Driving information prediction method and device and automatic driving vehicle

Abstract

The invention provides a driving information prediction method and device and an automatic driving vehicle, and relates to the field of automatic driving, in particular to the field of artificial intelligence, and in particular to the technical field of automatic driving and intelligent traffic. According to the specific implementation scheme, the method comprises the steps of determining a first master-slave relationship between a target vehicle and a first obstacle based on a motion parameter of the target vehicle at a current moment, path information of the target vehicle within a first duration, a motion parameter of the first obstacle at the current moment and predicted path information of the first obstacle within the first duration; obtaining first predicted driving information based on the first master-slave relationship; and determining first optimal driving information based on an evaluation result corresponding to the first predicted driving information.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving, in particular to the field of artificial intelligence, and in particular to the field of autonomous driving and intelligent transportation technology. Background Art

[0002] With the advancement of autonomous driving technology, autonomous vehicles are becoming more and more popular. Vehicle driving automation and intelligent transportation systems have become the dominant direction of current and future transportation development. Faced with complex traffic environments, autonomous vehicles need to have a high degree of intelligence and adaptability to cope with emergencies and complex traffic scenarios. Currently, they often rely on real-time data sensed by the vehicle's own sensors to assess collision risks and then predict vehicle driving information. However, with the increase in traffic participants, simply relying on the real-time data obtained by the vehicle's sensors to evaluate and regenerate the vehicle's driving information cannot guarantee the accuracy of the final predicted driving information. Summary of the invention

[0003] The present disclosure provides a driving information prediction method, device, equipment and storage medium.

[0004] According to one aspect of the present disclosure, a driving information prediction method is provided, comprising:

[0005] Determine a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period;

[0006] Based on the first master-slave relationship, first predicted driving information is obtained, wherein the first predicted driving information includes: first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and second predicted motion parameters of the first obstacle at multiple predicted moments within the first time length;

[0007] Based on the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined, wherein the first optimal driving information includes the best predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the best predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0008] According to one aspect of the present disclosure, there is provided a driving information prediction device, comprising:

[0009] A master-slave relationship determination module, used to determine a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period;

[0010] A driving information prediction module, configured to obtain first predicted driving information based on the first master-slave relationship, wherein the first predicted driving information includes: first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and second predicted motion parameters of the first obstacle at multiple predicted moments within the first time length;

[0011] The optimal driving information determination module is used to determine the first optimal driving information based on the evaluation result corresponding to the first predicted driving information, wherein the first optimal driving information includes the optimal predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the optimal predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0012] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0013] at least one processor; and

[0014] a memory communicatively connected to the at least one processor; wherein,

[0015] 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 execute any method in the embodiments of the present disclosure.

[0016] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any method according to the embodiments of the present disclosure.

[0017] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which implements any method according to the embodiments of the present disclosure when executed by a processor.

[0018] According to another aspect of the present disclosure, an autonomous driving vehicle is provided, comprising the above-mentioned electronic device.

[0019] By adopting the above-mentioned implementation mode, the first master-slave relationship between the target vehicle and the obstacle can be determined in real time by combining the motion parameters of the target vehicle and the obstacle at the current moment and the path information of both parties in the first future period of time, and then the first predicted driving information can be obtained based on the first master-slave relationship between the two. Finally, the predicted best driving information is determined based on the evaluation result of the first predicted driving information. Since the master-slave relationship of each traffic participant is analyzed according to the current status and future path of each traffic participant, and the driving information is predicted and evaluated to obtain the best driving information, the accuracy of the driving information finally predicted can be improved.

[0020] 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

[0021] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0022] Figure 1 is a flow chart of a driving information prediction method according to an embodiment of the present disclosure;

[0023] Figure 2 This is a schematic diagram of a scenario for obtaining the right-of-way priority of a target vehicle and a first obstacle within a first time period according to an embodiment of the present disclosure;

[0024] Figure 3 is a schematic diagram of a scenario for determining first optimal driving information according to an embodiment of the present disclosure;

[0025] Figure 4 is a schematic diagram of a scenario for determining a process of a decision instruction according to an embodiment of the present disclosure;

[0026] Figure 5 is a flow chart of a driving information prediction method according to an embodiment of the present disclosure;

[0027] Figure 6 is a schematic diagram of a traffic scene according to an embodiment of the present disclosure;

[0028] Figure 7 is a schematic block diagram of a driving information prediction device according to an embodiment of the present disclosure;

[0029] Figure 8 is a schematic block diagram of a driving information prediction device according to another embodiment of the present disclosure;

[0030] Fig. 9 is a schematic block diagram of a driving information prediction device according to another embodiment of the present disclosure;

[0031] Fig.10 is a block diagram of an electronic device for implementing an embodiment of the present disclosure. DETAILED DESCRIPTION

[0032] 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 of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0033] Figure 1 is a schematic flow chart of a driving information prediction method proposed in an embodiment of the present disclosure, comprising:

[0034] S110, determining a first master-slave relationship between the target vehicle and the first obstacle based on a motion parameter of the target vehicle at a current moment, path information of the target vehicle within a first time period, a motion parameter of a first obstacle at a current moment, and predicted path information of the first obstacle within the first time period;

[0035] S120, obtaining first predicted driving information based on the first master-slave relationship, wherein the first predicted driving information includes: first predicted motion parameters of the target vehicle at multiple predicted moments within a first time period, and second predicted motion parameters of the first obstacle at multiple predicted moments within the first time period;

[0036] S130, determining first optimal driving information based on the evaluation result corresponding to the first predicted driving information, wherein the first optimal driving information includes the best predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the best predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0037] The driving information prediction method of the disclosed embodiment can be executed by an electronic device. The electronic device can be an on-board terminal set in the target vehicle, or other terminal device with computing capabilities and capable of communicating with the target vehicle. It should be understood that the above is only an exemplary description of the electronic device, and the actual processing may not be limited to the devices mentioned in the above examples. As long as the electronic device can execute the driving information prediction method provided in this embodiment, it is within the protection scope of this embodiment.

[0038] The first obstacle may refer to any one of all obstacles within the specified range of the target vehicle. The specified range of the target vehicle may be a range centered on the target vehicle and having a specified length as a radius, and the specified length may be configured according to actual conditions, such as 200 meters, 500 meters, or longer or shorter, which is not limited here. The first obstacle may belong to any type of traffic participant, such as a type of the first obstacle may be any one of a vehicle, a pedestrian, an animal, and the like.

[0039] By adopting the above scheme, the first master-slave relationship between the target vehicle and the obstacle can be determined in real time by combining the motion parameters of the target vehicle and the obstacle at the current moment and the path information of both parties in the first future period of time, and then the first predicted driving information can be obtained based on the first master-slave relationship between the two. Finally, the predicted best driving information is determined based on the evaluation result of the first predicted driving information. Since the master-slave relationship of each traffic participant is analyzed according to the current status and future path of each traffic participant, the driving information is predicted and evaluated to obtain the best driving information, the accuracy of the driving information finally predicted can be improved.

[0040] In some possible implementations, the determining a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time length, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time length includes: determining the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time length, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time length; determining the first master-slave relationship between the target vehicle and the first obstacle based on the right-of-way priority of the target vehicle and the first obstacle within the first time length.

[0041] The method of determining the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time length, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time length includes: obtaining the driving type of the target vehicle within the first time length based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time length; obtaining the driving type of the first obstacle within the first time length based on the motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle within the first time length; obtaining the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the driving type of the target vehicle within the first time length and the driving type of the first obstacle within the first time length.

[0042] The motion parameters include at least one of the following: position, speed, direction, and acceleration.

[0043] The driving type can also be alternatively referred to as a driving mode. For example, the driving type of the target vehicle within the first time period can be any one of a plurality of optional driving types, and the plurality of optional driving types include at least one of going straight, turning left, turning right, changing lanes, etc.

[0044] Based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle in the first time period, the process of obtaining the driving type of the target vehicle in the first time period can be: using the high-precision map information, the position of the target vehicle at the current moment, the speed of the target vehicle at the current moment, the direction of the target vehicle at the current moment, and the path information of the target vehicle in the first time period as the first input parameter; using the first calculation function to calculate the first input parameter to obtain the driving type of the target vehicle in the first time period. Among them, the first calculation function can be configured according to actual needs, and this embodiment does not limit it.

[0045] Among them, the motion parameters of the target vehicle at the current moment can be determined by combining high-precision data and positioning information, and the specific determination method is not limited in this embodiment.

[0046] The path information of the target vehicle within the first duration can be determined based on high-precision data and positioning information, and pre-planning information provided by the pre-planning system. The path information of the target vehicle within the first duration can also be referred to as the path of the target vehicle within the first duration, or the pre-planned path of the target vehicle, etc., and this embodiment does not limit its determination method. The path information of the target vehicle within the first duration refers to the path that the target vehicle is about to pass within the first duration. The starting time of the first duration can be the current time, and the duration of the first duration can be configured according to actual conditions, such as the duration can be 10 seconds, or 1 minute, or longer or shorter, which is not limited here.

[0047] For example, the first input parameter is calculated using the first calculation function to obtain the driving type of the target vehicle within the first time period, which can be expressed as: Among them, D adc is the driving type of the target vehicle in the first duration; is the first calculation function; M is the high-precision map information; P adc is the position of the target vehicle at the current moment; V adc is the speed of the target vehicle at the current moment; H adc is the direction of the target vehicle at the current moment; T represents the first duration; S adc is the path information of the target vehicle within the first time period.

[0048] Based on the motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle in the first time period, the process of obtaining the driving type of the first obstacle in the first time period can be: using the high-precision map information, the position of the first obstacle at the current moment, the speed of the first obstacle at the current moment, the direction of the first obstacle at the current moment, the predicted path information of the first obstacle in the first time period, the driving type of the target vehicle in the first time period, and the traffic type of the first obstacle as the second input parameter; using the second calculation function to calculate the second input parameter to obtain the driving type of the first obstacle in the first time period. The second calculation function is the same as or different from the first calculation function, both within the protection scope of this embodiment, and the second calculation function can be configured according to actual needs, and this embodiment does not limit it.

[0049] The motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle within the first time period can be determined based on the perception and prediction information, high-precision data and positioning information, and the specific perception method is not limited in this embodiment. The predicted path information of the first obstacle within the first time period refers to the path that the first obstacle is predicted to pass within the first time period.

[0050] The traffic type of the first obstacle may be any one of a plurality of optional traffic types. The plurality of optional traffic types may be divided according to at least one of the driving direction, driving speed, etc. For example, the plurality of optional traffic types may be divided according to the driving direction to obtain at least one of the following: straight-moving vehicles, turning vehicles, U-turning vehicles, merging vehicles, and outgoing vehicles, etc.; for another example, the plurality of optional traffic types may be divided according to the speed to obtain at least one of the following: high-speed traffic, medium-speed traffic, low-speed traffic, etc.; for another example, the plurality of optional traffic types may be divided according to the driving direction and speed. It should be noted that the division method of the plurality of optional traffic types may include but is not limited to the above possibilities, and there may be other division methods to divide the plurality of optional traffic types, which are not limited or exhaustive in this embodiment.

[0051] The traffic flow type of the first obstacle may be any one of a plurality of optional traffic flow types determined based on a traffic flow division strategy. The traffic flow division strategy may be pre-configured, and this embodiment does not limit the traffic flow division strategy.

