A cloud control automatic driving fusion decision method and device

Abstract

The embodiment of the specification discloses a cloud control automatic driving fusion decision method and device, based on current driving state information and driving environment information of each connected vehicle, predicting the state of each connected vehicle after executing driving decision, obtaining a prediction result; generating driving decision instructions of each connected vehicle according to the prediction result, the current driving state information and the driving environment information; if the connected vehicle currently still has other driving decision instructions different from the driving decision instructions in the applicable scene function, then the driving decision instructions and the other driving decision instructions are fused to generate a fusion driving decision instruction; the fusion driving decision instruction is sent to the connected vehicle, so that the connected vehicle executes the fusion driving decision instruction. The cloud control automatic driving fusion decision method provided by the embodiment of the specification can cooperate with each connected vehicle to perform adaptive fusion of driving decision instructions, and can more efficiently and safely provide driving assistance for each connected vehicle.

Description

TECHNICAL FIELD

[0001] Embodiments of the present specification relate to the technical field of automatic driving, and in particular to a cloud control automatic driving fusion decision method and device. BACKGROUND

[0002] Traditional traffic infrastructure, traffic service capability and traffic management capability have been unable to cope with increasingly severe and complex traffic problems. Intelligent connected vehicle (ICV) technology based on vehicle networking and automatic driving technology can effectively improve traffic problems and improve traffic efficiency. Currently, a cloud control system uses vehicle-road fusion perception and cloud collaborative decision through the fusion of vehicle-road cloud capabilities, uses various intelligent connected driving functions to assist intelligent connected vehicles, and improves safety, efficiency and other performance, such as vehicle reverse warning, vehicle abnormal low speed warning, blind area collaborative perception, intersection merging coordination, signal light green wave band passage induction (GLOSA), lane changing coordination, etc. These functions may simultaneously act on the same intelligent connected vehicle to form a composite scenario. These composite scenarios need to be coordinated to provide the vehicle with the optimal unique assistance and avoid ambiguity.

[0003] Multiple collaborative decision functions may simultaneously issue decision instructions to vehicles for different independent scenarios. For an autonomous driving vehicle, in order to generate a decision for the vehicle to execute, multiple scenario collaborative decision instructions need to be fused to form a unique decision, which needs to be fused at the cloud or the vehicle. Decision fusion needs to be adaptive to different numbers and scenarios of decisions. The decision on the autonomous driving vehicle is based on a unified fusion of perception results at the vehicle or a decision based on multiple situations and rules to handle multiple situations. SUMMARY

[0004] Therefore, embodiments of the present specification provide a cloud control automatic driving fusion decision method and device to solve the problem that the existing automatic driving decision fusion cannot be applied to the fusion of various collaborative decisions on a cloud control platform.

[0005] Embodiments of the present specification adopt the following technical solutions:

[0006] The present specification provides a cloud control automatic driving fusion decision method, comprising:

[0007] Based on the current driving state information and the driving environment information of each connected vehicle, the state of each connected vehicle after executing a driving decision is predicted to obtain a prediction result;

[0008] According to the prediction result, the current driving state information and the driving environment information, a driving decision instruction for each connected vehicle is generated;

[0009] If there is another driving decision instruction different from the driving decision instruction applicable scene function of the connected vehicle at present, the driving decision instruction and the other driving decision instruction are fused to generate a fused driving decision instruction;

[0010] The fused driving decision instruction is sent to the connected vehicle, so that the connected vehicle executes the fused driving decision instruction.

[0011] Further, based on the current driving state information and the driving environment information of each connected vehicle, the state of each connected vehicle after executing the driving decision is predicted, including:

[0012] Based on the current driving state information and the driving environment information, it is judged whether the connected vehicle is safe after executing the driving decision, and a judgment result is obtained;

[0013] If the judgment result is safe, at least one feasible driving decision is generated;

[0014] The state of the connected vehicle after executing at least one feasible driving decision is predicted respectively, and a prediction result is obtained.

[0015] Further, the driving decision instruction of each connected vehicle is generated according to the prediction result, the current driving state information and the driving environment information, including:

[0016] The decision suggestion is generated according to the prediction result, the current driving state information and the driving environment information;

[0017] The decision suggestion is converted into an executable driving decision instruction.

[0018] Further, the method can further include:

[0019] After the driving decision instruction is generated, the driving decision instruction is cached.

[0020] Further, the driving decision instruction is cached, including:

[0021] The driving decision instruction on each lane where the connected vehicle is currently located and the driving decision instruction on other lanes are cached respectively.

[0022] Further, the method can further include:

[0023] If the driving decision instruction includes decision cancellation information, the corresponding driving decision instruction in the cache is cleared according to the decision cancellation information before the driving decision instruction and the other driving decision instruction are fused.

