Road network acquisition method and device of unmanned vehicle, equipment and medium

Abstract

The invention provides a road network obtaining method, device and equipment for an unmanned vehicle and a medium, and relates to the technical field of unmanned driving, and the method comprises the steps: obtaining a virtual object of the unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle; acquiring a virtual obstacle set in the first simulation environment, and controlling each virtual obstacle to move to obtain a second simulation environment; obtaining an initial road network of the virtual object, and controlling the virtual object to virtually travel in the second simulation environment based on the initial road network; in the virtual driving process of the virtual object, dynamic distance parameters between the virtual object and the virtual obstacles are obtained; and according to the dynamic distance parameter, identifying whether a risk exists in the virtual driving process of the virtual object so as to update the initial road network and obtain a target road network of the unmanned vehicle in the actual driving field.

Description

Technical Field

[0001] The present disclosure relates to the field of unmanned driving, and in particular to a method, device, equipment and medium for acquiring a road network of an unmanned vehicle. Background Art

[0002] With the development of technology, unmanned engineering transport vehicles are becoming increasingly important in construction operations in precast beam yards. During the daily driving of unmanned engineering transport vehicles, the moving obstacles that may exist in the precast beam yard may have a certain degree of impact on the driving safety of the unmanned engineering transport vehicles.

[0003] In the related technology, unmanned engineering transport vehicles can be tested and operated in the actual environment of the beam yard, which has high risks and costs. Summary of the invention

[0004] The present disclosure proposes a road network acquisition method, device, equipment and medium for an unmanned vehicle.

[0005] According to a first aspect of the present disclosure, a method for acquiring a road network of an unmanned vehicle is proposed, comprising: acquiring a virtual object of the unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle; acquiring a set of virtual obstacles in the first simulation environment, and controlling the movement of each virtual obstacle to obtain a second simulation environment; acquiring an initial road network of the virtual object, and controlling the virtual object to virtually drive in the second simulation environment based on the initial road network; acquiring dynamic distance parameters between the virtual object and each virtual obstacle during the virtual driving of the virtual object; and updating the initial road network according to the dynamic distance parameters to obtain a target road network of the unmanned vehicle in the actual driving site.

[0006] According to a second aspect of the present disclosure, a road network acquisition device for an unmanned vehicle is proposed, comprising: a first acquisition module, used to acquire a virtual object of the unmanned vehicle and a first simulation environment corresponding to the actual driving site of the unmanned vehicle; a second acquisition module, used to acquire a set of virtual obstacles in the first simulation environment, and control the movement of each virtual obstacle to obtain a second simulation environment; a virtual driving module, used to acquire an initial road network of the virtual object, and control the virtual object to virtually drive in the second simulation environment based on the initial road network; a third acquisition module, used to acquire dynamic distance parameters between the virtual object and each virtual obstacle during the virtual driving of the virtual object; and a fourth acquisition module, used to update the initial road network according to the dynamic distance parameters to obtain a target road network of the unmanned vehicle in the actual driving site.

[0007] According to the third aspect of the present disclosure, an electronic device is proposed, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the road network acquisition method for an unmanned vehicle proposed in the first aspect above.

[0008] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is proposed, wherein the computer instructions are used to enable the computer to execute the road network acquisition method for an unmanned vehicle proposed in the first aspect above.

[0009] According to a fifth aspect of the present disclosure, a computer program product is proposed, including a computer program, which, when executed by a processor, implements the road network acquisition method for an unmanned vehicle proposed in the first aspect above.

[0010] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0012] Figure 1 A schematic diagram of a process of a road network acquisition method for an unmanned vehicle according to an embodiment of the present disclosure;

[0013] Figure 2 A schematic diagram of a process flow of a method for acquiring a road network of an unmanned vehicle according to another embodiment of the present disclosure;

[0014] Figure 3 A schematic diagram of a process flow of a method for acquiring a road network of an unmanned vehicle according to another embodiment of the present disclosure;

[0015] Figure 4 A schematic diagram of the structure of a road network acquisition device for an unmanned vehicle according to an embodiment of the present disclosure;

[0016] Figure 5 A schematic block diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0017] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0018] Data processing is a basic part of system engineering and automatic control. Data is a form of expression of facts, concepts or instructions, which can be processed by manual or automatic devices. After data is interpreted and given a certain meaning, it becomes information. Data processing is the collection, storage, retrieval, processing, transformation and transmission of data. The basic purpose of data processing is to extract and derive valuable and meaningful data for certain specific people from a large amount of data that may be disorganized and difficult to understand.

[0019] Autonomous driving refers to the technology that enables the car to perceive the environment, plan its route and control the vehicle autonomously. It uses artificial intelligence and deep learning to "teach" vehicles equipped with various types of sensors to drive.

[0020] Figure 1 FIG. 1 is a flow chart of a method for acquiring a road network of an unmanned vehicle according to an embodiment of the present disclosure. Figure 1 As shown, the method includes:

[0021] S101, obtaining a virtual object of the unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle.

[0022] In the daily operation of the beam yard, unmanned engineering transport vehicles have become important engineering machinery vehicles in the daily operation of the beam yard. In this scenario, the safe driving of unmanned engineering transport vehicles in the beam yard environment is very important.

[0023] Optionally, the safe driving routes of unmanned engineering transport vehicles in the beam yard environment can be tested through virtual simulation, so as to obtain the safe driving routes of unmanned engineering transport vehicles in the beam yard environment, and then obtain the safe driving road network of unmanned concrete mixer trucks in the beam yard environment composed of various safe driving routes.

[0024] Among them, the unmanned engineering transport vehicle can be an unmanned concrete mixer transport vehicle, or it can be an unmanned engineering transport vehicle for transporting other materials, and no specific limitation is made here.

[0025] In the embodiments of the present disclosure, an unmanned vehicle for which a safe driving road network needs to be determined through virtual simulation can be obtained, and its projected object in the virtual simulation environment can be determined as a virtual object of the unmanned vehicle. The unmanned vehicle can be modeled based on the modeling method of virtual objects in related technologies, thereby obtaining a virtual object of the unmanned vehicle in the virtual simulation environment.