[0052] The second input parameter is calculated using the second calculation function, and the driving type of the first obstacle within the first time period can be expressed as:

[0053]

[0054] Among them, D obs is the driving type of the first obstacle in the first duration; is the second calculation function; P obs is the position of the first obstacle at the current moment; V obs is the speed of the first obstacle at the current moment; H obs is the direction of the first obstacle at the current moment; L obs represents the predicted path information of the first obstacle in the first time period; F obs The traffic type of the first obstacle.

[0055] The right-of-way priority of the target vehicle and the first obstacle within the first time period may include: the relative right-of-way priority of the target vehicle relative to the first obstacle within the first time period, and the relative right-of-way priority of the first obstacle relative to the target vehicle within the first time period. The relative right-of-way priority may be represented by a first value or a second value. For example, when the relative right-of-way priority of the target vehicle relative to the first obstacle within the first time period is the first value, and the relative right-of-way priority of the first obstacle relative to the target vehicle within the first time period is the second value, it indicates that the target vehicle has a high right-of-way priority relative to the first obstacle within the first time period, and the first obstacle has a low right-of-way priority relative to the target vehicle within the first time period. The first value and the second value are different, and the first value and the second value may be configured according to actual conditions. For example, the first value may be 1, the second value may be 0, or vice versa. All possible values ​​of the first value and the second value are not limited or enumerated here.

[0056] Obtaining the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the driving type of the target vehicle within the first time length and the driving type of the first obstacle within the first time length may include: obtaining the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the driving type of the target vehicle within the first time length, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, the driving type of the first obstacle within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0057] Here, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length can be determined based on the path information of the target vehicle within the first time length and the motion parameters of the target vehicle at the current moment. This embodiment does not limit the specific method of determining the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length. For example, it can be determined by the pre-decision module based on a pre-configured strategy, or it can be calculated by the pre-decision algorithm in the pre-decision module. This embodiment does not list them all.

[0058] The initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length can be determined based on the predicted path information of the first obstacle within the first time length and the motion parameters of the first obstacle at the current moment. This embodiment does not limit the specific method for determining the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0059] Based on the driving type of the target vehicle within the first time length, the initial predicted motion parameters of the target vehicle at multiple prediction moments within the first time length, the driving type of the first obstacle within the first time length, and the initial predicted motion parameters of the first obstacle at multiple prediction moments within the first time length, a specific processing method for the right of way priority of the target vehicle and the first obstacle within the first time length is obtained, which is not limited in this embodiment.

[0060] Combination Figure 2 The process of obtaining the right-of-way priority of the target vehicle and the first obstacle within the first time period is exemplified as follows:

[0061] First, the motion parameters of the target vehicle at the current moment are determined in combination with high-precision data and positioning information 201, and the path information of the target vehicle within the first time period is determined according to the pre-planning information provided by the pre-planning system 202. The motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time period can be used together as the target vehicle position and motion trend information 210; based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time period, the driving type of the target vehicle within the first time period is obtained.

[0062] Then, the position and motion trend information 211 of the first obstacle (i.e., the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle in the first time period in the future (for example, expressed as T, where T is a positive number)) are determined using high-precision data and positioning information 201 and perception and prediction information 203. Based on the motion parameters of the first obstacle at the current moment, the predicted path information of the first obstacle in the first time period, the traffic strategy 204 (i.e., the traffic type of the first obstacle), etc., the driving type of the first obstacle in the first time period is obtained.

[0063] The above processing not only relies on static high-precision data (such as high-precision map information), but also needs to obtain real-time perception, prediction data and output results of previous decisions, etc., to ensure accurate understanding of the surrounding environment, so as to accurately analyze the driving type of the target vehicle and the first obstacle in the first period of time. In this way, we can have a preliminary understanding of the intentions and potential risks of the target vehicle and the first obstacle. By integrating this information, the system can build a relatively comprehensive environmental model, laying a solid foundation for subsequent hyper-interactive simulation, collision risk and trajectory rationality assessment.

[0064] Finally, through an in-depth analysis of the driving types of the target vehicle and the first obstacle in the first time length, the right-of-way priority 205 of the target vehicle and the first obstacle in the first time length is determined. The determination of the right-of-way priority 205 of the target vehicle and the first obstacle in the first time length can be determined based on a pre-configured priority-related strategy or algorithm, which is not limited here. After determining the right-of-way priority 205 of the target vehicle and the first obstacle in the first time length, it can also be processed in an encoding manner to obtain the encoding content required for subsequent processing. For example, the encoding content may include at least one of the right-of-way priority 205 of the target vehicle and the first obstacle in the first time length, the initial predicted motion parameters of the target vehicle at multiple predicted moments in the first time length, the initial predicted motion parameters of the first obstacle at multiple predicted moments in the first time length, the traffic type of the first obstacle, etc.

[0065] In this way, the motion parameters of the target vehicle and the first obstacle at the current moment and the path information of the target vehicle and the first obstacle in the first time period in the future can be used to obtain the driving types of the target vehicle and the first obstacle respectively, and then based on the driving types of the target vehicle and the first obstacle, the relative right of way priority of both parties can be determined. In this way, by determining the driving type in combination with the motion parameters of the target vehicle and the first obstacle at the current moment, the real-time and accuracy of the driving types of both parties obtained by analysis can be guaranteed, thereby ensuring the real-time and accuracy of the relative right of way priority of both parties determined.

[0066] The first master-slave relationship between the target vehicle and the first obstacle may include an object in a dominant position and an object in a subordinate position between the target vehicle and the first obstacle within a first time period. The object in a dominant position may be an object that passes through the predicted intersection area first between the target vehicle and the first obstacle, and the object in a subordinate position may be an object that passes through the predicted intersection area later between the target vehicle and the first obstacle.

[0067] The first master-slave relationship between the target vehicle and the first obstacle can be determined based on a preset correlation strategy and the right-of-way priority of the target vehicle and the first obstacle within the first time period, or can be obtained by processing the right-of-way priority of the target vehicle and the first obstacle within the first time period using a preset master-slave relationship analysis model, which is not limited or exhaustive here.

[0068] Exemplarily, based on the right-of-way priorities of the target vehicle and the first obstacle within a first time length, a first master-slave relationship between the target vehicle and the first obstacle is determined, including: determining a predicted intersection area based on initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length; determining the first master-slave relationship between the target vehicle and the first obstacle based on the right-of-way priorities of the target vehicle and the first obstacle within the first time length, the predicted intersection area and reference information, the reference information including at least one of the following: the driving type of the target vehicle, the driving type of the first obstacle, the motion parameters of the target vehicle at the current moment, the motion parameters of the first obstacle at the current moment, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0069] For example, based on the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length, the predicted intersection area is determined; based on the right-of-way priority of the target vehicle and the first obstacle within the first time length, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length, the object that first passes through the predicted intersection area between the target vehicle and the first obstacle is determined, and the object that first passes through the predicted intersection area is used as the object in the dominant position in the first master-slave relationship, and the other object is used as the object in the subordinate position in the first master-slave relationship. The specific method for determining the object that first passes through the predicted intersection area between the target vehicle and the first obstacle is not limited here.

[0070] For example, based on the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length, the predicted intersection area is determined; based on the motion parameters of the target vehicle at the current moment and the motion parameters of the first obstacle at the current moment, the first distance between the target vehicle and the predicted intersection area and the second distance between the first obstacle and the predicted intersection area are determined; based on the first distance, the second distance, and the road right priorities of the target vehicle and the first obstacle within the first time length, the first master-slave relationship between the target vehicle and the first obstacle is determined. For example, assuming that the first distance is close to the second distance, that is, the difference between the first distance and the second distance is not greater than the specified distance threshold, the specified distance threshold can be configured according to actual conditions, and can be 10 meters, or 20 meters, or longer or shorter; in this case, since the difference between the first distance and the second distance is not large, it is possible to further predict the time it takes for the target vehicle and the first obstacle to reach the predicted intersection area based on the motion parameters of the target vehicle and the first obstacle at the current moment, and make the time it takes for the two to reach the predicted intersection area basically the same. It can be determined that the object with a higher relative road right priority is the object that passes through the predicted intersection area first, that is, the object in a dominant position in the first master-slave relationship, and the other party is the object in a subordinate position.

[0071] The above description is merely an exemplary description, and in actual processing, other methods or strategies may be used to determine the first master-slave relationship between the target vehicle and the first obstacle.

[0072] As an alternative example, the method for determining the first master-slave relationship may be: determining the predicted intersection area based on the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length; determining the first distance between the target vehicle and the predicted intersection area and the second distance between the first obstacle and the predicted intersection area based on the motion parameters of the target vehicle at the current moment and the motion parameters of the first obstacle at the current moment; determining the first master-slave relationship between the target vehicle and the first obstacle based on the first distance, the second distance and at least one of the following: the driving type of the target vehicle, the driving type of the first obstacle, the motion parameters of the target vehicle at the current moment, the motion parameters of the first obstacle at the current moment, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0073] For example, assuming that the first distance is much greater than the second distance, for example, the difference between the first distance and the second distance is greater than a specified distance threshold or the first distance is greater than the second distance multiplied by a specified multiple, the specified multiple can be configured according to actual conditions, for example, it can be 2 times, or 3 times, or greater, which is not limited or exhaustive here. In this case, since the first distance is much greater than the second distance, it can be determined that the target vehicle is much farther from the intersection position than the first obstacle, and the first master-slave relationship can be determined as the first obstacle being dominant and the target vehicle being subordinate.

[0074] Assume that the first distance is close to the second distance, for example, the difference between the first distance and the second distance is not greater than the specified distance threshold. In this case, since the difference between the first distance and the second distance is not large, the first master-slave relationship can be further determined based on the right-of-way priority of the target vehicle and the first obstacle within the first time period and at least one of the following: the driving type of the target vehicle, the driving type of the first obstacle, the motion parameters of the target vehicle at the current moment, the motion parameters of the first obstacle at the current moment, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time period, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time period. For example, based on the motion parameters of the target vehicle and the first obstacle at the current moment, the time it takes for the target vehicle and the first obstacle to reach the predicted intersection area is predicted, and the object with the shorter time to reach the predicted intersection area is taken as the dominant object in the first master-slave relationship, and the object with the longer time to reach the predicted intersection area is taken as the subordinate object in the first master-slave relationship.

[0075] It should be pointed out that the above are all exemplary descriptions, and other methods or strategies may be used to determine the first master-slave relationship between the target vehicle and the first obstacle in actual processing, and this embodiment does not limit or exhaustively list them.

[0076] In this way, the relative right-of-way priorities of the target vehicle and the first obstacle can be determined first, and then the first master-slave relationship between the two parties can be determined based on the right-of-way priorities of both parties. In this way, the determination of the first master-slave relationship between the target vehicle and the first obstacle can be more accurate, thereby ensuring the accuracy of the subsequent predicted driving information.

[0077] In some possible implementations, obtaining the first predicted driving information based on the first master-slave relationship includes: performing forward simulation based on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; using the first predicted motion parameters of the target vehicle at the multiple predicted moments within the first time length, and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length as the first predicted driving information.

[0078] Among them, forward simulation is performed based on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length, which may include: performing multiple iterative processing on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment based on the forward simulation method to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length.