[0024] Further, the driving decision instruction and the other driving decision instruction are fused, including:

[0025] The driving decision instructions currently cached by the connected vehicles in the same lane are fused.

[0026] Or,

[0027] The driving decision instructions currently cached by the connected vehicles in different lanes are fused.

[0028] Further, if the driving decision instructions currently cached by the connected vehicles in different lanes are fused, the other driving decision instruction includes the driving decision instruction cached by the connected vehicle in the current lane and the driving decision instruction cached by the connected vehicle in the other lane.

[0029] The driving decision instruction and the other driving decision instruction are fused to generate a fused driving decision instruction, including:

[0030] The driving decision instruction cached by the connected vehicle in the current lane and the driving decision instruction cached by the connected vehicle in the other lane are combined according to a preset decision urgency to obtain a combination result.

[0031] The fused driving decision instruction is generated according to the combination result.

[0032] Further, the driving decision instruction cached by the connected vehicle in the current lane and the driving decision instruction cached by the connected vehicle in the other lane are combined according to a preset decision urgency, including:

[0033] The speed, acceleration, lateral feasible region along the road, longitudinal feasible region along the road and feasible range of the target lane of each driving decision instruction are calculated by intersection, so as to obtain the fused driving decision instruction according to the calculation result.

[0034] The embodiments of the present specification also provide a cloud control automatic driving fusion decision device, including:

[0035] A prediction module predicts the state of each connected vehicle after executing a driving decision based on the current driving state information and driving environment information of each connected vehicle to obtain a prediction result.

[0036] An instruction generation module generates a driving decision instruction of each connected vehicle according to the prediction result, the current driving state information and the driving environment information.

[0037] A fusion module fuses the driving decision instruction and the other driving decision instruction to generate a fused driving decision instruction if the connected vehicle currently has other driving decision instructions different from the driving decision instruction in the applicable scene function.

[0038] send the fusion driving decision instruction to the connected vehicle, so that the connected vehicle executes the fusion driving decision instruction.

[0039] The above at least one technical solution adopted by the embodiments of the present specification can achieve the following beneficial effects:

[0040] By predicting the state of the connected vehicle after executing different driving decisions based on the current driving state information and driving environment information of the connected vehicle, generating the driving decision instruction of the connected vehicle according to the current driving state information and the prediction result, and fusing multiple driving decision instructions of the same vehicle to obtain the fusion driving decision instruction, the connected vehicle can execute the received fusion driving decision instruction.

[0041] In this way, the driving assistance for each connected vehicle can be provided more efficiently and more safely according to the real-time driving state and driving environment of the connected vehicle, considering the predicted state after the connected vehicle executes the feasible driving decision, and the adaptive fusion of the driving decision instructions of each connected vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0042] The accompanying drawings described herein are used to provide further understanding of the embodiments of the present specification, and form a part of the present specification. The illustrative embodiments of the present specification and their descriptions serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0043] Figure 1 A flowchart of a cloud-controlled automatic driving fusion decision method provided by the embodiments of the present specification;

[0044] Figure 2 A general architecture diagram of a cloud-controlled automatic driving fusion decision system provided by the embodiments of the present specification;

[0045] Figure 3 A structural schematic diagram of a feasible region decision fusion module provided by the embodiments of the present specification;

[0046] Figure 4 A flowchart of a cloud-controlled automatic driving fusion decision method provided by the embodiments of the present specification;

[0047] Figure 5 A structural schematic diagram of a cloud-controlled automatic driving fusion decision device provided by the embodiments of the present specification. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments in this specification, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in this specification without creative effort are within the scope of protection of this application.

[0049] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0050] like Figure 1 The diagram shown is a flowchart illustrating a cloud-controlled autonomous driving fusion decision-making method provided in an embodiment of this specification. The executing entity of the cloud-controlled autonomous driving fusion decision-making method provided in this embodiment can be a cloud control platform, a server, or an in-vehicle computing platform, or other entities capable of executing cloud-controlled autonomous driving fusion decision-making software.

[0051] The cloud-controlled autonomous driving fusion decision-making method may include the following steps:

[0052] S101: Based on the current driving status information and driving environment information of each connected vehicle, predict the state of each connected vehicle after executing a driving decision, and obtain the prediction result;

[0053] S102: Generate driving decision instructions for each of the connected vehicles based on the prediction results, the current driving status information, and the driving environment information;

[0054] S103: If the connected vehicle currently has other driving decision instructions that are different from the applicable scenario of the driving decision instruction, then the driving decision instruction and the other driving decision instructions are merged to generate a merged driving decision instruction;

[0055] S104: Send the fusion driving decision command to the connected vehicle, so that the connected vehicle executes the fusion driving decision command.