[0026] In this scenario, the actual driving site of the unmanned vehicle can be obtained, and a virtual simulation environment can be modeled based on the actual driving site, so as to obtain a simulation environment corresponding to the actual driving site in the virtual simulation environment, and the simulation environment can be determined as the first simulation environment.

[0027] S102, obtaining a set of virtual obstacles in the first simulation environment, and controlling the movement of each virtual obstacle to obtain a second simulation environment.

[0028] In the disclosed embodiment, there are various types of obstacles in the actual driving field of the unmanned concrete mixer truck. Therefore, the projection objects of each obstacle in the actual driving field in the first simulation environment can be determined as virtual obstacles in the first simulation environment, thereby obtaining a virtual obstacle set composed of each virtual obstacle.

[0029] During the actual driving of an unmanned vehicle, the presence of movable obstacles in the actual driving site may affect the safe driving of the unmanned vehicle. In this scenario, in order to obtain a safe driving road network for the unmanned vehicle, the movement of each virtual obstacle in the first simulation environment can be simulated based on the actual movement of the movable obstacles in the actual driving site.

[0030] Among them, relevant movement information such as the movement trajectory of the movable obstacles in the actual driving field can be obtained, and based on the obtained movement information, each virtual obstacle is controlled to move in the first simulation environment, and the first simulation environment after each virtual obstacle starts to move is determined as the second simulation environment.

[0031] S103, obtaining an initial road network of the virtual object, and controlling the virtual object to virtually travel in the second simulation environment based on the initial road network.

[0032] In the embodiment of the present disclosure, there is a pre-planned initial route for the virtual object in the second simulation environment, and a road network composed of the initial routes can be determined as the initial road network of the virtual object.

[0033] Among them, the initial road network can be determined based on the relative position relationship between the target operating areas of the unmanned vehicle in the actual driving site, the vehicle departure area and the non-moving obstacles in the site, or based on other route planning methods, which are not specifically limited here.

[0034] Optionally, based on the virtual simulation method in the relevant technology, the virtual simulation of the virtual object in the current round can be started. It can be understood that after the virtual simulation of the current round is started, the virtual object will virtually travel on the corresponding formal route based on the preset driving information.

[0035] Among them, the virtual driving of the virtual object in the second simulation environment can be realized by writing corresponding code snippets, and the virtual driving of the virtual object in the second simulation environment can also be realized by other methods, which are not specifically limited here.

[0036] S104, during the virtual driving of the virtual object, obtaining dynamic distance parameters between the virtual object and each virtual obstacle.

[0037] In the disclosed embodiment, during the virtual driving process of the virtual object, the distance relationship between the virtual object and each virtual obstacle can be continuously detected, and based on the detected distance relationship, it can be identified whether there are obstacles among the virtual obstacles that affect the safe driving of the virtual object, and corresponding processing can be performed to obtain a safe driving road network for the unmanned vehicle corresponding to the virtual object in the actual driving field corresponding to the second simulation environment.

[0038] Optionally, the parameter describing the distance relationship between the virtual object and each virtual obstacle may be marked as a dynamic distance parameter, wherein the dynamic distance parameter may continuously change with the movement of the virtual obstacle and the movement of the virtual object.

[0039] As an example, an object bounding box of a virtual object and an obstacle bounding box of each virtual obstacle may be obtained, and a dynamic distance relationship between the virtual object and each virtual obstacle may be obtained by detecting a dynamic distance between the object bounding box and each obstacle bounding box.

[0040] S105, updating the initial road network according to the dynamic distance parameter to obtain the target road network of the unmanned vehicle in the actual driving field.

[0041] In the disclosed embodiment, for any virtual obstacle, some obstacles that affect the safe driving of the virtual object can be screened out from the virtual obstacles based on the dynamic distance parameter between the virtual obstacle and the virtual object, and then some driving routes that cannot meet the safe driving requirements of the virtual object can be screened out from the initial road network, and the initial road network can be updated based on the partial driving routes.

[0042] Among them, the part of the driving route can be deleted from the initial road network, or other marking processing can be performed to achieve the update of the initial road network.

[0043] Furthermore, based on the updated initial road network, a safe driving route is determined in which the virtual object can safely drive in the second simulation environment and can reach the relevant target operation area, thereby obtaining a safe driving road network composed of the safe driving routes.

[0044] In this scenario, the safe driving road network is the target road network for safe driving of the unmanned vehicle corresponding to the virtual object in the actual driving field corresponding to the second simulation environment.

[0045] The road network acquisition method for an unmanned vehicle proposed in the present disclosure obtains a second simulation environment by moving each virtual obstacle in a first simulation environment, realizes virtual driving simulation of the virtual object corresponding to the unmanned vehicle in the second simulation environment, and obtains a target road network according to the result of the virtual driving simulation. Compared with the target road network obtained by test driving in an actual driving field in the related art, the acquisition cost and risk of the target road network for safe driving of the unmanned vehicle are reduced. The target road network is obtained by the second simulation environment composed of each virtual obstacle in a moving state, which reduces the influence of each obstacle in the actual driving field on the safe driving of the unmanned vehicle, improves the accuracy of the acquisition of the target road network of the unmanned vehicle in the actual driving field, improves the safety of the unmanned vehicle driving based on the target road network, reduces the probability of abnormal situation in which the unmanned vehicle changes lanes due to driving safety being affected, shortens the driving time for the unmanned vehicle to reach the operating area, improves the operating efficiency of the unmanned vehicle, thereby improving the safety, stability and operating efficiency of the project to which the unmanned vehicle belongs, and optimizes the driving method of the unmanned vehicle.

[0046] In the above embodiment, the acquisition of the target road network can also be combined with Figure 2 understand, Figure 2 FIG. 1 is a flow chart of a method for acquiring a road network of an unmanned vehicle according to another embodiment of the present disclosure. Figure 2 As shown, the method includes:

[0047] S201, obtaining a current initial driving route of a virtual object from an initial road network.