[0079] Exemplarily, performing multiple iterative processing on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment based on a forward simulation method to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length may include: in the i-th iterative processing, taking the first master-slave relationship, the motion parameters of the target vehicle at the i-th moment, and the motion parameters of the first obstacle at the i-th moment as the i-th input parameters, processing the i-th input parameters based on a forward simulation method, and predicting the first predicted motion parameters of the target vehicle at the i+1-th predicted moment and the second predicted motion parameters of the first obstacle at the i+1-th predicted moment; when the i+1-th predicted moment is the last predicted moment within the first time length, determining that multiple iterative processing is completed to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length. Among them, when i is equal to 1, the i-th moment is the current moment; when i is an integer greater than 1, the i-th moment is the i-th predicted moment.

[0080] In addition, it may also include: when the i+1th prediction moment is not the last prediction moment in the first duration, returning to perform the i+1th iteration process, the i+1th iteration process is similar to the i-th iteration process and will not be described in detail.

[0081] The forward simulation method may include: Stackelberg game or Monte Carlo tree search method, etc., and all possible algorithms or methods that may be used in the forward simulation method are not limited or enumerated here.

[0082] The first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length may include: the first predicted motion parameters of the target vehicle at each predicted moment in the multiple predicted moments within the first time length, wherein the first predicted motion parameters at any predicted moment include at least one of the following: the first predicted speed at any predicted moment, the first predicted position at any predicted moment, and the first predicted direction at any predicted moment.

[0083] The second predicted motion parameters of the first obstacle at multiple predicted moments within the first time length may include: the second predicted motion parameters of the first obstacle at each predicted moment among the multiple predicted moments within the first time length, wherein the second predicted motion parameters at any predicted moment include at least one of the following: the second predicted speed at any predicted moment, the second predicted position at any predicted moment, and the second predicted direction at any predicted moment.

[0084] In this way, forward simulation is performed through the current motion parameters of the target vehicle and the current motion parameters of the first obstacle perceived in real time, combined with the first master-slave relationship between the target vehicle and the first obstacle determined in real time, to predict the motion parameters of the target vehicle and the first obstacle at multiple predicted moments in the future, thereby ensuring the real-time and accuracy of the prediction results.

[0085] In some possible embodiments, determining the first optimal driving information based on the evaluation result corresponding to the first predicted driving information includes one of the following: when the evaluation result corresponding to the first predicted driving information satisfies a first condition, using the first predicted driving information as the first optimal driving information; when the evaluation result corresponding to the first predicted driving information does not satisfy the first condition, alternating the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle, obtaining second predicted driving information based on the second master-slave relationship, and determining the first optimal driving information from the first predicted driving information and the second predicted driving information based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information.

[0086] The first condition includes at least one of the following: a first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than a safety threshold, and a first somatosensory assessment value in the assessment result corresponding to the first predicted driving information is higher than a somatosensory assessment threshold. The first condition can be used to measure whether the first predicted driving information meets safety requirements and / or somatosensory requirements. Among them, the safety threshold can also be referred to as a safety requirement threshold, and the somatosensory assessment threshold can also be referred to as a somatosensory requirement threshold. Both the safety threshold and the somatosensory assessment threshold can be configured according to actual conditions, and this embodiment does not limit them.

[0087] The method also includes at least one of the following: based on a safety assessment formula, the first master-slave relationship and the first predicted driving information, calculating a first safety assessment value in an assessment result corresponding to the first predicted driving information; based on a somatosensory assessment formula, the first master-slave relationship and the first predicted driving information, calculating a first somatosensory assessment value in an assessment result corresponding to the first predicted driving information; based on the safety assessment formula, the second master-slave relationship and the second predicted driving information, calculating a second safety assessment value in an assessment result corresponding to the second predicted driving information; based on the somatosensory assessment formula, the second master-slave relationship and the second predicted driving information, calculating a second somatosensory assessment value in an assessment result corresponding to the second predicted driving information.

[0088] Preferably, the evaluation result corresponding to the first predicted driving information may include a first safety evaluation value and a first somatosensory evaluation value. In this case, the first condition includes that the first safety evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the safety threshold, and the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the somatosensory evaluation threshold.

[0089] The first safety assessment value may be used to indicate the level of collision risk between the target vehicle corresponding to the first predicted driving information and the first obstacle. A higher first safety assessment value indicates a lower collision risk; a lower first safety assessment value indicates a higher collision risk.

[0090] The method of obtaining the first safety assessment value in the assessment result corresponding to the first predicted driving information may include: calculating the first safety assessment value in the assessment result corresponding to the first predicted driving information based on a safety assessment formula, the first master-slave relationship, and the first predicted driving information. The safety assessment formula may be configured according to actual conditions.

[0091] For example, based on the safety assessment formula, the first master-slave relationship and the first predicted driving information, a first safety assessment value in the assessment result corresponding to the first predicted driving information is calculated and can be expressed as:

[0092]

[0093] in, represents the first safety assessment value in the assessment result corresponding to the first predicted driving information, s R () represents the safety assessment formula, P adc-P1 It can represent the first predicted position of the target vehicle at multiple predicted moments within the first time period in the first predicted driving information, P obs-P1 represents the second predicted position of the first obstacle in the first predicted driving information at multiple predicted moments within the first time period, V adc-P1 It can represent the first predicted speed of the target vehicle at multiple predicted moments within the first time period in the first predicted driving information, V obs-P1 represents the second predicted speed of the first obstacle at multiple predicted moments within the first time period in the first predicted driving information, H adc-P1 It can represent the first predicted direction of the target vehicle at multiple predicted moments within the first time period in the first predicted driving information, H obs-P1 represents the second predicted direction of the first obstacle in the first predicted driving information at multiple predicted moments within the first time period, + Indicates the primary master-slave relationship.

[0094] The first somatosensory evaluation value can be used to represent the somatosensory feeling of the target vehicle and the first obstacle corresponding to the first predicted driving information during the interaction process. Here, the somatosensory feeling may refer to the feeling of being pushed back by the person caused by acceleration during the interaction process between the target vehicle and the first obstacle, and / or the forward leaning state of the person caused by deceleration during the interaction process between the target vehicle and the first obstacle, wherein the person may include at least one of the driver in the target vehicle, the passenger in the target vehicle, the driver in the first obstacle, and the passenger in the first obstacle. The higher the first somatosensory evaluation value, the higher the somatosensory feeling of the target vehicle and the first obstacle during the interaction process; if the first somatosensory evaluation value is lower, the worse the somatosensory feeling of the target vehicle and the first obstacle during the interaction process.

[0095] The method of obtaining the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information may include: calculating the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information based on the somatosensory evaluation formula, the first master-slave relationship and the first predicted driving information.

[0096] For example, based on the somatosensory evaluation formula, the first master-slave relationship and the first predicted driving information, the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is calculated and can be expressed as:

[0097]

[0098] in, represents the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information, S L () represents the somatosensory evaluation formula. The meanings of the remaining items are the same as those in the previous example and are not elaborated on herein.

[0099] Determining whether the evaluation result corresponding to the first predicted driving information satisfies the first condition may be: determining whether the first safety evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the safety threshold, and determining whether the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the somatosensory evaluation threshold. Further, when the first safety evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the safety threshold, and the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the somatosensory evaluation threshold, it is determined that the first condition is satisfied; or, when the first safety evaluation value in the evaluation result corresponding to the first predicted driving information is not higher than the safety threshold, and / or the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is not higher than the somatosensory evaluation threshold, it is determined that the first condition is not satisfied.

[0100] Optionally, the evaluation result corresponding to the first predicted driving information may include a first safety evaluation value. The processing method for obtaining the first safety evaluation value in the evaluation result corresponding to the first predicted driving information is the same as that in the above embodiment and will not be described in detail.

[0101] In this case, the first condition includes that the first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than the safety threshold. Specifically, judging whether the assessment result corresponding to the first predicted driving information satisfies the first condition may be: judging whether the first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than the safety threshold. Further, when the first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than the safety threshold, it is determined that the first condition is satisfied; or, when the first safety assessment value in the assessment result corresponding to the first predicted driving information is not higher than the safety threshold, it is determined that the first condition is not satisfied.

[0102] Optionally, the evaluation result corresponding to the first predicted driving information may include a first somatosensory evaluation value. The processing method for obtaining the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is the same as the above example and will not be described in detail.

[0103] In this case, the first condition includes: the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the somatosensory evaluation threshold. Specifically, judging whether the evaluation result corresponding to the first predicted driving information satisfies the first condition may be: judging whether the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the somatosensory evaluation threshold. Further, in the case where the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is higher than the somatosensory evaluation threshold, it is determined that the first condition is satisfied; or, in the case where the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is not higher than the somatosensory evaluation threshold, it is determined that the first condition is not satisfied.

[0104] In one example, the evaluation result corresponding to the first predicted driving information satisfies the first condition, which may indicate that the first predicted driving information may meet safety requirements and / or physical requirements. Therefore, in this example, the first predicted driving information may be directly used as the first optimal driving information, that is, the best predicted motion parameter of the target vehicle at each predicted moment in the first optimal driving information among multiple predicted moments in the first time length is the first predicted motion parameter of the target vehicle at each predicted moment in the first time length, and the best predicted motion parameter of the first obstacle at multiple predicted moments in the first optimal driving information within the first time length is the second predicted motion parameter of the first obstacle at each predicted moment in the first time length.

[0105] In one example, if the evaluation result corresponding to the first predicted driving information does not meet the first condition, it can mean that the first predicted driving information does not meet the safety requirements and / or physical requirements. Therefore, in this example, the following processing needs to be further performed: alternating the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle, based on the second master-slave relationship, obtaining the second predicted driving information, based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, determining the first optimal driving information from the first predicted driving information and the second predicted driving information.

[0106] In this way, by using the first condition that can be used to measure whether the predicted driving information meets the safety requirements and / or physical requirements, the first predicted driving information corresponding to the first master-slave relationship can be evaluated, thereby ensuring the safety and / or physical sensation of the first optimal driving information finally obtained.

[0107] Alternating the first master-slave relationship to obtain the second master-slave relationship between the target vehicle and the first obstacle may refer to swapping the object in a dominant position and the object in a subordinate position between the target vehicle and the first obstacle in the first time period in the first master-slave relationship to obtain the second master-slave relationship between the target vehicle and the first obstacle. For example, in the first master-slave relationship, the target vehicle is the object in a dominant position and the first obstacle is the object in a subordinate position; after alternating the first master-slave relationship, the first obstacle is the object in a dominant position and the target vehicle is the object in a subordinate position in the second master-slave relationship.

[0108] Obtaining second predicted driving information based on the second master-slave relationship may include: performing forward simulation based on the second master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; using the third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length as the second predicted driving information.

[0109] Here, a forward simulation method is used to obtain the third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length. The processing method is similar to the method of obtaining the first predicted motion parameters of the target vehicle at multiple predicted moments within the first time length and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length in the aforementioned embodiment, so they are not repeated.

[0110] The third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length may include: the third predicted motion parameters of the target vehicle at each predicted moment in the multiple predicted moments within the first time length, wherein the third predicted operating parameters at any predicted moment include at least one of the following: the third predicted speed at any predicted moment, the third predicted position at any predicted moment, and the third predicted direction at any predicted moment.