[0056] In the embodiments of this specification, driving environment information may refer to the environmental information around the location of each connected vehicle, which may include map data, road user status information, road condition information, weather information, traffic information, dynamic events, etc. Among them, dynamic events may refer to events that can affect the driving status of connected vehicles, such as traffic accidents, sudden weather changes, traffic lights, traffic control, abnormal vehicles, vehicle malfunctions, etc., which will not be elaborated here.

[0057] As one application embodiment, for step S101, predicting the state of each connected vehicle after executing a driving decision based on the current driving state information and driving environment information of each connected vehicle may include:

[0058] determine whether the connected vehicle is safe after executing the driving decision based on the current driving state information and the driving environment information, and obtain a determination result;

[0059] if the determination result is safe, generate at least one feasible driving decision;

[0060] respectively predict the state of the connected vehicle after executing at least one feasible driving decision, and obtain a prediction result.

[0061] In the embodiments of the present application, the at least one feasible driving decision generated can refer to the driving decision that the connected vehicle can execute at present under the condition of ensuring the driving safety of the connected vehicle, which can include, for example, a feasible lane-changing target lane, acceleration or deceleration, lateral offset along the road, longitudinal driving distance along the road, etc.

[0062] In this case, there can be more than two feasible driving decisions, for example, left lane-changing or right lane-changing is feasible, and then by predicting the driving state of the connected vehicle after left lane-changing and right lane-changing, the cloud control platform can make reasonable decision suggestions according to the prediction result, for example, by state prediction, it is found that the driving state of left lane-changing is better based on the current driving environment, and then the cloud control platform will give a decision suggestion of left lane-changing according to the prediction result.

[0063] The state of the connected vehicle after executing the driving decision can refer to the driving state of the connected vehicle after executing the driving decision, and can specifically refer to the driving speed, driving position, etc. of the connected vehicle after executing the driving decision.

[0064] For the step S102, generating the driving decision instruction of each connected vehicle according to the prediction result, the current driving state information and the driving environment information can include:

[0065] generating a decision suggestion according to the prediction result, the current driving state information and the driving environment information;

[0066] converting the decision suggestion into an executable driving decision instruction.

[0067] In the embodiments of the present application, the executable driving decision instruction can include specific driving data required by the connected vehicle to execute the driving decision, which can refer to specific driving speed, acceleration, deceleration, lane-changing target lane, lateral offset along the road, longitudinal driving distance along the road, etc. The driving decision instruction can also include the time and space state or range required to execute the instruction, such as executing in the next three minutes, specifically, uniformly changing lanes to the left lane after three minutes, etc.

[0068] Further, the method can further include:

[0069] After the driving decision instruction is generated, the driving decision instruction is cached.

[0070] If the driving decision instruction includes decision cancellation information, the corresponding driving decision instruction in the cache is cleared according to the decision cancellation information before the driving decision instruction and the other driving decision instruction are fused.

[0071] Specifically, the driving decision instruction is cached, which can include:

[0072] For each connected vehicle, the driving decision instruction on the current lane and the driving decision instruction on other lanes are cached respectively.

[0073] In the embodiments of the present disclosure, for each connected vehicle, the cloud control platform uniformly records all single driving decision instructions available for decision fusion in the cache. In addition, the single driving decision instructions for the current lane, the left lanes and the right lanes of each connected vehicle are cached respectively, and each single driving decision instruction has a unique identifier or number in the cache.

[0074] If a single driving decision instruction is newly added, the single driving decision instruction is updated and cached in the corresponding lane cache. Specifically, only the latest single driving decision instruction for the decision suggestion of the same scene function is retained, for example, for the decision suggestion of the emergency braking vehicle in front, only the latest received deceleration or lane changing instruction issued by the scene function is retained in the cache.

[0075] If a single driving decision cancellation instruction is added, the corresponding single driving decision in the cache is deleted.

[0076] Further, the method can further include:

[0077] If the connected vehicle receives decision cancellation information during the execution of the fused driving decision instruction, causing the type of the fused driving decision instruction to change, the execution of the original fused driving decision is cancelled, and a new fused driving decision instruction is calculated and issued as needed.

[0078] In specific application scenarios, as the connected vehicle travels, the driving state and driving environment information of the connected vehicle are constantly changing. Therefore, during the execution of the fused driving decision instruction by the connected vehicle, new single driving decision instructions can be generated due to changes in the driving state and driving environment of the vehicle.

[0079] For example, in the process of executing the acceleration decision instruction by the connected vehicle, the cloud control platform detects that the traffic light in front of the current lane will turn red. If the connected vehicle continues to execute the current acceleration decision instruction, the connected vehicle may not be able to stop in time when it reaches the intersection, and a traffic accident may occur. At this time, the cloud control platform generates a new driving decision instruction for canceling the acceleration decision instruction according to the situation, and sends it to the connected vehicle. The connected vehicle can cancel the currently executed acceleration decision instruction according to the new driving decision instruction, so that the connected vehicle can stop in time when it reaches the intersection.