[0048] In the embodiment of the present disclosure, the initial road network may include multiple driving routes that the virtual object can virtually travel in the second simulation environment. In this scenario, the driving route that the virtual object currently needs to travel in the second simulation environment can be determined based on the virtual operation area determined by the current round of virtual simulation, and this route can be determined as the initial driving route.

[0049] S202: Controlling a virtual object to virtually drive on the initial driving route based on a first virtual driving strategy corresponding to the initial driving route, wherein the first virtual driving strategy at least includes a speed strategy and a direction strategy of the virtual object on the initial driving route.

[0050] In the disclosed embodiment, a constraint speed set is provided for the virtual object and each virtual obstacle in the second simulation environment, wherein the constraint data set may include an environmental speed upper limit value and an environmental acceleration upper limit value limited by the second simulation environment, and a model speed upper limit value and a model acceleration upper limit value limited by the vehicle model corresponding to the virtual object.

[0051] In this scenario, the speed information and acceleration information related to the virtual object can be obtained from the constraint data set to obtain the expected speed information and expected acceleration information of the virtual object when performing virtual driving on the initial driving route, thereby obtaining the speed strategy of the virtual object on the initial driving route.

[0052] Optionally, the moving direction strategy of the virtual object on the initial driving route can also be obtained from the constraint data set. As an example, the wheel angle of the virtual object can be calculated by the kinematic model in the relevant technology, wherein the wheel angle calculation formula can be as follows:

[0053]

[0054] In the above formula, Δθ represents the angle of change in speed direction, Φ represents the steering wheel angle (which can be understood as the wheel angle), L represents the wheelbase, v represents the vehicle speed, and Δt represents the update time step.

[0055] It should be noted that the wheel angle obtained based on the above algorithm formula can be smaller than the upper limit of the wheel angle corresponding to the virtual object, and the wheel angle angular velocity is smaller than the upper limit of the angular velocity corresponding to the virtual object. In this scenario, the conventional angular acceleration of the virtual object can be deflected based on the preset trajectory deviation data to achieve the adjustment of the driving direction of the virtual object.

[0056] In this scenario, the direction strategy used by the virtual object when performing virtual driving on the initial driving route can be obtained based on the acquired wheel angle and wheel angle angular velocity of the virtual object.

[0057] Furthermore, the acquired speed strategy and direction strategy are integrated to obtain a first virtual driving strategy adopted by the virtual object when performing virtual driving on the initial driving route.

[0058] It should be noted that the first virtual driving strategy may include a direction strategy and a speed strategy of the virtual object, and may also include other strategies required by the virtual object when performing virtual driving, which are not specifically limited here.

[0059] S203, during the virtual driving of the virtual object, obtaining dynamic distance parameters between the virtual object and each virtual obstacle.

[0060] Optionally, a first speed of the virtual object and a second speed of each virtual obstacle are acquired, and a dynamic distance parameter between the virtual object and each virtual obstacle is obtained according to the first speed and the second speed.

[0061] As an example, it can be understood in combination with the following formula:

[0062] s=p0+v0*Δt-p1+v1*Δt

[0063] s p =p0+v0*Δt*ξ p -p1+v1*Δt*ξ p

[0064] In the above formula, s represents the distance between the virtual object and the virtual obstacle at the next moment when the virtual object is virtually driving based on the initial driving route, and s p represents the predicted distance between the virtual object and each virtual obstacle at the next moment when the virtual object performs virtual driving based on the initial driving route, p0 represents the current position of the virtual object, p1 represents the current position of the virtual obstacle, v0 represents the first speed of the virtual object, v1 represents the second speed of the virtual obstacle, Δt represents the update time step, ξ p Indicates the number of predicted frames.

[0065] In this example, the distance between the virtual object and each virtual obstacle at the next moment and the predicted distance at the next moment can be calculated according to the algorithm formula shown in the above content, and the distance at the next moment and the predicted distance can be used as the dynamic distance parameters between the virtual object and each virtual obstacle.

[0066] S204 , for any dynamic distance parameter, in response to identifying that a virtual driving process of the virtual object has a risk according to the dynamic distance parameter, obtaining a risk handling strategy for the virtual object.

[0067] Optionally, in response to the dynamic distance parameter being less than a preset collision distance threshold, a deceleration and avoidance strategy of the virtual object is obtained as a risk handling strategy, wherein the deceleration and avoidance strategy is obtained based on a preset deviation distance and offset direction of the virtual object.

[0068] In the embodiment of the present disclosure, the virtual obstacle is a mobile obstacle in the second simulation environment. In this scenario, when the virtual object is performing virtual driving on the initial driving route, a virtual obstacle may enter the front area of ​​the virtual object on the route. In this scenario, it can be identified whether the virtual obstacle entering the front area will affect the safe driving of the virtual object based on the dynamic distance parameter between the two.

[0069] Among them, for any virtual obstacle entering the area, when the dynamic distance parameter between the virtual object and the virtual obstacle is less than the preset collision distance threshold, it can be determined that the virtual object may collide with the virtual obstacle based on the current driving speed and driving direction.

[0070] In this scenario, the virtual obstacle needs to be avoided, wherein the virtual object can be decelerated and its driving direction can be changed so that the virtual object can avoid the virtual obstacle. This strategy can be determined as a deceleration and avoidance strategy for the virtual object. Further, the deceleration and avoidance strategy is determined as a risk handling strategy for the virtual object in the current scenario.

[0071] Among them, the preset deceleration information of the virtual object in the current scene can be obtained from the first virtual driving strategy, and the driving speed of the virtual object can be adjusted based on the obtained deceleration information to achieve deceleration of the virtual object.

[0072] Furthermore, the preset offset direction and deviation distance of the virtual object in the current scene can be obtained from the first virtual driving strategy, and the virtual object can be controlled to turn based on the offset direction until the virtual object reaches the deviation distance, thereby achieving the virtual object's avoidance of the virtual obstacle.

[0073] It should be noted that, based on the deceleration and avoidance strategy, the virtual object can be decelerated based on a preset deceleration upper limit value, thereby achieving obstacle avoidance of the virtual object to the virtual obstacle.