[0111] The fourth predicted motion parameters of the first obstacle at multiple predicted moments within the first time length may include: the fourth predicted motion parameters of the first obstacle at each predicted moment among the multiple predicted moments within the first time length, wherein the fourth predicted motion parameters at any predicted moment include at least one of the following: the fourth predicted speed at any predicted moment, the fourth predicted position at any predicted moment, and the fourth predicted direction at any predicted moment.

[0112] It should be pointed out that in the process of obtaining the second predicted driving information and the first predicted driving information, although the motion parameters of the target vehicle at the current moment and the motion parameters of the first obstacle at the current moment are used as input, the two use different master-slave relationships respectively. For example, in the first master-slave relationship, the target vehicle is in a dominant position and the first obstacle is in a subordinate position, and in the second master-slave relationship, the target vehicle is in a subordinate position and the first obstacle is in a dominant position. In order to meet the needs of different objects in the first master-slave relationship and the second master-slave relationship that are in a dominant position first pass through the predicted intersection area, the predicted motion parameters of the target vehicle and the first obstacle contained in the second predicted driving information and the first predicted driving information at the same predicted moment may be different. Specifically, the third predicted motion parameter of the target vehicle at the first predicted moment within the first time length and the first predicted motion parameter of the target vehicle at the first predicted moment within the first time length may be the same or different. For example, at least one of the third predicted speed, the third predicted position, and the third predicted direction of the target vehicle at the first predicted moment within the first time length may be different from at least one of the first predicted speed, the first predicted position, and the first predicted direction of the target vehicle at the first predicted moment within the first time length. The fourth predicted motion parameter of the first obstacle at the first predicted moment within the first time length and the second predicted motion parameter of the first obstacle at the first predicted moment within the first time length may be the same or different. For example, at least one of the fourth predicted speed, fourth predicted position, and fourth predicted direction of the first obstacle at the first predicted moment within the first time length may be different from at least one of the second predicted speed, second predicted position, and second predicted direction of the first obstacle at the first predicted moment within the first time length.

[0113] Preferably, the evaluation result corresponding to the first predicted driving information includes a first safety evaluation value and a first somatosensory evaluation value, and the evaluation result corresponding to the second predicted driving information may include a second safety evaluation value and a second somatosensory evaluation value.

[0114] The relevant descriptions of the first safety assessment value and the first somatosensory assessment value are the same as those in the above-mentioned embodiment and are not elaborated herein.

[0115] The second safety assessment value may be used to indicate the level of collision risk between the target vehicle corresponding to the second predicted driving information and the first obstacle.

[0116] The method of obtaining the second safety assessment value may include: calculating the second safety assessment value in the assessment result corresponding to the second predicted driving information based on the safety assessment formula, the second master-slave relationship and the second predicted driving information.

[0117] For example, the second master-slave relationship and the second predicted driving information are calculated based on the safety assessment formula to obtain a safety assessment value between the target vehicle and the first obstacle corresponding to the second predicted driving information, which can be expressed as:

[0118]

[0119] in, represents the second safety assessment value, S R () represents the safety assessment formula, P adc-P2 It can represent the third predicted position of the target vehicle at multiple predicted moments within the first time period in the second predicted driving information, P obs-P2 represents the fourth predicted position of the first obstacle in the second predicted driving information at multiple predicted moments within the first time period, V adc-P2 V may represent the third predicted speed of the target vehicle at multiple predicted moments within the first time period in the second predicted driving information, obs-P2 represents the fourth predicted speed of the first obstacle at multiple predicted moments within the first time period in the second predicted driving information, H adc-P2 It can represent the third predicted direction of the target vehicle at multiple predicted moments within the first time period in the second predicted driving information, H obs-P2 represents the fourth predicted direction of the first obstacle at multiple predicted time points within the first time period in the second predicted driving information, - Indicates the second master-slave relationship.

[0120] The second somatosensory evaluation value may be used to represent the somatosensory feeling during the interaction between the target vehicle and the first obstacle corresponding to the second predicted driving information.

[0121] The method of obtaining the second somatosensory evaluation value may include: calculating the second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information based on the somatosensory evaluation formula, the second master-slave relationship and the second predicted driving information.

[0122] For example, the second master-slave relationship and the second predicted driving information are calculated based on the somatosensory evaluation formula to obtain a somatosensory evaluation value corresponding to the second predicted driving information during the interaction between the target vehicle and the first obstacle, which can be expressed as:

[0123] in, It represents the second somatosensory evaluation value, and the meanings of the other contents are the same as those in the above example and will not be elaborated on.

[0124] In this case, the first best driving information is determined from the first predicted driving information and the second predicted driving information based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, including: determining a first reference value corresponding to the first predicted driving information based on the evaluation result corresponding to the first predicted driving information; determining a second reference value corresponding to the second predicted driving information based on the evaluation result corresponding to the second predicted driving information; and selecting the first best driving information from the first predicted driving information and the second predicted driving information based on the maximum value of the first reference value and the second reference value.

[0125] Among them, based on the evaluation result corresponding to the first predicted driving information, determining the first reference value corresponding to the first predicted driving information may include: multiplying a first factor by a first safety evaluation value in the evaluation result corresponding to the first predicted driving information to obtain a first value, multiplying a second factor by a first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information to obtain a second value, and adding the first value and the second value to obtain the first reference value corresponding to the first predicted driving information.

[0126] Based on the evaluation result corresponding to the second predicted driving information, determining the second reference value corresponding to the second predicted driving information may include: multiplying the first factor by the second safety evaluation value in the evaluation result corresponding to the second predicted driving information to obtain a third value, multiplying the second factor by the second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information to obtain a fourth value, and adding the third value and the fourth value to obtain the second reference value corresponding to the second predicted driving information.

[0127] The first factor and the second factor may be determined based on the balance factor. The balance factor may be configured according to actual conditions, and the value of the balance factor may be greater than 0 and less than 1. For example, the balance factor is represented by γ, and the value range of the balance factor is represented by γ∈(0,1). The first factor may be equal to the balance factor (for example, the first factor is represented by γ), and the second factor may be equal to 1 minus the balance factor, for example, the second factor is represented by (1-γ).

[0128] Selecting the first optimal driving information from the first predicted driving information and the second predicted driving information based on the maximum value between the first reference value and the second reference value may include: selecting the maximum reference value from the first reference value and the second reference value; selecting the predicted driving information corresponding to the maximum reference value from the first predicted driving information and the second predicted driving information as the first optimal driving information.

[0129] In this example, the first reference value is calculated by combining the first safety assessment value and the first somatosensory assessment value. Therefore, the first reference value can be a numerical value used to represent the comprehensive evaluation situation of the first predicted driving information, and the size of the first reference value is positively correlated with the quality of the comprehensive evaluation situation of the first predicted driving information.

[0130] Since the second reference value is calculated by combining the second safety assessment value and the second somatosensory assessment value, the second reference value can be a numerical value used to represent the comprehensive assessment situation of the second predicted driving information, and the size of the second reference value is positively correlated with the quality of the comprehensive assessment situation of the second predicted driving information.

[0131] Furthermore, the sizes of the first reference value and the second reference value are positively correlated with the comprehensive evaluation conditions of the corresponding predicted driving information respectively. Therefore, the predicted driving information corresponding to the maximum value between the first reference value and the second reference value has a relatively better comprehensive evaluation condition among the two. Based on this, this example can select the predicted driving information with the best comprehensive evaluation condition as the first optimal driving information through the maximum value between the first reference value and the second reference value.

[0132] For example, based on the maximum value of the first reference value and the second reference value, the first optimal driving information is selected from the first predicted driving information and the second predicted driving information, which can be expressed by the following formula:

[0133]

[0134] Among them, τ op( Indicates the predicted driving information corresponding to the maximum value, i.e., the first optimal driving information; argmax() indicates taking the maximum value; τ + represents the first predicted driving information obtained based on the first master-slave relationship; τ - represents the second predicted driving information obtained based on the second master-slave relationship; It can represent the calculation of the first reference value, and the meaning of each parameter in the formula is the same as that in the above embodiment; It can represent the calculation of the second reference value. The meaning of each parameter in the formula is the same as that in the aforementioned embodiment and will not be described in detail.

[0135] For example, if the first predicted driving information is determined to be the first optimal driving information from the first predicted driving information and the second predicted driving information, the best predicted motion parameter of the target vehicle at each predicted moment in the first predicted moments in the first optimal driving information is the first predicted motion parameter of the target vehicle at each predicted moment in the first time length, and the best predicted motion parameter of the first obstacle at multiple predicted moments in the first optimal driving information is the second predicted motion parameter of the first obstacle at each predicted moment in the first time length. For another example, if the second predicted driving information is determined to be the first optimal driving information from the first predicted driving information and the second predicted driving information, the best predicted motion parameter of the target vehicle at each predicted moment in the first predicted moments in the first predicted moments in the first time length in the first optimal driving information is the third predicted motion parameter of the target vehicle at each predicted moment in the first time length, and the best predicted motion parameter of the first obstacle at multiple predicted moments in the first optimal driving information is the fourth predicted motion parameter of the first obstacle at each predicted moment in the first time length.

[0136] Optionally, the evaluation result corresponding to the first predicted driving information includes a first safety evaluation value; and the evaluation result corresponding to the second predicted driving information may include a second safety evaluation value. The description of the first safety evaluation value and the second safety evaluation value is the same as that of the above embodiment and will not be repeated.

[0137] In this case, the first best driving information is determined from the first predicted driving information and the second predicted driving information based on the evaluation results corresponding to the second predicted driving information and the evaluation results corresponding to the first predicted driving information, including: determining the first reference value corresponding to the first predicted driving information based on the first safety evaluation value in the evaluation results corresponding to the first predicted driving information; determining the second reference value corresponding to the second predicted driving information based on the second safety evaluation value in the evaluation results corresponding to the second predicted driving information; and selecting the first best driving information from the first predicted driving information and the second predicted driving information based on the maximum value of the first reference value and the second reference value.

[0138] Determining the first reference value corresponding to the first predicted driving information based on the first safety assessment value in the assessment result corresponding to the first predicted driving information may mean using the first safety assessment value in the assessment result corresponding to the first predicted driving information as the first reference value corresponding to the first predicted driving information.

[0139] Determining the second reference value corresponding to the second predicted driving information based on the second safety assessment value in the assessment result corresponding to the second predicted driving information may mean using the second safety assessment value in the assessment result corresponding to the second predicted driving information as the second reference value corresponding to the second predicted driving information.

[0140] Selecting the first optimal driving information from the first predicted driving information and the second predicted driving information based on the maximum value between the first reference value and the second reference value may include: selecting the maximum reference value from the first reference value and the second reference value; selecting the predicted driving information corresponding to the maximum reference value from the first predicted driving information and the second predicted driving information as the first optimal driving information.

[0141] In this example, the first reference value is equal to the first safety assessment value, so the first reference value and the first safety assessment value have the same meaning, both are numerical values ​​used to represent the safety assessment of the first predicted driving information, and the size of the first reference value is positively correlated with the quality of the safety assessment of the first predicted driving information. The second reference value is equal to the second safety assessment value, so the second reference value and the second safety assessment value have the same meaning, both are numerical values ​​used to represent the safety assessment of the second predicted driving information, and the size of the second reference value is positively correlated with the quality of the safety assessment of the second predicted driving information.