[0080] Further, for step S103, the fusion of the driving decision instruction and the other driving decision instruction can include:

[0081] fusing the driving decision instructions currently cached in the same lane of the connected vehicle;

[0082] Or,

[0083] fusing the driving decision instructions currently cached in different lanes of the connected vehicle.

[0084] In a specific application scenario, if the driving decision instructions currently cached in the same lane of the connected vehicle are fused, at least two driving decision instructions currently cached in the same lane are combined. Specifically, the intersection of the speed, acceleration, lateral road feasible domain, longitudinal road feasible domain and feasible range of the lane change target lane included in the at least two driving decision instructions can be calculated, and the fusion driving decision instruction can be obtained according to the calculation result, that is, the fusion driving decision instruction that meets different scene single driving decisions at the same time.

[0085] In the embodiments of the present specification, for each driving decision instruction cached in the same lane, the fusion of different decisions in the same lane can be considered in time sequence before fusion, or each driving decision instruction can be executed respectively. After executing a round, if there are still at least two driving decision instructions cached in the lane, the fusion of the driving decision instructions is triggered.

[0086] For example, for the driving decision instruction currently cached in the lane of the connected vehicle, including the acceleration decision instruction, the left lane change decision instruction and the parking decision instruction, the three instructions need to be completed to achieve the final parking operation, that is, acceleration is needed in the current lane, the driving safety between the connected vehicle and the vehicle in the left lane is ensured when changing lanes, and then left lane change and parking are performed. After the respective execution of each driving decision instruction, due to the change of the driving environment, the left lane change decision instruction and the parking decision instruction have not been completed and are cached in the cloud control platform. At this time, the fusion of the driving decision instruction is triggered, that is, the left lane change decision instruction and the parking decision instruction are fused to complete the parking operation more quickly.

[0087] It should be noted that the at least two driving decision instructions currently cached by the same vehicle have different decision functions, such as acceleration decision instructions, lane change decision instructions, parking decision instructions, etc.

[0088] In a specific application scenario, if each driving decision instruction currently cached in different lanes of the connected vehicle is fused, the other driving decision instructions can include the driving decision instruction currently cached in the lane where the connected vehicle is located and the driving decision instruction cached in other lanes.

[0089] In the embodiments of the present specification, if there are other driving decision instructions for the connected vehicle at present, the driving decision instruction and the other driving decision instructions are fused to generate a fused driving decision instruction, which can specifically include:

[0090] According to the preset decision emergency degree, the driving decision instruction currently cached in the lane where the connected vehicle is located and the driving decision instruction cached in other lanes are merged to obtain a merging result;

[0091] The fused driving decision instruction is generated according to the merging result.

[0092] In the embodiments of the present specification, the preset decision emergency degree can refer to the division of the emergency importance degree of each driving decision instruction of different scene functions in advance. In this way, when the driving decision instructions currently cached in the lane of the connected vehicle and other lanes are fused, the driving decision instructions of different scene functions can be fused in order according to the preset decision emergency degree.

[0093] In a specific application scenario, the emergency importance degree of each driving decision instruction of different scene functions can be listed in advance. In this way, when fusion is performed, the emergency importance degree of each driving decision instruction of different scene functions can be determined by looking up the table, which is more clear and explicit.

[0094] Specifically, merging driving decision instructions cached in the lane where the connected vehicle is currently located with driving decision instructions cached in other lanes, based on a preset decision urgency level, may include:

[0095] The intersection of the speed, acceleration, lateral feasible region along the road, longitudinal feasible region along the road, and feasible range of the target lane for lane changing contained in each driving decision command is calculated to obtain the calculation result;

[0096] The fusion driving decision command is generated based on the calculation results.

[0097] The cloud-controlled autonomous driving fusion decision-making method provided in the embodiments of this specification predicts the state of the connected vehicle after executing different driving decisions based on the current driving state information and driving environment information of the connected vehicle. Then, it generates driving decision instructions for each connected vehicle based on the current driving state information and the prediction results, and merges multiple driving decision instructions of the same vehicle to obtain fused driving decision instructions, so that the connected vehicle can execute the received fused driving decision instructions.

[0098] In this way, based on the real-time driving status and driving environment of connected vehicles, and considering the predicted state after the connected vehicles make feasible driving decisions, and by coordinating the adaptive fusion of driving decision commands among connected vehicles, driving assistance can be provided to each connected vehicle more efficiently and safely.

[0099] like Figure 2 The diagram shown is an overall architecture diagram of a cloud-controlled autonomous driving fusion decision-making system provided in an embodiment of this specification.