[0074] Optionally, in response to the dynamic distance parameter being greater than or equal to a preset collision distance threshold and less than a preset safety distance threshold, a deceleration strategy of the virtual object is acquired as a risk handling strategy.

[0075] It can be understood that when the dynamic distance parameter is greater than or equal to the collision distance threshold and less than the safety distance threshold, it can be determined that the virtual object can avoid collision with the virtual obstacle by slowing down the current driving speed. In this scenario, the preset deceleration strategy of the virtual object can be obtained and determined as the risk handling strategy of the virtual object in the current scenario.

[0076] The deceleration strategy may be a conventional deceleration process implemented by the virtual object based on a deceleration rate used in daily life, or may be a conventional deceleration process implemented by the virtual object based on other set deceleration rates, which is not specifically limited here.

[0077] Optionally, in response to the dynamic distance parameter being greater than or equal to a safety distance threshold, the virtual object is kept moving based on the current driving state.

[0078] It can be understood that when the dynamic distance parameter is greater than or equal to the safety distance threshold, based on the current driving direction and the current driving speed, the virtual object will not collide with the virtual obstacle. In this scenario, there is no need to adjust the driving state of the virtual object. In other words, the virtual object can continue to perform virtual driving on the initial driving route based on the current driving state.

[0079] S205, in response to obtaining an updated driving route of the initial driving route based on the risk processing strategy, updating the initial road network according to the updated driving route to obtain a target road network.

[0080] Optionally, in response to the risk handling strategy being a deceleration and avoidance strategy, the virtual object is controlled to avoid obstacles based on the deceleration and avoidance strategy to obtain a new obstacle avoidance driving route, and the obstacle avoidance route is determined as the updated driving route.

[0081] In the embodiment of the present disclosure, in a scenario where the virtual object executes a deceleration and avoidance strategy, the driving route of the virtual object changes. In this scenario, the driving route of the virtual object based on the signal generated by the execution of the deceleration and avoidance strategy can be determined as the obstacle avoidance driving route obtained by the virtual object to avoid obstacles, wherein the obstacle avoidance driving route can be determined as the updated driving route.

[0082] Optionally, the initial road network is updated according to the updated driving route to obtain an updated candidate road network.

[0083] In the disclosed embodiment, the current initial driving route of the virtual object can be modified based on the updated driving route to achieve an update of the initial road network, wherein the initial driving route can be deleted from the initial road network, or the current obstacle avoidance section can be deleted from the initial road network based on the updated driving route, thereby achieving an update of the initial road network and determining the updated road network as a candidate road network.

[0084] Optionally, a third simulation environment of the virtual object is obtained, and the candidate road network is three-dimensionally projected onto the third simulation environment, so as to obtain absolute coordinate information of each candidate driving route in the candidate road network through the third simulation environment.

[0085] In the embodiment of the present disclosure, the first simulation environment and the second simulation environment may be two-dimensional simulation environments. In this scenario, the candidate road network obtained in the two-dimensional simulation environment may be projected into a three-dimensional simulation environment.

[0086] Optionally, a corresponding three-dimensional third simulation environment can be constructed based on the actual driving site of the unmanned vehicle corresponding to the virtual object, and the candidate road network in the two-dimensional simulation environment can be projected into the three-dimensional third simulation environment based on the three-dimensional projection method in related technologies.

[0087] In this scenario, based on the three-dimensional projection, the virtual simulation results of each candidate driving route of the candidate road network in the three-dimensional third simulation environment can be obtained. It can be understood that based on the three-dimensional projection, the virtual simulation results of each candidate driving route in the three-dimensional environment such as uphill and downhill can be obtained, thereby obtaining a safe driving route for the virtual object in the three-dimensional simulation environment.

[0088] Among them, after the candidate road network is projected into the third simulation environment, the absolute coordinate information of each candidate driving route in the third simulation environment can be obtained, and the road network can be constructed according to the absolute coordinate information of each candidate driving route to obtain the target road network of the unmanned vehicle in the actual driving site.

[0089] Optionally, based on a route construction algorithm in related technologies, the absolute coordinate information of each candidate driving route may be processed algorithmically to obtain a target driving route for safe driving of each candidate driving route in a three-dimensional environment.

[0090] Furthermore, the road network composed of the target driving routes is determined as the target road network for the unmanned vehicle corresponding to the virtual object to drive safely in the actual driving field corresponding to the third simulation environment.

[0091] It should be noted that when the virtual object is performing virtual driving, it may switch the driving route in the initial road network. In response to the virtual object switching the driving route, the first virtual driving strategy of the virtual object on the initial driving route is obtained, and it is identified whether the first virtual driving strategy matches the new switched driving route.

[0092] In the disclosed embodiment, when the virtual object switches from the initial driving route to another route in the initial road network, the switched driving route can be determined as the new switched driving route of the virtual object.

[0093] In this scenario, it is necessary to obtain the first virtual driving strategy of the virtual object, which is a virtual driving strategy adopted when the virtual object drives on the initial driving route. It can be identified whether the direction strategy and the speed strategy in the first virtual driving strategy can enable the virtual object to drive virtually on the switching driving route, thereby identifying whether the first virtual driving strategy matches the switching driving route.

[0094] Optionally, in response to identifying that the first virtual driving strategy does not match the switching driving route, a second virtual driving strategy of the virtual object is acquired according to the switching driving route, and the virtual object is controlled to perform virtual driving on the switching driving route based on the second virtual driving strategy.

[0095] When it is identified that the first virtual driving strategy does not match the new driving route after switching, the virtual driving strategy of the driving route after switching can be obtained.

[0096] Optionally, when the virtual object switches its driving route and enters a new switching driving route, it is necessary to obtain the route identification of the switching driving route, and based on the route identification, obtain the speed information, acceleration information, deceleration information, deviation data information, and angular acceleration information and angular deceleration information of the wheel angles required for the virtual object to drive on the new switching driving route from relevant data sets, so as to obtain the virtual driving strategy required for the virtual object to perform virtual driving on the new switching driving route, which is determined as the second virtual driving strategy.