[0142] Furthermore, the sizes of the first reference value and the second reference value are positively correlated with the safety assessment conditions of the corresponding predicted driving information respectively. Therefore, the predicted driving information corresponding to the maximum value between the first reference value and the second reference value has a relatively better safety assessment condition between the two. Based on this, this example can select the predicted driving information with the best safety assessment condition as the first optimal driving information through the maximum value between the first reference value and the second reference value.

[0143] For example, based on the maximum value of the first reference value and the second reference value, the first optimal driving information is selected from the first predicted driving information and the second predicted driving information, which can be expressed by the following formula: in, It can be indicated that the first safety assessment value is used as the first reference value. It can be indicated that the second safety assessment value is used as the second reference value. The meanings of the remaining items in the formula are the same as those in the aforementioned embodiment and are not described in detail.

[0144] Optionally, the evaluation result corresponding to the first predicted driving information includes a first somatosensory evaluation value; and the evaluation result corresponding to the second predicted driving information may include a second somatosensory evaluation value. The relevant descriptions of the first somatosensory evaluation value and the second somatosensory evaluation value are the same as those in the above embodiment and are not repeated here.

[0145] In this case, the first best driving information is determined from the first predicted driving information and the second predicted driving information based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, including: determining the first reference value corresponding to the first predicted driving information based on the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information; determining the second reference value corresponding to the second predicted driving information based on the second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information; and selecting the first best driving information from the first predicted driving information and the second predicted driving information based on the maximum value of the first reference value and the second reference value.

[0146] Determining the first reference value corresponding to the first predicted driving information based on the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information may mean using the first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information as the first reference value corresponding to the first predicted driving information.

[0147] Determining the second reference value corresponding to the second predicted driving information based on the second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information may mean using the second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information as the second reference value corresponding to the second predicted driving information.

[0148] Selecting the first optimal driving information from the first predicted driving information and the second predicted driving information based on the maximum value between the first reference value and the second reference value may include: selecting the maximum reference value from the first reference value and the second reference value; selecting the predicted driving information corresponding to the maximum reference value from the first predicted driving information and the second predicted driving information as the first optimal driving information.

[0149] In this example, the first reference value is equal to the first somatosensory evaluation value, so the first reference value and the first somatosensory evaluation value have the same meaning, both are numerical values ​​used to represent the somatosensory evaluation of the first predicted driving information, and the size of the first reference value is positively correlated with the quality of the somatosensory evaluation of the first predicted driving information. The second reference value is equal to the second somatosensory evaluation value, so the second reference value and the second somatosensory evaluation value have the same meaning, both are numerical values ​​used to represent the somatosensory evaluation of the second predicted driving information, and the size of the second reference value is positively correlated with the quality of the somatosensory evaluation of the second predicted driving information.

[0150] Furthermore, the sizes of the first reference value and the second reference value are respectively positively correlated with the quality of the physical perception evaluation of the corresponding predicted driving information. Therefore, the predicted driving information corresponding to the maximum value between the first reference value and the second reference value is the one with relatively better physical perception evaluation between the two. Based on this, this example can select the predicted driving information with the best physical perception evaluation as the first optimal driving information through the maximum value between the first reference value and the second reference value.

[0151] Combination Figure 3 The above process of determining the first best driving information is exemplarily described as follows:

[0152] First, based on the right-of-way priority 301 of the target vehicle and the first obstacle in the first time length, the initial predicted motion parameters of the target vehicle at multiple predicted moments in the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments in the first time length, a master-slave relationship 302 is determined, specifically, the first master-slave relationship 302 between the target vehicle and the first obstacle is determined. For example, according to the relative position, speed, acceleration of the target vehicle and the first obstacle at multiple predicted moments in the first time length, the traffic type and the right-of-way priority of the first obstacle, the master-slave relationship between the target vehicle and the first obstacle in the first time length is established. + .

[0153] Then, based on the existing kinematic model, combined with forward mode methods such as Stackelberg game or Monte Carlo tree search method, based on the first master-slave relationship between the target vehicle and the first obstacle, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, a hyper-interactive forward simulation (forward simulation) 303 is performed to obtain the first predicted driving information.

[0154] Next, the collision risk and trajectory rationality evaluation 304 is performed based on the first predicted driving information, that is, the safety evaluation value corresponding to the first predicted driving information is calculated (used to evaluate the collision risk of both parties). ), somatosensory evaluation value (used to evaluate the somatosensory loss of overtaking or giving way ), the specific calculation method is the same as that in the previous embodiment and will not be described in detail.

[0155] If the evaluation result determined based on the first predicted driving information indicates that the collision risk and physical loss of both parties are high, and the risk borne by the decision of the self-vehicle and the other vehicle to give way or overtake is high, it proves that the rationality of the current first predicted driving information is not good, and it is necessary to re-examine the first master-slave relationship and make appropriate adjustments. Therefore, further executing the secondary game 305, the secondary game may include: alternating the first master-slave relationship to obtain the second master-slave relationship, so that both parties can obtain more favorable decision-making conditions. This adjustment is not only based on the response of risk assessment, but also to achieve more efficient interaction and risk control in an uncertain traffic environment. Based on the second master-slave relationship after the exchange - Perform forward simulation again to obtain the second predicted driving information, and then perform collision risk and trajectory rationality assessment based on the second predicted driving information, that is, calculate the safety assessment value corresponding to the second predicted driving information (used to assess the collision risk of both parties). ), somatosensory evaluation value (used to evaluate the somatosensory loss of overtaking or giving way ), the specific calculation method is the same as that in the previous embodiment and will not be described in detail.

[0156] Finally, based on the evaluation results of the two predicted driving information before and after the master-slave relationship is exchanged, the first optimal driving information is selected to ensure that both parties can complete the super interaction in a safe and comfortable manner.

[0157] In this way, the first master-slave relationship and the first predicted driving information contain the predicted motion parameters of the target vehicle and the first obstacle at each prediction moment, respectively, for physical perception and / or safety assessment, thereby ensuring the comprehensiveness and timeliness of the assessment of the first predicted driving information; similarly, the second master-slave relationship and the second predicted driving information contain the predicted motion parameters of the target vehicle and the first obstacle at each prediction moment, respectively, for physical perception and / or safety assessment, thereby ensuring the comprehensiveness and timeliness of the assessment of the second predicted driving information.

[0158] Furthermore, the above processing can combine the evaluation results of the first predicted driving information and the evaluation results of the second predicted driving information to determine the first best driving information among the two predicted driving information. Since the two predicted driving information are determined based on two master-slave relationships respectively, the most comprehensive evaluation results of various possible predicted driving information of each traffic participant can be given, thereby ensuring the accuracy of the first best driving information finally determined.

[0159] By adopting the scheme provided in the above implementation mode, through the first condition that can be used to measure whether the predicted driving information meets the safety requirements and / or physical requirements, the first predicted driving information corresponding to the first master-slave relationship can be evaluated, and then combined with the evaluation results to determine whether to further alternate the master-slave relationship and then determine the first optimal driving information, or directly use the first predicted driving information as the first optimal driving information, thereby ensuring the accuracy of the evaluation of the predicted driving information and the safety and / or physical sensation of the first optimal driving information finally obtained.

[0160] In some possible implementations, determining the first optimal driving information based on the evaluation result corresponding to the first predicted driving information includes: alternating the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle; obtaining second predicted driving information based on the second master-slave relationship; determining the first optimal driving information from the first predicted driving information and the second predicted driving information based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information.

[0161] The present embodiment involves alternating the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle; based on the second master-slave relationship, second predicted driving information is obtained; based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined from the first predicted driving information and the second predicted driving information. The specific descriptions of the above each processing are the same as those in the aforementioned embodiment, and therefore will not be repeated herein.

[0162] The present embodiment is different from the aforementioned embodiment in that, before alternating the first master-slave relationship, the present embodiment does not need to pre-evaluate whether the evaluation result corresponding to the first predicted driving information satisfies the first condition. Instead, after obtaining the first predicted driving information under the first master-slave relationship, the first master-slave relationship is directly alternated to obtain the second master-slave relationship. After obtaining the two types of predicted driving information under the two master-slave relationships, the evaluation results of the two predicted driving information are combined to select the first optimal driving information.

[0163] The solution provided in this embodiment can reduce the processing of judging whether the first condition is met, and directly evaluate the two predicted driving information under the two master-slave relationships to select the first optimal driving information, thereby ensuring both the accuracy of the evaluation of the predicted driving information and the safety and physical feel of the first optimal driving information finally obtained.

[0164] In some possible implementations, after determining the first optimal driving information based on the evaluation result corresponding to the first predicted driving information, it may also include: generating a decision instruction based on the first optimal driving information and the evaluation result corresponding to the first optimal driving information, wherein the decision instruction is used to control the target vehicle to perform a yielding process or an overtaking process.

[0165] Specifically, the decision instruction may include the speed planning results of the target vehicle at each predicted moment in multiple predicted moments within the first time length, and the path planning results of the target vehicle within the first time length. The path planning results of the target vehicle within the first time length may include at least one of the position, orientation, etc. of the target vehicle at each predicted moment in multiple predicted moments within the first time length.

[0166] For a target vehicle, the number of all obstacles within a specified range of the target vehicle may be one or more.

[0167] In one example, for a target vehicle, the number of all obstacles within a specified range of the target vehicle may be one, that is, there is only a first obstacle within the specified range of the target vehicle.

[0168] In this example, generating a decision instruction based on the first optimal driving information and an evaluation result corresponding to the first optimal driving information may include: generating a decision instruction based on the optimal motion parameters of the first obstacle at multiple predicted moments within the first time period included in the first optimal driving information, the optimal motion parameters of the target vehicle at multiple predicted moments within the first time period, and an evaluation result corresponding to the first optimal driving information.

[0169] Based on the first optimal driving information, including the best motion parameters of the first obstacle at multiple predicted moments within the first time period, the best motion parameters of the target vehicle at multiple predicted moments within the first time period, and the evaluation result corresponding to the first optimal driving information, the processing of generating a decision instruction may be executed based on a preset strategy or by a decision model, which is not limited in this embodiment.

[0170] For example, generating a decision instruction based on the best motion parameters of the first obstacle at multiple predicted moments within the first time length included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time length may include: judging whether there is an intersection area between the target vehicle and the first obstacle based on the best motion parameters of the first obstacle at multiple predicted moments within the first time length included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time length; in the case of determining that there is an intersection area, judging whether the predicted behavior of the target vehicle is overtaking or giving way based on the best motion parameters of the first obstacle at multiple predicted moments within the first time length included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time length; if the predicted behavior of the target vehicle is overtaking, judging whether the safety evaluation value between the target vehicle and the first obstacle is higher than the safety threshold based on the evaluation result corresponding to the first best driving information, and if it is higher, generating a decision instruction for controlling the target vehicle to perform overtaking processing. Wherein, the safety evaluation value higher than the safety threshold may indicate that the collision risk between the target vehicle and the first obstacle is low, that is, the target vehicle can safely pass through the intersection area.

[0171] For example, the decision instruction may include instructions for controlling the target vehicle to perform overtaking processing relative to the first obstacle, the speed planning results of the target vehicle at each prediction moment in multiple prediction moments within the first time length, and the path planning results of the target vehicle within the first time length; or, the decision instruction may only include the speed planning results of the target vehicle at each prediction moment in multiple prediction moments within the first time length, and the path planning results of the target vehicle within the first time length, so that the target vehicle can overtake by executing the decision instruction.