[0100] The cloud-controlled autonomous driving fusion decision system provided in the embodiments of this specification includes a perception bus 21, a dynamic event recognition module 22, a filtering module 23, an auxiliary decision module 24, a prediction module 25, a safety check module 26, a decision format conversion module 27, and a feasible domain decision fusion module 28.

[0101] The sensing bus 21 can sense the current driving status information and driving environment information of each connected vehicle. The dynamic event recognition module 22 obtains the current driving status information and driving environment information of each connected vehicle from the sensing bus 21, identifies dynamic events from them, and outputs them to the filtering module 23.

[0102] The safety inspection module 25 obtains the current driving status information of the connected vehicle to be served from the perception bus 21, judges the safety of the connected vehicle after executing each driving decision, and outputs the feasible driving decision to the prediction module 25, which then predicts the state of the connected vehicle after executing the feasible driving decision.

[0103] For example, the safety check module 25 can determine the safety of each connected vehicle performing lane changing based on the current driving state information, driving environment information, and map of the connected vehicle, and output the feasible lane changing target lane to the prediction module 25, so that the prediction module 25 predicts the state of left and right lane changing.

[0104] Specifically, for the state prediction after lane changing, the following methods can be included:

[0105] (1) The safety check module 25 queries the map to determine whether the lane where the vehicle is located is a straight lane. If yes, the lane changing can be performed, and the process continues. Otherwise, the lane changing cannot be performed, and the process exits.

[0106] (2) The safety check module 25 queries the map to obtain the road heading of the lane where the vehicle is located, the lane width, and the distance to the stop line or the distance to the curve (or other nodes). If the distance traveled at the current vehicle speed for the lane changing time is less than the distance to the stop line (or other nodes), the lane changing can be performed, and the process continues. Otherwise, the lane changing cannot be performed, and the process exits.

[0107] (3) The safety check module 25 determines whether the state of the road users (motor vehicles, non-motor vehicles, pedestrians, etc.) around the connected vehicle meets the space-time requirements (such as the distance between vehicles and the time distance being greater than a threshold) for the lane changing of the connected vehicle based on the current driving state information, driving environment information, and map of the connected vehicle. If yes, the lane changing can be performed, and the process continues. Otherwise, the lane changing cannot be performed, and the process exits.

[0108] (4) The vehicle position in latitude and longitude is converted into a rectangular coordinate. After the lane changing, the rectangular coordinate position is (x, y). The current coordinate is moved forward along the road heading by the distance traveled at the current vehicle speed for the lane changing time, and then moved by the lane width in the direction perpendicular to the road heading. The direction is determined according to the left or right direction of the lane changing. The rectangular coordinate after the lane changing is converted into latitude and longitude and output.

[0109] The screening module 23 obtains the current driving state information and the predicted state of each connected vehicle, screens out the set of each connected vehicle and its current driving state information and predicted state affected by each dynamic event, and outputs to the auxiliary decision-making module 24.

[0110] The auxiliary decision-making module 24 calculates the decision-making suggestion and decision cancellation information of each connected vehicle based on the dynamic event, the set of current driving state information and predicted state of the connected vehicle, and sends them to the decision format conversion module 27. In specific application scenarios, the auxiliary decision-making module 24 can consider or not consider the output results of the safety check module 26 when calculating the decision-making suggestion and decision cancellation information.

[0111] The decision format conversion module 27 converts the decision-making suggestion and decision cancellation information into a feasible region decision mode in a standard format, generates a driving decision instruction, and outputs it to the feasible region decision fusion module 28.

[0112] The feasible domain decision fusion module 28 can clear the driving decision instructions for the corresponding scenario function in the cache according to the decision cancellation information. For each connected vehicle, when a new single driving decision instruction is input, if there are other driving decision instructions for that connected vehicle in the cache, the fusion function is triggered to calculate the fusion decision for that connected vehicle and send it to that connected vehicle.

[0113] For each vehicle, when a fusion decision is received, it is executed accordingly; when a decision cancellation is received, if a fusion decision is being executed, its execution is cancelled.

[0114] like Figure 3 The diagram shown is the overall architecture of the feasible domain decision fusion module 28 in the above embodiment.

[0115] The feasible domain decision fusion module 28 includes a decision cache module 281, a decision fusion module 282, a fusion output module 283, and an output cancellation module 284.

[0116] After the decision format conversion module 27 generates the driving decision instruction, it outputs it to the decision cache module 281 to cache the driving decision instruction.

[0117] Determine whether each individual driving decision instruction in the cache has completed one round of execution. If so, and there are multiple decision instructions in the current cache, the decision fusion module 282 will be triggered to fuse the multiple individual driving decision instructions in the cache, and the fusion output module 283 will output the fused driving decision instruction.