[0097] Further, the virtual object is controlled to switch from the first virtual driving strategy to the second virtual driving strategy, and continues to perform virtual driving on the new switched driving route based on the second virtual driving strategy.

[0098] The road network acquisition method for an unmanned vehicle proposed in the present invention identifies whether there is a risk between a virtual object and a virtual obstacle through a dynamic distance parameter, and updates the initial road network to obtain a candidate road network according to a new route generated when the virtual object performs risk processing, thereby reducing the difficulty and cost of acquiring the candidate road network, and performing three-dimensional projection on the candidate road network to obtain a target road network, thereby improving the acquisition precision and accuracy of the target road network. Compared with the test driving in the actual driving field in the related technology, the acquisition cost and risk of the target road network for the safe driving of the unmanned vehicle are reduced, the influence of various obstacles in the actual driving field on the safe driving of the unmanned vehicle is reduced, and the safety of the unmanned vehicle is improved.

[0099] In the above embodiment, regarding the construction of the second virtual environment, it can be combined with Figure 3 understand, Figure 3 FIG. 1 is a flow chart of a method for acquiring a driving road network of an unmanned vehicle according to another embodiment of the present disclosure. Figure 3 As shown, the method includes:

[0100] S301, obtaining a virtual object of the unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle.

[0101] Optionally, a candidate modeling data set of an actual driving site is collected, wherein the candidate modeling data set includes at least lidar data, real-time dynamic data, and inertial measurement unit data of the actual driving site;

[0102] In the disclosed embodiment, data can be collected on the actual driving site of the unmanned vehicle according to a preset data collection device, and the collected data used for simulated environment modeling can be determined as candidate modeling data of the actual driving site, thereby obtaining a candidate modeling data set composed of the candidate modeling data.

[0103] Among them, the lidar data, real-time kinematic data (Real-Time Kinematic Data, RTK Data) and inertial measurement unit data (InertialMeasurement Unit Data, IMU Data) of the actual driving site can be collected through the preset data collection equipment, and the collected lidar data, IMU data and RTK data can be determined as candidate modeling data of the actual driving site, so as to obtain the corresponding candidate modeling data set.

[0104] Optionally, data preprocessing is performed on the candidate modeling data set to obtain a target modeling data set of the actual driving site, and a first simulation environment of the actual driving site is acquired based on the target modeling data set.

[0105] In the embodiments of the present disclosure, data preprocessing can be performed on the candidate modeling data set based on preset data preprocessing means, wherein the candidate modeling data set can be preprocessed based on preprocessing methods such as time synchronization, filtering, denoising and alignment processing, and the data set obtained after preprocessing can be determined as the target modeling data set.

[0106] Furthermore, based on the modeling method in the relevant technology, the modeling processing of the simulation environment for virtual simulation is performed based on the target modeling data set, and the simulation environment for virtual simulation of the unmanned vehicle obtained by modeling is determined as the first simulation environment of the actual driving site.

[0107] It should be noted that the first simulation environment may be a two-dimensional simulation environment or a three-dimensional simulation environment, which is not specifically limited here.

[0108] As an example, based on the Extended Kalman Filter with External Speed ​​Estimation (ES-EKF) algorithm, the lidar three-dimensional point cloud, RTK absolute coordinate information and IMU high-frequency inertial data included in the target modeling data set can be fused to obtain a virtual simulation environment corresponding to the actual driving site.

[0109] S302, obtaining a set of virtual obstacles in the first simulation environment, and controlling the movement of each virtual obstacle to obtain a second simulation environment.

[0110] Optionally, a subset of high-speed regularly moving obstacles and a subset of low-speed randomly moving obstacles in the first simulation environment are obtained to obtain a set of virtual obstacles in the first simulation environment;

[0111] In the disclosed embodiment, a preset image acquisition device can be used to collect data on the daily running trajectory, moving speed, acceleration and related position information of each obstacle in the actual driving site, and some obstacles with faster speeds and regular moving trajectories in the actual driving site can be extracted based on the collected data to form a high-speed regularly moving obstacle subset.

[0112] Furthermore, some obstacles with low moving speed and random moving trajectories in the actual driving field are extracted to form a low-speed random moving obstacle subset.

[0113] Furthermore, a set consisting of the projection objects of the high-speed regularly moving obstacle subset and the low-speed randomly moving obstacle subset in the first simulation environment is determined as a virtual obstacle set in the first simulation environment.

[0114] Optionally, obtaining moving direction information, moving speed information, and obstacle position information of each virtual obstacle in the virtual obstacle set to obtain an obstacle movement strategy for each virtual obstacle;

[0115] Among them, the information of each obstacle in the actual driving field can be collected according to the data collection method in the relevant technology, and the moving direction information, moving speed information and obstacle position information of each obstacle can be obtained through the collected data, which can be used as the moving direction information, moving speed information and obstacle position information of the virtual obstacles projected by each obstacle in the first simulation environment.

[0116] Furthermore, an obstacle movement data set preset in the first simulation environment may be obtained, and the movement direction information, movement speed information and obstacle position information of each virtual obstacle may be determined from the obstacle movement data set.

[0117] As an example, the information of each obstacle in the actual driving field can be collected, and the obstacle trajectory of each obstacle, the operation information of the obstacle on different roads, etc. can be obtained through deep learning model detection. The operation information may include speed statistics, acceleration / deceleration statistics, steering statistics, and trajectory deviation data. Furthermore, the corresponding speed data set ρ is obtained according to the peak value of the statistical results. speed , acceleration / deceleration data set ρ acc , turn to the dataset ρ dir and trajectory deviation dataset ρ pos .

[0118] Furthermore, the moving direction information, moving speed information and obstacle position information of each virtual obstacle are determined from the data set.

[0119] It should be noted that based on the speed data set ρ speed , acceleration / deceleration data set ρ acc , turn to the dataset ρ dir and trajectory deviation dataset ρ pos A constraint data set of the first simulation environment may be obtained, and in this scenario, information related to the speed and direction of the virtual object may be obtained from the constraint data set.

[0120] Further, based on the acquired relevant information, a movement strategy required for each virtual obstacle to move in the first simulation environment is generated and marked as an obstacle movement strategy of each virtual obstacle.