[0172] For example, generating a decision instruction based on the best motion parameters of the first obstacle at multiple predicted moments within the first time period included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time period may include: judging whether there is an intersection area between the target vehicle and the first obstacle based on the best motion parameters of the first obstacle at multiple predicted moments within the first time period included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time period; in the case of determining that there is an intersection area, judging whether the predicted behavior of the target vehicle is overtaking or yielding based on the best motion parameters of the first obstacle at multiple predicted moments within the first time period included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time period; if the predicted behavior of the target vehicle is overtaking, judging whether the safety evaluation value between the target vehicle and the first obstacle is higher than the safety threshold based on the evaluation result corresponding to the first best driving information, and if not, generating a decision instruction for controlling the target vehicle to perform yielding. Wherein, the safety evaluation value not being higher than the safety threshold may indicate that the collision risk between the target vehicle and the first obstacle is high, that is, the target vehicle may not be able to safely pass through the intersection area based on the predicted motion parameters.

[0173] For example, the decision instruction may include instructions for controlling the target vehicle to perform the yielding process relative to the first obstacle, the speed planning result of the target vehicle at each of the multiple predicted moments within the first time length, and the path planning result of the target vehicle within the first time length; or, the decision instruction may only include the speed planning result of the target vehicle at each of the multiple predicted moments within the first time length, and the path planning result of the target vehicle within the first time length, so that the target vehicle yields by executing the decision instruction.

[0174] For example, generating a decision instruction based on the best motion parameters of the first obstacle at multiple predicted moments within the first time length included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time length may include: judging whether there is an intersection area between the target vehicle and the first obstacle based on the best motion parameters of the first obstacle at multiple predicted moments within the first time length included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time length; when it is determined that there is an intersection area, judging whether the predicted behavior of the target vehicle is overtaking or giving way based on the best motion parameters of the first obstacle at multiple predicted moments within the first time length included in the first best driving information and the best motion parameters of the target vehicle at multiple predicted moments within the first time length; if the predicted behavior of the target vehicle is giving way, judging whether the safety assessment value between the target vehicle and the first obstacle is higher than the safety threshold based on the assessment result corresponding to the first best driving information, and if so, generating a decision instruction for controlling the target vehicle to perform the giving way process.

[0175] It should be pointed out that the above is only an exemplary description. In actual processing, other decision-making methods may be used or a pre-trained decision model may be used to perform the above processing. The specific method of generating decision instructions is not limited or exhaustive here.

[0176] In one example, for a target vehicle, the number of all obstacles within a specified range of the target vehicle may be multiple, that is, there are multiple obstacles within the specified range of the target vehicle, and the first obstacle is any one of the multiple obstacles.

[0177] In this example, generating a decision instruction based on the first optimal driving information and an evaluation result corresponding to the first optimal driving information may include: generating a decision instruction based on the best motion parameters of the obstacle included in each of multiple optimal driving information at multiple predicted moments within the first time period, the best motion parameters of the target vehicle included in each of the optimal driving information at multiple predicted moments within the first time period, and an evaluation result corresponding to each of the optimal driving information, wherein the multiple optimal driving information includes the first optimal driving information, and different optimal driving trajectories correspond to different obstacles.

[0178] Taking the case where the number of all obstacles within the specified range of the target vehicle is 3 as an example, the 3 obstacles include a first obstacle, a second obstacle and a third obstacle. The target vehicle and the first obstacle are processed by the above implementation mode to obtain the first optimal driving information, which will not be elaborated here. The target vehicle and the second obstacle are also processed by the above implementation mode to obtain the second optimal driving information, which may include the best motion parameters of the second obstacle at multiple predicted moments within the first time length, and the best motion parameters of the target vehicle at multiple predicted moments within the first time length. The target vehicle and the third obstacle are also processed by the above implementation mode to obtain the third optimal driving information, which may include the best motion parameters of the third obstacle at multiple predicted moments within the first time length, and the best motion parameters of the target vehicle at multiple predicted moments within the first time length.

[0179] The above three optimal driving information all include the optimal motion parameters of the target vehicle at multiple predicted moments within the first time period, but the optimal motion parameters of the target vehicle at multiple predicted moments within the first time period contained in the three optimal driving information may be the same or different. This is because the above processing is performed separately on the target vehicle and each of its corresponding obstacles to generate the optimal driving information of the target vehicle and each of its corresponding obstacles. Therefore, it is necessary to further analyze and generate decision instructions in combination with the various contents contained in each optimal driving information.

[0180] For example, generating a decision instruction based on the best motion parameters of the obstacle at multiple predicted moments within the first time period included in each of the multiple best driving information, the best motion parameters of the target vehicle at multiple predicted moments within the first time period included in each of the best driving information, and the evaluation results corresponding to each of the best driving information may include: determining the best predicted position of each obstacle at each predicted moment based on the best motion parameters of the obstacle at multiple predicted moments within the first time period included in each of the multiple best driving information; vertically projecting the best predicted position of each obstacle at each predicted moment onto the path of the target vehicle within the first time period to obtain one or more projection points corresponding to each obstacle that fall into the path of the target vehicle within the first time period; determining the predicted entry moment of each obstacle into the path of the target vehicle based on the one or more projection points corresponding to each obstacle that fall into the path of the target vehicle within the first time period; calculating the time difference between the predicted entry moment of each obstacle into the path of the target vehicle and the current moment; generating a decision instruction based on the time difference corresponding to each obstacle, the best motion parameters of the target vehicle at multiple predicted moments within the first time period included in each of the best driving information, and the evaluation results corresponding to each of the best driving information.

[0181] Among them, based on one or more projection points corresponding to each obstacle in the path of the target vehicle within the first time length, determining the predicted entry time of each obstacle into the path of the target vehicle can include: determining the predicted time corresponding to each of the one or more projection points corresponding to each obstacle in the path of the target vehicle within the first time length, and taking the earliest predicted time among the predicted times corresponding to each projection point corresponding to each obstacle in the path of the target vehicle within the first time length as the predicted entry time of each obstacle into the path of the target vehicle.

[0182] Optionally, generating a decision instruction based on the time difference corresponding to each obstacle, the best motion parameters of the target vehicle at multiple predicted moments within the first time length included in each best driving information, and an evaluation result corresponding to each best driving information may include: determining the target best driving information containing the target obstacle when there is a target obstacle with a time difference less than a time difference threshold among multiple obstacles; determining the overtaking relationship between the target obstacle and the target vehicle based on the evaluation result corresponding to the target best driving information, the best motion parameters of the target obstacle at multiple predicted moments within the first time length included in the target best driving information, and the best motion parameters of the target vehicle at multiple predicted moments within the first time length; generating a decision instruction based on the overtaking relationship between the target obstacle and the target vehicle, and the best motion parameters at multiple predicted moments within the first time length corresponding to other obstacles except the target obstacle.

[0183] For example, generating a decision instruction based on the yielding relationship between the target obstacle and the target vehicle and the best motion parameters of obstacles other than the target obstacle at multiple predicted moments within the first time period may include: when the yielding relationship between the target obstacle and the target vehicle is that the target vehicle exceeds the target obstacle, adjusting the best motion parameters of the target obstacle at one or more to-be-adjusted moments after the predicted cut-in moment among the multiple predicted moments within the first time period to obtain adjusted driving information of the target obstacle; generating a decision instruction based on the adjusted driving information of the target obstacle and the best motion parameters of obstacles other than the target obstacle at multiple predicted moments within the first time period.

[0184] Adjusting the best motion parameters of the target obstacle at one or more to-be-adjusted moments after the predicted cut-in moment among multiple predicted moments within the first time length to obtain the adjusted driving information of the target obstacle may include: taking the predicted cut-in moment and each predicted moment after the target obstacle within the first time length as the to-be-adjusted moment, delaying each to-be-adjusted moment by the first time length, and obtaining each adjusted moment corresponding to each to-be-adjusted moment; taking the best motion parameters at each to-be-adjusted moment as the best motion parameters at each adjusted moment corresponding to each to-be-adjusted moment; taking the best motion parameters at each adjusted moment of the target obstacle within the first time length as the adjusted driving information of the target obstacle. The first time length may be greater than or equal to the time difference threshold. It should also be noted that in the case where the adjusted moment exceeds the range of the first time length, the adjusted moment may be deleted.

[0185] Generating a decision instruction based on the adjusted driving information of the target obstacle and the best motion parameters of other obstacles except the target obstacle at multiple predicted moments within the first time period may include: determining the yielding relationship between the target vehicle and other obstacles based on the best motion parameters of other obstacles except the target obstacle at multiple predicted moments within the first time period, and generating a decision instruction based on the adjusted driving information of the target obstacle and the yielding relationship between the target vehicle and other obstacles. The specific processing method for generating a decision instruction based on the adjusted driving information of the target obstacle and the yielding relationship between the target vehicle and other obstacles is not limited in this embodiment. For example, the decision instruction may include an instruction for controlling the target vehicle to perform overtaking processing relative to a target obstacle, an instruction for controlling the target vehicle to perform yielding processing or overtaking processing relative to each other obstacle, a speed planning result of the target vehicle at each predicted moment in a plurality of predicted moments within a first time length, and a path planning result of the target vehicle within the first time length; or, the decision instruction may only include a speed planning result of the target vehicle at each predicted moment in a plurality of predicted moments within the first time length, and a path planning result of the target vehicle within the first time length, so that the target vehicle achieves overtaking processing relative to the target obstacle, yielding processing relative to each other obstacle, or overtaking processing by executing the decision instruction.

[0186] For example, generating a decision instruction based on the yielding relationship between the target obstacle and the target vehicle and the best motion parameters at multiple predicted moments within the first duration corresponding to other obstacles except the target obstacle may include: generating a decision instruction based on the best motion parameters at multiple predicted moments within the first duration corresponding to the target obstacle and the best motion parameters at multiple predicted moments within the first duration corresponding to other obstacles except the target obstacle, when the yielding relationship between the target obstacle and the target vehicle is that the target vehicle yields to the target obstacle. In this case, the specific processing method for generating a decision instruction based on the best motion parameters at multiple predicted moments within the first duration corresponding to the target obstacle and the best motion parameters at multiple predicted moments within the first duration corresponding to other obstacles except the target obstacle is not limited in this embodiment. For example, the decision instruction may include an instruction for controlling the target vehicle to perform a yielding process relative to a target obstacle, an instruction for controlling the target vehicle to perform a yielding process relative to each other obstacle or an instruction for overtaking, a speed planning result of the target vehicle at each predicted moment in a plurality of predicted moments within a first time period, and a path planning result of the target vehicle within the first time period; or, the decision instruction may only include a speed planning result of the target vehicle at each predicted moment in a plurality of predicted moments within the first time period, and a path planning result of the target vehicle within the first time period, so that the target vehicle achieves a yielding process relative to the target obstacle, a yielding process relative to each other obstacle, or an overtaking process by executing the decision instruction.