[0118] If there are no multiple decision instructions in the current cache, and a fusion driving decision instruction has been issued before, then the output cancellation module 284 will output a cancellation instruction.

[0119] In addition, before the fusion output module 283 outputs the fusion driving decision command, it can consider the safety check results of the safety check module 26 to further ensure the safety of the fusion driving decision command.

[0120] like Figure 4 The diagram shown is a flowchart illustrating a specific application of a cloud-controlled autonomous driving fusion decision-making method provided in this embodiment. In this embodiment, lane-changing decision-making is used as a specific application scenario to illustrate the execution process of the fusion decision-making.

[0121] S401: Decision-making is integrated for each lane;

[0122] Specifically, the decision fusion for each lane can refer to merging the single decisions for connected vehicles in their own lane and adjacent lanes. The algorithm is to find the intersection of speed, acceleration, lateral feasible region along the road, longitudinal feasible region along the road, and the feasible range of lane changing.

[0123] S402: Determine whether the single driving decision instruction of the connected vehicle in the current lane is to change lanes, and obtain a first determination result;

[0124] S403: If the first determination result is no, indicating that there is a single driving decision instruction requiring deceleration, in which case the connected vehicle must decelerate to avoid, then output the fusion decision instruction as deceleration in the current lane;

[0125] S404: If the first determination result is yes, determine whether the current lane can change lanes, and obtain a second determination result;

[0126] S405: If the second determination result is yes, determine whether the adjacent lane has an event, and obtain a third determination result;

[0127] S406: If the third determination result is no, output the fusion decision instruction as changing lanes to the adjacent lane;

[0128] S407: If the third determination result is yes, determine whether the reachable adjacent lane is decided to change lanes, and obtain a fourth determination result;

[0129] S408: If the fourth determination result is no, output the fusion decision instruction as normal driving in the current lane;

[0130] S409: If the fourth determination result is yes, compare the emergency importance of the events;

[0131] Specifically, the emergency importance of different events can be weighted and calculated according to the dynamic event type and distance or time interval, or the emergency importance of different dynamic events can be listed in advance, so that the table comparison can be directly performed. The time interval represents the time difference of the vehicle front end reaching the dynamic event position (movable), which can generally be calculated by dividing the vehicle head interval to the dynamic event by the relative speed of the vehicle and the dynamic event.

[0132] S410: Determine whether the adjacent lane is more urgent, and obtain a fifth determination result;

[0133] S411: If the fifth determination result is no, output the fusion decision instruction as changing lanes to the adjacent lane;

[0134] If the fifth determination result is yes, output the fusion decision instruction as normal driving in the current lane.

[0135] S412: If the second determination result is no, output the fusion decision instruction as normal driving in the current lane.

[0136] As Figure 5 shown, based on the same inventive idea, a structural schematic diagram of a cloud control automatic driving fusion decision device provided by an embodiment of the present specification is provided.

[0137] The cloud control automatic driving fusion decision device can comprise:

[0138] A prediction module 501 configured to predict a state of each of the connected vehicles after performing a driving decision based on current driving state information and driving environment information of each of the connected vehicles, and obtain a prediction result;

[0139] An instruction generation module 502 configured to generate a driving decision instruction of each of the connected vehicles according to the prediction result, the current driving state information and the driving environment information;

[0140] A fusion module 503 configured to fuse the driving decision instruction and another driving decision instruction different from the driving decision instruction in a function of a scenario currently existing in the connected vehicle, and generate a fused driving decision instruction;

[0141] A sending module 504 configured to send the fused driving decision instruction to the connected vehicle, so that the connected vehicle performs the fused driving decision instruction.

[0142] Further, the prediction of the state of each of the connected vehicles after performing a driving decision based on the current driving state information and the driving environment information of each of the connected vehicles can comprise:

[0143] Judging whether the connected vehicle is safe after performing a driving decision based on the current driving state information and the driving environment information, and obtaining a judgment result;

[0144] If the judgment result is safe, generating at least one feasible driving decision;

[0145] Respectively predicting a state of the connected vehicle after performing at least one feasible driving decision, and obtaining a prediction result.

[0146] Further, the generation of the driving decision instruction of each of the connected vehicles according to the prediction result, the current driving state information and the driving environment information can comprise:

[0147] Generating a decision suggestion according to the prediction result, the current driving state information and the driving environment information;

[0148] Converting the decision suggestion into an executable driving decision instruction.

[0149] Further, the device can further comprise:

[0150] After generating the driving decision instruction, caching the driving decision instruction.

[0151] Further, the caching of the driving decision instruction can comprise:

[0152] The driving decision instructions on the current lane of each of the connected vehicles and the driving decision instructions on other lanes are respectively cached.