[0121] Optionally, based on the obstacle movement strategy, each virtual obstacle is controlled to move in the first simulation environment to obtain a second simulation environment in an obstacle movement state.

[0122] In the embodiment of the present disclosure, based on a method for controlling the movement of various elements in a preset virtual simulation environment, the movement process of each virtual obstacle can be started based on an obstacle movement strategy to control the movement of each virtual obstacle in a first simulation environment, and the simulation environment after each virtual obstacle starts to move is determined as a second simulation environment.

[0123] The road network acquisition method for unmanned vehicles proposed in the present invention improves the matching degree between the first simulation environment and the actual driving site by collecting the modeling data set of the actual driving site, and provides a precise virtual simulation environment and accurate data support for the virtual simulation of safe driving of the virtual object corresponding to the unmanned vehicle, thereby improving the accuracy and precision of the target road network acquired based on the virtual simulation.

[0124] An embodiment of the present disclosure further proposes a road network acquisition device for an unmanned vehicle. Since the road network acquisition device for an unmanned vehicle proposed in the embodiment of the present disclosure corresponds to the road network acquisition methods for unmanned vehicles proposed in the above-mentioned embodiments, the implementation methods of the above-mentioned road network acquisition methods for unmanned vehicles are also applicable to the road network acquisition device for an unmanned vehicle proposed in the embodiment of the present disclosure, and will not be described in detail in the following embodiments.

[0125] Figure 4 FIG. 1 is a schematic diagram of a road network acquisition device for an unmanned vehicle according to an embodiment of the present disclosure. Figure 4 As shown, the road network acquisition device 400 of the unmanned vehicle includes a first acquisition module 41, a second acquisition module 42, a virtual driving module 43, a third acquisition module 44 and a fourth acquisition module 45, wherein:

[0126] A first acquisition module 41 is used to acquire a virtual object of the unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle;

[0127] The second acquisition module 42 is used to acquire a set of virtual obstacles in the first simulation environment and control the movement of each virtual obstacle to obtain a second simulation environment;

[0128] A virtual driving module 43, used for obtaining an initial road network of a virtual object, and controlling the virtual object to virtually drive in a second simulation environment based on the initial road network;

[0129] A third acquisition module 44 is used to acquire dynamic distance parameters between the virtual object and each virtual obstacle during the virtual driving of the virtual object;

[0130] The fourth acquisition module 45 is used to update the initial road network according to the dynamic distance parameter to obtain the target road network of the unmanned vehicle in the actual driving field.

[0131] In the disclosed embodiment, the fourth acquisition module 45 is also used to: for any dynamic distance parameter, in response to identifying that there is a risk in the virtual driving process of the virtual object based on the dynamic distance parameter, obtain a risk handling strategy for the virtual object; in response to obtaining an updated driving route of the initial driving route based on the risk handling strategy, update the initial road network according to the updated driving route to obtain a target road network.

[0132] In the embodiment of the present disclosure, the fourth acquisition module 45 is also used to: in response to the dynamic distance parameter being less than a preset collision distance threshold, obtain the deceleration and avoidance strategy of the virtual object as a risk handling strategy, wherein the deceleration and avoidance strategy is obtained based on a preset deviation distance and offset direction of the virtual object; in response to the dynamic distance parameter being greater than or equal to a preset collision distance threshold and less than a preset safety distance threshold, obtain the deceleration strategy of the virtual object as a risk handling strategy.

[0133] In the disclosed embodiment, the fourth acquisition module 45 is further used to: in response to the dynamic distance parameter being greater than or equal to the safety distance threshold, keep the virtual object moving based on the current driving state.

[0134] In the disclosed embodiment, the fourth acquisition module 45 is also used for: in response to the risk handling strategy being a deceleration and avoidance strategy, controlling the virtual object to avoid obstacles based on the deceleration and avoidance strategy to obtain a new obstacle avoidance driving route, and determining the obstacle avoidance route as an updated driving route; updating the initial road network according to the updated driving route to obtain an updated candidate road network; acquiring a third simulation environment of the virtual object, and projecting the candidate road network in three dimensions to the third simulation environment, so as to obtain the absolute coordinate information of each candidate driving route in the candidate road network through the third simulation environment, and constructing a road network according to the absolute coordinate information of each candidate driving route to obtain the target road network of the unmanned vehicle in the actual driving site.

[0135] In the disclosed embodiment, the virtual driving module 43 is further used to: obtain the current initial driving route of the virtual object from the initial road network; control the virtual object to virtually drive on the initial driving route based on a first virtual driving strategy corresponding to the initial driving route, wherein the first virtual driving strategy includes at least a speed strategy and a direction strategy of the virtual object on the initial driving route.

[0136] In the disclosed embodiment, the virtual driving module 43 is further used to: obtain a first speed of the virtual object and a second speed of each virtual obstacle; and obtain a dynamic distance parameter between the virtual object and each virtual obstacle according to the first speed and the second speed.

[0137] In the disclosed embodiment, the second acquisition module 42 is further used to: acquire a subset of high-speed regularly moving obstacles and a subset of low-speed randomly moving obstacles in the first simulation environment to obtain a virtual obstacle set in the first simulation environment; acquire moving direction information, moving speed information and obstacle position information of each virtual obstacle in the virtual obstacle set to obtain an obstacle movement strategy of each virtual obstacle; and control each virtual obstacle to move in the first simulation environment based on the simulated movement strategy to obtain a second simulation environment in an obstacle movement state.

[0138] In the disclosed embodiment, the first acquisition module 41 is also used to: collect a candidate modeling data set of the actual driving site, wherein the candidate modeling data set includes at least lidar data, real-time dynamic data and inertial measurement unit data of the actual driving site; perform data preprocessing on the candidate modeling data set to obtain a target modeling data set of the actual driving site; and obtain a first simulation environment of the actual driving site based on the target modeling data set.