[0187] Optionally, generating a decision instruction based on the time difference corresponding to each obstacle, the best motion parameters of the target vehicle at multiple predicted moments within the first time length included in each optimal driving information, and the evaluation result corresponding to each optimal driving information may include: in the case where there is no target obstacle with a time difference less than a time difference threshold among the multiple obstacles, determining the overtaking relationship between each obstacle and the target vehicle based on the evaluation result corresponding to each optimal driving information, the best motion parameters of the obstacle at multiple predicted moments within the first time length included in each optimal driving information, and the best motion parameters of the target vehicle at multiple predicted moments within the first time length; generating a decision instruction based on the overtaking relationship between each obstacle and the target vehicle. For example, the decision instruction may include instructions for controlling the target vehicle to perform yielding processing or overtaking processing relative to different obstacles of each obstacle, the speed planning result of the target vehicle at each predicted moment among multiple predicted moments within the first time length, and the path planning result of the target vehicle within the first time length; or, the decision instruction may only include the speed planning result of the target vehicle at each predicted moment among multiple predicted moments within the first time length, and the path planning result of the target vehicle within the first time length, so that the target vehicle can achieve yielding processing or overtaking processing relative to different obstacles of each obstacle by executing the decision instruction.

[0188] It should be pointed out that the above is only an exemplary description. In actual processing, processing can be performed based on a preset strategy or decision model to ultimately obtain a decision instruction, and no limitation or exhaustive list is made here.

[0189] Combination Figure 4 The following is an exemplary description of the process of determining a decision instruction:

[0190] First, the planning system 402 obtains input information 401, which may include evaluation information (ie, evaluation results corresponding to the first best driving information) and optimal decision trajectory reference information (ie, the first best driving information).

[0191] Then, the planning system processes the input information 401 to finally obtain a decision instruction, which may include a speed planning result 4021 and a path planning result 4022. It should be noted that in order to achieve efficient transmission of evaluation and decision information, it is necessary to open the interface between the decision module in the electronic device and the planning system to ensure that the planning system can implement overtaking according to the instructions of the decision module. For example, when deciding to overtake, the overtaking cost of the planning system can be reduced to ensure that the underlying logic plans the overtaking trajectory as much as possible. If the decision is to give way, the corresponding deceleration cost is increased using the public interface to avoid acceleration of the vehicle.

[0192] Finally, the driver of the target vehicle can also intervene in the planning system 402 through the driver input processing system 403. For example, in the case of deciding to overtake, if the acceleration is not timely, the driver can intervene in the throttle 4031, such as by lightly stepping on the accelerator pedal to help the vehicle speed up; if the steering is not timely, the driver can intervene in the steering wheel 4032 to timely adjust the driving direction of the target vehicle, etc. In this way, the driving safety and efficiency are guaranteed, while taking into account the driving experience of the user.

[0193] By adopting the above scheme, it is possible to combine the currently determined optimal driving information and its corresponding evaluation results to generate a decision execution to control the target vehicle to give way or overtake. Since the optimal driving information is selected by evaluating the predicted behaviors of multiple traffic participants, the accuracy of the final generated decision instructions can be guaranteed, thereby ensuring the safety of the target vehicle executing corresponding processing based on the decision instructions.

[0194] Combination Figure 5 and Figure 6 , the driving information prediction method provided in the embodiment of the present application is exemplarily described:

[0195] S510, interaction environment analysis and upstream signal extraction.

[0196] Specific as Figure 5 As shown, it may include: combining high-precision data and positioning information 511 to determine the motion parameters of the target vehicle at the current moment, and determining the path information of the target vehicle within the first time length according to the pre-planning system information 512, and obtaining the driving type 513 of the target vehicle within the first time length based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time length. Using perception and prediction information 514 to determine the motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle in the first time length in the future, and obtaining the driving type 515 of the first obstacle within the first time length based on the motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle in the first time length. In this way, the intentions and potential risks of the target vehicle and the first obstacle can be preliminarily understood. By integrating this information, the system can build a relatively comprehensive environmental model, laying a solid foundation for the subsequent hyper-interactive simulation, collision risk and trajectory rationality assessment.

[0197] Combination Figure 6 For example, in Figure 6 Several possible intersection scenarios between traffic participants are shown in Figure 1. Figure 6 The vehicle A in FIG. 1 represents the first obstacle, the vehicle B represents the target vehicle, the arrow of each vehicle represents the driving direction of the vehicle, and the dotted line extending from the arrow can represent the driving path or predicted position within the first time period. Figure 6In the scene 601 in FIG. 1 , the driving type of vehicle B may be going straight, and the driving type of vehicle A may be turning left. Figure 6 In the scene 602 in FIG. 1 , the driving type of vehicle B may be a right turn, and the driving type of vehicle A is a straight drive. Figure 6 In the scene 603 in FIG. 1 , the driving type of vehicle B may be a left turn, and the driving type of vehicle A may be a right turn. Figure 6 In the scene 604, the driving type of vehicle B may be straight driving, and the driving type of vehicle A may be lane changing.

[0198] S520, integrating road right information and determining the master-slave relationship.

[0199] Specific as Figure 5 As shown, it may include: based on the driving type 513 of the target vehicle within the first time period, the driving type 515 of the first obstacle within the first time period, and the traffic flow strategy 521 (the traffic flow strategy is the traffic type of the first obstacle), obtaining the road right priority 522 of the target vehicle and the first obstacle.

[0200] S530, allows super interactive simulation and secondary game deduction.

[0201] Specific as Figure 5 As shown, it may include: based on the right-of-way priority 522 of the target vehicle and the first obstacle, establishing a master-slave relationship 531 to obtain a first master-slave relationship;

[0202] Performing forward simulation 532 based on the first master-slave relationship to obtain first predicted driving information, for example, possible collision risks and trajectory changes may be predicted through forward simulation, and an optimal driving trajectory, namely the first predicted driving information, may be determined through iterative calculation;

[0203] The first predicted driving information is evaluated and a secondary game is performed 533. For example, an evaluation result of the first predicted driving information is obtained by evaluating the first predicted driving information. According to the evaluation result, a decision is made whether a master-slave relationship swap and a secondary game are required, and finally the first optimal driving information is obtained, thereby improving the evaluation stability and decision reliability of the vehicle in an uncertain environment.

[0204] by Figure 6 Taking the scenario 601 shown in FIG. 1 as an example, vehicle B has a higher right of way priority than vehicle A. However, based on the initial predicted motion parameters of vehicle A and vehicle B, it is found that vehicle A will arrive at the predicted intersection position earlier than vehicle B (i.e. Figure 6 In 601, the intersection of the driving paths (i.e., the two dashed lines) of vehicle A and vehicle B in the first time period is determined, then in the first master-slave relationship finally determined, vehicle A is in a dominant position and vehicle B is in a subordinate position. Figure 6 The exemplary descriptions of other scenarios are similar to scenario 601 and are not described in detail.

[0205] S540, the evaluation results are output and the decision instructions are issued.

[0206] Specific as Figure 5 As shown, it may include: planning and processing based on the first optimal driving information obtained by evaluation and secondary game 533 and its corresponding evaluation result 541 to obtain path planning results and speed planning results 542; in addition, the driver can also intervene in this process and input intervention information 543 to the path planning results and speed planning results, such as throttle intervention or steering wheel intervention. It should be understood that in this process, the transmission of information is not only a simple overtaking result and collision probability, but more importantly, a signal that the planning system can understand and use. In order to achieve this goal, the decision-making and evaluation system needs to be fully adapted to the planning module, and information transmission needs to use a dedicated interface and follow a specific format and protocol to ensure the integrity and timeliness of the data, thereby ensuring the efficient coordination of the decision-making and planning systems.

[0207] Figure 7 A schematic block diagram of a driving information prediction device provided by an embodiment of the present disclosure is shown.

[0208] like Figure 7 As shown, including:

[0209] A master-slave relationship determination module 701 is used to determine a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period;

[0210] A driving information prediction module 702 is used to obtain first predicted driving information based on the first master-slave relationship, wherein the first predicted driving information includes: first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and second predicted motion parameters of the first obstacle at multiple predicted moments within the first time length;

[0211] The optimal driving information determination module 703 is used to determine the first optimal driving information based on the evaluation result corresponding to the first predicted driving information, wherein the first optimal driving information includes the optimal predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the optimal predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0212] The driving information prediction module is used to perform one of the following: when the evaluation result corresponding to the first predicted driving information meets the first condition, use the first predicted driving information as the first optimal driving information; when the evaluation result corresponding to the first predicted driving information does not meet the first condition, alternate the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle, obtain second predicted driving information based on the second master-slave relationship, and determine the first optimal driving information from the first predicted driving information and the second predicted driving information based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information.

[0213] The first condition includes at least one of the following: a first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than a safety threshold, and a first somatosensory assessment value in the assessment result corresponding to the first predicted driving information is higher than a somatosensory assessment threshold.

[0214] The optimal driving information determination module is used to alternate the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle; based on the second master-slave relationship, obtain second predicted driving information; based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, determine the first optimal driving information from the first predicted driving information and the second predicted driving information.

[0215] The second predicted driving information includes: third predicted motion parameters of the target vehicle at multiple predicted moments within the first time period, and fourth predicted motion parameters of the first obstacle at multiple predicted moments within the first time period.

[0216] like Figure 8 As shown, the device further includes a security assessment module 801.

[0217] The safety assessment module is used to perform at least one of the following: based on the safety assessment formula, the first master-slave relationship and the first predicted driving information, calculate a first safety assessment value in the assessment result corresponding to the first predicted driving information; based on the somatosensory assessment formula, the first master-slave relationship and the first predicted driving information, calculate a first somatosensory assessment value in the assessment result corresponding to the first predicted driving information; based on the safety assessment formula, the second master-slave relationship and the second predicted driving information, calculate a second safety assessment value in the assessment result corresponding to the second predicted driving information; based on the somatosensory assessment formula, the second master-slave relationship and the second predicted driving information, calculate a second somatosensory assessment value in the assessment result corresponding to the second predicted driving information.

[0218] The optimal driving information determination module is used to determine a first reference value corresponding to the first predicted driving information based on an evaluation result corresponding to the first predicted driving information; determine a second reference value corresponding to the second predicted driving information based on an evaluation result corresponding to the second predicted driving information; and select the first optimal driving information from the first predicted driving information and the second predicted driving information based on the maximum value of the first reference value and the second reference value.

[0219] The driving information prediction module is used to perform forward simulation based on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; the first predicted motion parameters of the target vehicle at the multiple predicted moments within the first time length, and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length are used as the first predicted driving information.

[0220] The driving information prediction module is used to perform forward simulation based on the second master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain the third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; the third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length are used as the second predicted driving information.

[0221] The master-slave relationship determination module is used to determine the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time length, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time length; based on the right-of-way priority of the target vehicle and the first obstacle within the first time length, determine the first master-slave relationship between the target vehicle and the first obstacle.

[0222] The master-slave relationship determination module is used to obtain the driving type of the target vehicle within the first time length based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time length; obtain the driving type of the first obstacle within the first time length based on the motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle within the first time length; obtain the road right priority of the target vehicle and the first obstacle within the first time length based on the driving type of the target vehicle within the first time length and the driving type of the first obstacle within the first time length.

[0223] The master-slave relationship determination module is used to determine the predicted intersection area based on the initial predicted motion parameters of the target vehicle at multiple predicted moments within a first time length and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length; determine the first master-slave relationship between the target vehicle and the first obstacle based on the road right priority of the target vehicle and the first obstacle within the first time length, the predicted intersection area and reference information, and the reference information includes at least one of the following: the driving type of the target vehicle, the driving type of the first obstacle, the motion parameters of the target vehicle at the current moment, the motion parameters of the first obstacle at the current moment, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

[0224] like Fig. 9 As shown, the device further includes a decision instruction generating module 901.