[0153] Further, the apparatus can further include:

[0154] If the driving decision instruction includes decision cancellation information, the corresponding driving decision instruction in the cache is cleared according to the decision cancellation information before the driving decision instruction and the other driving decision instructions are fused.

[0155] Further, the fusion of the driving decision instruction and the other driving decision instructions can include:

[0156] fusing each of the driving decision instructions currently cached in the same lane of the connected vehicle;

[0157] or,

[0158] fusing each of the driving decision instructions currently cached in different lanes of the connected vehicle.

[0159] Further, if each of the driving decision instructions currently cached in different lanes of the connected vehicle is fused, the other driving decision instructions include the driving decision instructions cached in the current lane of the connected vehicle and the driving decision instructions cached in other lanes;

[0160] The fusion of the driving decision instruction and the other driving decision instructions to generate a fused driving decision instruction can include:

[0161] combining the driving decision instructions cached in the current lane of the connected vehicle and the driving decision instructions cached in other lanes according to a preset decision urgency level to obtain a combination result;

[0162] generating the fused driving decision instruction according to the combination result.

[0163] Further, the combination of the driving decision instructions cached in the current lane of the connected vehicle and the driving decision instructions cached in other lanes according to a preset decision urgency level can include:

[0164] calculating the intersection of the speed, acceleration, lateral feasible region along the road, longitudinal feasible region along the road, and feasible range of the target lane in each of the driving decision instructions to obtain the fused driving decision instruction according to the calculation result.

[0165] The cloud control automatic driving fusion decision device provided by the embodiments of the present specification can predict the state of the connected vehicle after executing different driving decisions based on the current driving state information and driving environment information of the connected vehicle, generate driving decision instructions of the connected vehicle according to the current driving state information and the prediction result, and fuse multiple driving decision instructions of the same vehicle to obtain a fused driving decision instruction, so that the connected vehicle can execute the received fused driving decision instruction.

[0166] In this way, the driving state and driving environment of the connected vehicle can be considered in real time, the predicted state of the connected vehicle after executing a feasible driving decision is considered, and adaptive fusion of driving decision instructions of each connected vehicle is performed, so that driving assistance can be more efficiently and safely provided to each connected vehicle.

[0167] In the 1990s, it was relatively easy to distinguish whether an improvement in a technology was a hardware improvement (e.g., an improvement in the circuit structure of a diode, transistor, switch, etc.) or a software improvement (e.g., an improvement in a method flow). However, as technology has evolved, many improvements in method flows today can be considered as direct improvements in hardware circuit structures. Designers almost always obtain the corresponding hardware circuit structures by programming the improved method flows into hardware circuits. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using hardware entity modules. For example, a programmable logic device (PLD) (e.g., a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by user programming of the device. A digital system is "integrated" on a PLD by the designer programming the PLD, rather than by ordering a custom integrated circuit chip from a chip fabricator. Moreover, instead of manually fabricating an integrated circuit chip, this programming is now most often implemented using "logic compiler" software, which is similar to software compilers used in program development, and the original code to be compiled is written in a specific programming language, which is called a hardware description language (HDL), and there are many such languages, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), and the most commonly used are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should be aware that, by simply logically programming a method flow in one of the above-mentioned hardware description languages and programming it into an integrated circuit, a hardware circuit implementing the logical method flow can be easily obtained.

[0168] The controller can be implemented in any suitable way, for example, the controller can take the form of a microprocessor or processor and a computer readable medium storing computer readable program code, such as software or firmware, executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that, in addition to being implemented in pure computer readable program code, the controller can also be implemented to perform the same functions in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers, etc. by logically programming the method steps. Therefore, such a controller can be considered as a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can even be considered as both a software module implementing a method and a structure within a hardware component.

[0169] The systems, apparatuses, modules or units illustrated by the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0170] For the sake of description, the above apparatuses are described in various units by functions respectively. Of course, the functions of each unit can be implemented in the same or multiple software and / or hardware in the implementation of the present application.

[0171] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0172] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0173] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0174] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof. Figure 1 one or more flowcharts and / or blocks in the flowcharts and / or combination thereof.

[0175] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0176] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory. The memory is an example of computer-readable media.

[0177] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0178] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to encompass non-exclusive inclusion, such that processes, methods, articles or devices that comprise a list of elements not only include those elements, but also include other elements not expressly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0179] The present application can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform particular tasks or implement particular abstract data types. The present application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are connected through a communication network. In a distributed computing environment, program modules can be located in both local and remote computer storage media including storage devices.