[0139] In the disclosed embodiment, the virtual driving module 43 is further used to: in response to the virtual object switching the driving route, obtain a first virtual driving strategy of the virtual object on the initial driving route, and identify whether the first virtual driving strategy matches the new switching driving route; in response to identifying that the first virtual driving strategy does not match the switching driving route, obtain a second virtual driving strategy of the virtual object according to the switching driving route, and control the virtual object to perform virtual driving on the switching driving route based on the second virtual driving strategy.

[0140] The road network acquisition device for an unmanned vehicle proposed in the present disclosure obtains a second simulation environment by moving each virtual obstacle in a first simulation environment, realizes virtual driving simulation of the virtual object corresponding to the unmanned vehicle in the second simulation environment, and obtains a target road network according to the result of the virtual driving simulation. Compared with the related art in which the target road network is obtained by test driving in an actual driving field, the acquisition cost and risk of the target road network for safe driving of the unmanned vehicle are reduced. The target road network is obtained by the second simulation environment composed of each virtual obstacle in a moving state, which reduces the influence of each obstacle in the actual driving field on the safe driving of the unmanned vehicle, improves the accuracy of the acquisition of the target road network of the unmanned vehicle in the actual driving field, improves the safety of the unmanned vehicle driving based on the target road network, reduces the probability of abnormal situation in which the unmanned vehicle changes lanes due to driving safety being affected, shortens the driving time for the unmanned vehicle to reach the operating area, improves the operating efficiency of the unmanned vehicle, thereby improving the safety, stability and operating efficiency of the project to which the unmanned vehicle belongs, and optimizes the driving method of the unmanned vehicle.

[0141] According to an embodiment of the present disclosure, the present disclosure also proposes an electronic device, a readable storage medium, and a computer program product.

[0142] Figure 5A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0143] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 509 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0144] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 506, such as various types of displays, speakers, etc.; a storage unit 509, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0145] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the road network acquisition method for an unmanned vehicle. For example, in some embodiments, the road network acquisition method for an unmanned vehicle may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 509. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the road network acquisition method for the unmanned vehicle described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to execute the road network acquisition method for the unmanned vehicle in any other appropriate manner (for example, by means of firmware).

[0146] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0147] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be presented to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, implements the functions / operations specified in the flow chart and / or block diagram. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0148] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0149] To propose interactions with a user account, the systems and techniques described herein may be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user account; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user account can propose input to the computer. Other types of devices may also be used to propose interactions with the user account; for example, feedback proposed to the user account may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user account may be received in any form (including acoustic input, voice input, or tactile input).

[0150] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user account computer with a graphical user account interface or a web browser through which a user account can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0151] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0152] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0153] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for obtaining a driving road network of an unmanned vehicle, wherein: The method comprises: Acquire a virtual object of an unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle; Acquire a set of virtual obstacles in the first simulation environment, and control the movement of each virtual obstacle to obtain a second simulation environment; Acquire an initial road network of the virtual object, and control the virtual object to virtually travel in the second simulation environment based on the initial road network; During the virtual driving of the virtual object, obtaining dynamic distance parameters between the virtual object and each virtual obstacle; The initial road network is updated according to the dynamic distance parameter to obtain a target road network for the unmanned vehicle in the actual driving field.

2. The method according to claim 1, wherein: The updating of the initial road network according to the dynamic distance parameter to obtain a target road network of the unmanned vehicle in the actual driving site includes: For any dynamic distance parameter, in response to identifying that a virtual driving process of the virtual object has a risk according to the dynamic distance parameter, obtaining a risk handling strategy for the virtual object; In response to obtaining an updated driving route of the initial driving route based on the risk handling strategy, the initial road network is updated according to the updated driving route to obtain the target road network.

3. The method according to claim 2, wherein: For any dynamic distance parameter, in response to identifying that a risk exists in the virtual driving process of the virtual object according to the dynamic distance parameter, obtaining a risk handling strategy for the virtual object includes: In response to the dynamic distance parameter being less than a preset collision distance threshold, obtaining a deceleration and avoidance strategy of the virtual object as the risk handling strategy, wherein the deceleration and avoidance strategy is obtained based on a preset deviation distance and deviation direction of the virtual object; In response to the dynamic distance parameter being greater than or equal to a preset collision distance threshold and less than a preset safety distance threshold, a deceleration strategy of the virtual object is acquired as the risk handling strategy.

4. The method according to claim 3, wherein: The method further comprises: In response to the dynamic distance parameter being greater than or equal to the safety distance threshold, the virtual object is kept moving based on a current driving state.

5. The method according to claim 3, wherein: In response to obtaining an updated driving route of the initial driving route based on the risk processing strategy, the initial road network is updated according to the updated driving route to obtain the target road network, including: In response to the risk handling strategy being the deceleration and avoidance strategy, controlling the virtual object to avoid obstacles based on the deceleration and avoidance strategy to obtain a new obstacle avoidance driving route, and determining the obstacle avoidance route as the updated driving route; Updating the initial road network according to the updated driving route to obtain an updated candidate road network; Acquire a third simulation environment of the virtual object, and project the candidate road network three-dimensionally onto the third simulation environment, so as to acquire absolute coordinate information of each candidate driving route in the candidate road network through the third simulation environment; A road network is constructed according to the absolute coordinate information of each candidate driving route to obtain the target road network of the unmanned vehicle in the actual driving site.

6. The method according to claim 1, wherein: The obtaining of the initial road network of the virtual object, and controlling the virtual object to virtually travel in the second simulation environment based on the initial road network, comprises: Acquiring a current initial driving route of the virtual object from the initial road network; The virtual object is controlled to virtually drive on the initial driving route based on a first virtual driving strategy corresponding to the initial driving route, wherein the first virtual driving strategy at least includes a speed strategy and a direction strategy of the virtual object on the initial driving route.

7. The method according to claim 6, wherein: The step of obtaining the dynamic distance parameters between the virtual object and each virtual obstacle during the virtual driving of the virtual object includes: Acquire a first speed of the virtual object and a second speed of each virtual obstacle; A dynamic distance parameter between the virtual object and each virtual obstacle is obtained according to the first speed and the second speed.

