. If so, issue the traffic strategy of the own vehicle: follow the vehicle or drive. If the monitored vehicle type is a return vehicle, determine whether the current route ID of the monitored vehicle is greater than the maximum value of the locked task route ID issued by the own vehicle. If so, the traffic strategy of the own vehicle is: follow the vehicle and drive; Step S12 specifically includes: S121, determining whether there is a return vehicle within the locked mission route of the own vehicle, if yes, issuing the own vehicle's traffic strategy: queuing at the station, parking, and waiting for the return vehicle to pass, if no, issuing the own vehicle's traffic strategy: passing, driving; S122, determining the vehicle type of the monitored vehicle, and when the vehicle type is a departure vehicle, comparing the route ID of the own vehicle with the route ID of the monitored vehicle, and issuing a traffic strategy for the own vehicle based on the comparison result; Step S2: Based on the vehicle information of the monitored vehicle, the locked route array is calculated. The locked route array is obtained by matching the locked task route of the monitored vehicle with the global task route of the own vehicle. Then, it is determined whether the current route ID of the vehicle is equal to the first route ID of the locked route array. If so, the own vehicle stops and waits for the monitored vehicle to arrive at the station before continuing to the next station. Otherwise, the own vehicle continues to move forward to the next station.

2. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 1 is characterized in that: Step S2 also includes: In addition, a judgment condition is added to determine whether the current station of the own vehicle is equal to the next station of the monitored vehicle. If so, the passage strategy of the own vehicle is: pass, drive.

3. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 1 is characterized in that: Use lidar to collect map data of the transportation scene and create a high-precision point cloud map. According to the actual route from the departure point to each target location in the transportation scene, establish a corresponding route network on the high-precision point cloud map. The route network should enable the unmanned vehicle to reach any one or more target locations from the departure point and return to the departure point.

4. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 1 is characterized in that: The specific method of marking the route ID with numbers is as follows: the first route ID of the departure point is marked as 1, and then the ID of each route section is increased by 1, and the numbers are marked in sequence until the return to the departure point.

5. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 1 is characterized in that: When dispatching vehicles, it is set that only one vehicle can enter each target location, and other vehicles need to wait outside. At the same time, the route is restricted to a single-vehicle lane, and the next target location is not empty and cannot leave the current target location. The current station is defined as the station closest to the vehicle, and the next station is defined as the target location to which the vehicle will go.

6. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 5 is characterized in that: The vehicle type is determined by the next stop of the vehicle. When the next stop of the vehicle is the departure location, the vehicle is a departure vehicle, otherwise the vehicle is a return vehicle.

7. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 1 is characterized in that: In step S1, the specific method for determining whether the next station of the own vehicle is occupied is as follows: determine whether the next station of the own vehicle is the same as the current station of the listening vehicle. If they are the same, it is determined that the next station of the own vehicle is occupied; otherwise, it is determined that the next station of the own vehicle is not occupied.

8. The route conflict scheduling method for unmanned logistics vehicles under the route network structure according to claim 1 is characterized in that: In step S122, based on the comparison result, the traffic strategy of the own vehicle is issued, specifically: When the route ID of the own vehicle is greater than the route ID of the monitoring vehicle, the traffic strategy of the own vehicle is: pass, drive; When the route ID of the own vehicle is equal to the route ID of the monitored vehicle, the vehicle priority is assigned according to the vehicle serial number, so that the vehicle with high priority can depart and the vehicle with low priority can wait; When the route ID of the own vehicle is smaller than the route ID of the monitoring vehicle, the own vehicle first waits for the vehicle in front to drive away, and then determines the route ID of the monitoring vehicle. When the route ID of the monitoring vehicle minus 1 is greater than the route ID of the own vehicle, the traffic strategy of the own vehicle is: follow the vehicle and drive.

9. A device for implementing the route conflict scheduling method for unmanned logistics vehicles under the route network structure described in any one of claims 1 to 8, characterized in that: It includes a memory, a V2X vehicle-mounted terminal, and a processor. The memory is used to store route networks, the V2X vehicle-mounted terminal is used to transmit real-time data of vehicles, and the processor is used to execute vehicle scheduling strategies.

Images