[0180] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0181] The above merely provides an example of the present application, but is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A cloud-controlled automatic driving fusion decision method, characterized in that, The method comprises the following steps: predicting the state of each connected vehicle after executing driving decision based on the current driving state information and driving environment information of each connected vehicle, and obtaining a prediction result; generating a driving decision instruction for each connected vehicle according to the prediction result, the current driving state information and the driving environment information; if there is another driving decision instruction different from the driving decision instruction in the applicable scene function of the connected vehicle, fusing the driving decision instruction and the other driving decision instruction to generate a fused driving decision instruction; sending the fused driving decision instruction to the connected vehicle, so that the connected vehicle executes the fused driving decision instruction; fusing the driving decision instruction and the other driving decision instruction comprises: fusing each driving decision instruction currently cached in different lanes of the connected vehicle; the other driving decision instruction comprises a driving decision instruction cached in the lane where the connected vehicle is currently located and a driving decision instruction cached in other lanes; fusing the driving decision instruction and the other driving decision instruction to generate a fused driving decision instruction comprises: combining the driving decision instruction cached in the lane where the connected vehicle is currently located and the driving decision instruction cached in other lanes according to a preset decision emergency level to obtain a combination result; generating the fused driving decision instruction according to the combination result; for each connected vehicle, all single driving decision instructions currently available for decision fusion are recorded in the cache, and each connected vehicle has single driving decision instructions for the lane where the vehicle is currently located and each lane below, each single driving decision instruction has a unique identifier or number in the cache; if a new single driving decision instruction is added, the single driving decision instruction is updated and cached in the corresponding lane cache, and only the latest single driving decision instruction for the decision suggestion of the same scene function is retained; if a single driving decision cancellation instruction is added, the corresponding single driving decision in the cache is deleted.

2. The method of claim 1, wherein, The method comprises the following steps: based on the current driving state information and the driving environment information, judging whether the connected vehicle is safe after executing driving decision, and obtaining a judgment result; if the judgment result is safe, generating at least one feasible driving decision; respectively predicting the state of the connected vehicle after executing at least one feasible driving decision, and obtaining a prediction result.

3. The method of claim 1, wherein, The method comprises the following steps: generating a decision suggestion according to the prediction result, the current driving state information and the driving environment information; converting the decision suggestion into an executable driving decision instruction.

4. The method of claim 3, wherein, The method further comprises the following steps: after generating the driving decision instruction, caching the driving decision instruction.

5. The method of claim 4, wherein, The method further comprises the following steps: The driving decision instructions on the current lane of each of the connected vehicles and the driving decision instructions on other lanes are respectively cached.

6. The method of claim 4, wherein, The method further includes: If the driving decision instruction includes decision cancellation information, the corresponding driving decision instruction in the cache is cleared according to the decision cancellation information before the driving decision instruction and the other driving decision instructions are fused.

7. The method of claim 1, wherein The fusing of the driving decision instruction and the other driving decision instructions further includes: The driving decision instructions currently cached in the same lane of the connected vehicle are fused.

8. The method of claim 1, wherein, The driving decision instructions currently cached in the same lane of the connected vehicle and the driving decision instructions currently cached in other lanes are combined according to a preset decision urgency level, including: The speed, acceleration, lateral feasible region along the road, longitudinal feasible region along the road and feasible range of the target lane in each of the driving decision instructions are calculated by intersection, so as to obtain the fused driving decision instruction according to the calculation result.

9. A cloud control automatic driving fusion decision device, characterized in that, including: A prediction module that predicts the state of each of the connected vehicles after the driving decision is executed based on the current driving state information and the driving environment information of each of the connected vehicles, and obtains a prediction result; An instruction generation module that generates the driving decision instruction of each of the connected vehicles according to the prediction result, the current driving state information and the driving environment information; A fusion module that fuses the driving decision instruction and other driving decision instructions different from the driving decision instruction in the applicable scene function of the connected vehicle, and generates a fused driving decision instruction; A sending module that sends the fused driving decision instruction to the connected vehicle, so that the connected vehicle executes the fused driving decision instruction; The fusion module is specifically configured to: The driving decision instructions currently cached in different lanes of the connected vehicle are fused. The other driving decision instructions include the driving decision instructions currently cached in the same lane of the connected vehicle and the driving decision instructions currently cached in other lanes. The fusion module is specifically configured to: The driving decision instructions currently cached in the same lane of the connected vehicle and the driving decision instructions currently cached in other lanes are combined according to a preset decision urgency level, and a combination result is obtained; The fused driving decision instruction is generated according to the combination result; For each connected vehicle, all single driving decision instructions currently running and available for decision fusion are recorded in the cache, and each connected vehicle has single driving decision instructions for the current lane and the following lanes, each single driving decision instruction has a unique identifier or number in the cache; If a single driving decision instruction is added, the single driving decision instruction is updated and cached in the corresponding lane cache, and only the latest single driving decision instruction for the decision suggestion of the same scene function is retained; If a single driving decision cancellation instruction is added, the corresponding single driving decision in the cache is deleted.

Images