8. The method according to claim 1, wherein: The step of obtaining a set of virtual obstacles in the first simulation environment and controlling the movement of each virtual obstacle to obtain a second simulation environment includes: Acquire a subset of high-speed regularly moving obstacles and a subset of low-speed randomly moving obstacles in the first simulation environment to obtain a set of virtual obstacles in the first simulation environment; Acquire moving direction information, moving speed information, and obstacle position information of each virtual obstacle in the virtual obstacle set to obtain an obstacle movement strategy for each virtual obstacle; Based on the simulated movement strategy, each virtual obstacle is controlled to move in the first simulation environment to obtain the second simulation environment in an obstacle moving state.

9. The method according to claim 1, wherein: The step of obtaining a virtual object of the unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle includes: Collecting a candidate modeling data set of the actual driving site, wherein the candidate modeling data set includes at least laser radar data, real-time dynamic data, and inertial measurement unit data of the actual driving site; Performing data preprocessing on the candidate modeling data set to obtain a target modeling data set for the actual driving site; The first simulation environment of the actual driving site is acquired according to the target modeling data set.

10. The method according to claim 1, wherein: The method further comprises: In response to the virtual object switching the driving route, acquiring a first virtual driving strategy of the virtual object on the initial driving route, and identifying whether the first virtual driving strategy matches the new switched driving route; In response to identifying that the first virtual driving strategy does not match the switching driving route, a second virtual driving strategy of the virtual object is acquired according to the switching driving route, and the virtual object is controlled to perform virtual driving on the switching driving route based on the second virtual driving strategy.

11. A road network acquisition device for an unmanned vehicle, wherein: The device comprises: A first acquisition module, used to acquire a virtual object of an unmanned vehicle and a first simulation environment corresponding to an actual driving site of the unmanned vehicle; A second acquisition module, used to acquire a set of virtual obstacles in the first simulation environment, and control the movement of each virtual obstacle to obtain a second simulation environment; A virtual driving module, used for obtaining an initial road network of the virtual object, and controlling the virtual object to virtually drive in the second simulation environment based on the initial road network; A third acquisition module, used for acquiring dynamic distance parameters between the virtual object and each virtual obstacle during the virtual driving of the virtual object; The fourth acquisition module is used to update the initial road network according to the dynamic distance parameter to obtain the target road network of the unmanned vehicle in the actual driving site.

12. The device according to claim 11, wherein The fourth acquisition module is further used for: For any dynamic distance parameter, in response to identifying that a virtual driving process of the virtual object has a risk according to the dynamic distance parameter, obtaining a risk handling strategy for the virtual object; In response to obtaining an updated driving route of the initial driving route based on the risk handling strategy, the initial road network is updated according to the updated driving route to obtain the target road network.

13. The device according to claim 12, wherein: The fourth acquisition module is further used for: In response to the dynamic distance parameter being less than a preset collision distance threshold, obtaining a deceleration and avoidance strategy of the virtual object as the risk handling strategy, wherein the deceleration and avoidance strategy is obtained based on a preset deviation distance and deviation direction of the virtual object; In response to the dynamic distance parameter being greater than or equal to a preset collision distance threshold and less than a preset safety distance threshold, a deceleration strategy of the virtual object is acquired as the risk handling strategy.

14. The device according to claim 13, wherein: The fourth acquisition module is further used for: In response to the dynamic distance parameter being greater than or equal to the safety distance threshold, the virtual object is kept moving based on a current driving state.

15. The device according to claim 13, wherein: The fourth acquisition module is further used for: In response to the risk handling strategy being the deceleration and avoidance strategy, controlling the virtual object to avoid obstacles based on the deceleration and avoidance strategy to obtain a new obstacle avoidance driving route, and determining the obstacle avoidance route as the updated driving route; Updating the initial road network according to the updated driving route to obtain an updated candidate road network; Acquire a third simulation environment of the virtual object, and project the candidate road network three-dimensionally onto the third simulation environment, so as to acquire absolute coordinate information of each candidate driving route in the candidate road network through the third simulation environment; A road network is constructed according to the absolute coordinate information of each candidate driving route to obtain the target road network of the unmanned vehicle in the actual driving site.

16. The device according to claim 11, wherein The virtual driving module is further used for: Acquiring a current initial driving route of the virtual object from the initial road network; The virtual object is controlled to virtually drive on the initial driving route based on a first virtual driving strategy corresponding to the initial driving route, wherein the first virtual driving strategy at least includes a speed strategy and a direction strategy of the virtual object on the initial driving route.

17. The device according to claim 16, wherein: The virtual driving module is further used for: Acquire a first speed of the virtual object and a second speed of each virtual obstacle; A dynamic distance parameter between the virtual object and each virtual obstacle is obtained according to the first speed and the second speed.

18. The device according to claim 11, wherein The second acquisition module is further used for: Acquire a subset of high-speed regularly moving obstacles and a subset of low-speed randomly moving obstacles in the first simulation environment to obtain a set of virtual obstacles in the first simulation environment; Acquire moving direction information, moving speed information, and obstacle position information of each virtual obstacle in the virtual obstacle set to obtain an obstacle movement strategy for each virtual obstacle; Based on the simulated movement strategy, each virtual obstacle is controlled to move in the first simulation environment to obtain the second simulation environment in an obstacle moving state.

19. The device according to claim 11, wherein: The first acquisition module is further used for: Collecting a candidate modeling data set of the actual driving site, wherein the candidate modeling data set includes at least laser radar data, real-time dynamic data, and inertial measurement unit data of the actual driving site; Performing data preprocessing on the candidate modeling data set to obtain a target modeling data set for the actual driving site; The first simulation environment of the actual driving site is acquired according to the target modeling data set.

20. The device according to claim 11, wherein The virtual driving module is further used for: In response to the virtual object switching the driving route, acquiring a first virtual driving strategy of the virtual object on the initial driving route, and identifying whether the first virtual driving strategy matches the new switched driving route; In response to identifying that the first virtual driving strategy does not match the switching driving route, a second virtual driving strategy of the virtual object is acquired according to the switching driving route, and the virtual object is controlled to perform virtual driving on the switching driving route based on the second virtual driving strategy.

21. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

22. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-10.

23. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images