[0225] The decision instruction generation module is used to generate a decision instruction based on the first optimal driving information and an evaluation result corresponding to the first optimal driving information, wherein the decision instruction is used to control the target vehicle to perform a yielding process or an overtaking process.

[0226] For the description of specific functions and examples of each module and submodule of the device in the embodiment of the present disclosure, reference can be made to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0227] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0228] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0229] Fig.10A schematic block diagram of an example electronic device 1000 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 assistants, 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.

[0230] like Fig.10 As shown, the electronic device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the electronic device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0231] Multiple components in the electronic device 1000 are connected to the I / O interface 1005, including: an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the electronic device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0232] The computing unit 1001 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1001 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 1001 performs the various methods and processes described above. For example, in some embodiments, the above method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1000 via ROM 1002 and / or communication unit 1009. When the computer program is loaded into RAM 1003 and executed by the computing unit 1001, at least one step of the method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to perform the above method in any other appropriate manner (e.g., by means of firmware).

[0233] According to another aspect of the present disclosure, an autonomous driving vehicle is provided, comprising the above-mentioned electronic device.

[0234] 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.

[0235] The program code for implementing the method of the present disclosure may be written in any combination of at least one programming language. These program codes may be provided 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, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0236] 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 at least one line, 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.

[0237] To provide interaction with a user, the systems and techniques described herein can 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; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0238] 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 computer with a graphical user interface or a web browser through which a user 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.

[0239] 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.

[0240] 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.

[0241] 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 principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A driving information prediction method, comprising: Determine a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period; Based on the first master-slave relationship, first predicted driving information is obtained, wherein the first predicted driving information includes: first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and second predicted motion parameters of the first obstacle at multiple predicted moments within the first time length; Based on the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined, wherein the first optimal driving information includes the best predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the best predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

2. The method according to claim 1, wherein: The determining the first optimal driving information based on the evaluation result corresponding to the first predicted driving information includes one of the following: When the evaluation result corresponding to the first predicted driving information satisfies a first condition, using the first predicted driving information as the first optimal driving information; When the evaluation result corresponding to the first predicted driving information does not meet the first condition, the first master-slave relationship is alternated to obtain a second master-slave relationship between the target vehicle and the first obstacle, and second predicted driving information is obtained based on the second master-slave relationship. Based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined from the first predicted driving information and the second predicted driving information.

3. The method according to claim 2, wherein: The first condition includes at least one of the following: a first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than a safety threshold, and a first somatosensory assessment value in the assessment result corresponding to the first predicted driving information is higher than a somatosensory assessment threshold.

4. The method according to claim 1, wherein: The determining the first optimal driving information based on the evaluation result corresponding to the first predicted driving information includes: Alternating the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle; Based on the second master-slave relationship, obtaining second predicted driving information; Based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined from the first predicted driving information and the second predicted driving information.

5. The method according to any one of claims 2 to 4, wherein: The second predicted driving information includes: third predicted motion parameters of the target vehicle at multiple predicted moments within the first time period, and fourth predicted motion parameters of the first obstacle at multiple predicted moments within the first time period.

6. The method according to any one of claims 2 to 4, further comprising at least one of the following: Based on the safety assessment formula, the first master-slave relationship and the first predicted driving information, a first safety assessment value in the assessment result corresponding to the first predicted driving information is calculated; Based on the somatosensory evaluation formula, the first master-slave relationship and the first predicted driving information, a first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is calculated; Based on the safety assessment formula, the second master-slave relationship and the second predicted driving information, a second safety assessment value in the assessment result corresponding to the second predicted driving information is calculated; Based on the somatosensory evaluation formula, the second master-slave relationship and the second predicted driving information, a second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information is calculated.

7. The method according to any one of claims 2 to 4, wherein: The determining the first optimal driving information from the first predicted driving information and the second predicted driving information based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information includes: determining a first reference value corresponding to the first predicted driving information based on an evaluation result corresponding to the first predicted driving information; determining a second reference value corresponding to the second predicted driving information based on an evaluation result corresponding to the second predicted driving information; Based on the maximum value of the first reference value and the second reference value, the first optimal driving information is selected from the first predicted driving information and the second predicted driving information.

8. The method according to claim 1, wherein: The obtaining first predicted driving information based on the first master-slave relationship includes: Performing forward simulation based on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; The first predicted motion parameters of the target vehicle at multiple predicted moments within the first time period and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time period are used as the first predicted driving information.

9. The method according to any one of claims 2 to 4, wherein: The obtaining second predicted driving information based on the second master-slave relationship includes: Performing forward simulation based on the second master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; The third predicted motion parameters of the target vehicle at multiple predicted moments within the first time period and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time period are used as the second predicted driving information.

10. The method according to claim 1, wherein: The determining of a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period includes: Determine the right-of-way priority of the target vehicle and the first obstacle within the first time period based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period; A first master-slave relationship between the target vehicle and the first obstacle is determined based on the right-of-way priorities of the target vehicle and the first obstacle within a first time period.

11. The method according to claim 10, wherein: The determining the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the motion parameter of the target vehicle at the current moment, the path information of the target vehicle within the first time length, the motion parameter of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time length includes: Based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time period, the driving type of the target vehicle within the first time period is obtained; Obtaining a travel type of the first obstacle within the first time period based on a motion parameter of the first obstacle at a current moment and predicted path information of the first obstacle within a first time period; Based on the driving type of the target vehicle within the first time period and the driving type of the first obstacle within the first time period, the right-of-way priorities of the target vehicle and the first obstacle within the first time period are obtained.

12. The method according to claim 10, wherein: The determining a first master-slave relationship between the target vehicle and the first obstacle based on the right-of-way priorities of the target vehicle and the first obstacle within a first time period includes: Determining a predicted intersection area based on initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time period and initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time period; The first master-slave relationship between the target vehicle and the first obstacle is determined based on the right-of-way priority of the target vehicle and the first obstacle within a first time length, the predicted intersection area and reference information, wherein the reference information includes at least one of the following: the driving type of the target vehicle, the driving type of the first obstacle, the motion parameters of the target vehicle at a current moment, the motion parameters of the first obstacle at the current moment, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

13. The method according to claim 1, wherein: After determining the first optimal driving information based on the evaluation result corresponding to the first predicted driving information, the method further includes: Based on the first optimal driving information and an evaluation result corresponding to the first optimal driving information, a decision instruction is generated, wherein the decision instruction is used to control the target vehicle to perform a yielding process or an overtaking process.

14. A driving information prediction device, comprising: A master-slave relationship determination module, used to determine a first master-slave relationship between the target vehicle and the first obstacle based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time period, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time period; A driving information prediction module, configured to obtain first predicted driving information based on the first master-slave relationship, wherein the first predicted driving information includes: first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and second predicted motion parameters of the first obstacle at multiple predicted moments within the first time length; The optimal driving information determination module is used to determine the first optimal driving information based on the evaluation result corresponding to the first predicted driving information, wherein the first optimal driving information includes the optimal predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the optimal predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

15. The device according to claim 14, wherein: The driving information prediction module is used to perform one of the following: When the evaluation result corresponding to the first predicted driving information satisfies a first condition, using the first predicted driving information as the first optimal driving information; When the evaluation result corresponding to the first predicted driving information does not meet the first condition, the first master-slave relationship is alternated to obtain a second master-slave relationship between the target vehicle and the first obstacle, and second predicted driving information is obtained based on the second master-slave relationship. Based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined from the first predicted driving information and the second predicted driving information.

16. The device according to claim 15, wherein: The first condition includes at least one of the following: a first safety assessment value in the assessment result corresponding to the first predicted driving information is higher than a safety threshold, and a first somatosensory assessment value in the assessment result corresponding to the first predicted driving information is higher than a somatosensory assessment threshold.

17. The device according to claim 14, wherein: The optimal driving information determination module is used to alternate the first master-slave relationship to obtain a second master-slave relationship between the target vehicle and the first obstacle; Based on the second master-slave relationship, obtaining second predicted driving information; Based on the evaluation result corresponding to the second predicted driving information and the evaluation result corresponding to the first predicted driving information, the first optimal driving information is determined from the first predicted driving information and the second predicted driving information.

18. The device according to any one of claims 15 to 17, wherein: The second predicted driving information includes: third predicted motion parameters of the target vehicle at multiple predicted moments within the first time period, and fourth predicted motion parameters of the first obstacle at multiple predicted moments within the first time period.

19. The apparatus according to any one of claims 15 to 17, further comprising a security assessment module; the security assessment module is configured to perform at least one of the following: Based on the safety assessment formula, the first master-slave relationship and the first predicted driving information, a first safety assessment value in the assessment result corresponding to the first predicted driving information is calculated; Based on the somatosensory evaluation formula, the first master-slave relationship and the first predicted driving information, a first somatosensory evaluation value in the evaluation result corresponding to the first predicted driving information is calculated; Based on the safety assessment formula, the second master-slave relationship and the second predicted driving information, a second safety assessment value in the assessment result corresponding to the second predicted driving information is calculated; Based on the somatosensory evaluation formula, the second master-slave relationship and the second predicted driving information, a second somatosensory evaluation value in the evaluation result corresponding to the second predicted driving information is calculated.

20. The device according to any one of claims 15 to 17, wherein: The optimal driving information determination module is used to determine a first reference value corresponding to the first predicted driving information based on an evaluation result corresponding to the first predicted driving information; Based on the evaluation result corresponding to the second predicted driving information, determine the second reference value corresponding to the second predicted driving information; based on the maximum value of the first reference value and the second reference value, select the first optimal driving information from the first predicted driving information and the second predicted driving information.

21. The device according to claim 14, wherein: The driving information prediction module is used to perform forward simulation based on the first master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain the first predicted motion parameters of the target vehicle at multiple predicted moments within a first time length, and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; the first predicted motion parameters of the target vehicle at the multiple predicted moments within the first time length, and the second predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length are used as the first predicted driving information.

22. The device according to any one of claims 15 to 17, wherein: The driving information prediction module is used to perform forward simulation based on the second master-slave relationship, the motion parameters of the target vehicle at the current moment, and the motion parameters of the first obstacle at the current moment, to obtain the third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length; the third predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the fourth predicted motion parameters of the first obstacle at the multiple predicted moments within the first time length are used as the second predicted driving information.

23. The device according to claim 14, wherein: The master-slave relationship determination module is used to determine the right-of-way priority of the target vehicle and the first obstacle within the first time length based on the motion parameters of the target vehicle at the current moment, the path information of the target vehicle within the first time length, the motion parameters of the first obstacle at the current moment, and the predicted path information of the first obstacle within the first time length; A first master-slave relationship between the target vehicle and the first obstacle is determined based on the right-of-way priorities of the target vehicle and the first obstacle within a first time period.

24. The device according to claim 23, wherein: The master-slave relationship determination module is used to obtain the driving type of the target vehicle within the first time length based on the motion parameters of the target vehicle at the current moment and the path information of the target vehicle within the first time length; and obtain the driving type of the first obstacle within the first time length based on the motion parameters of the first obstacle at the current moment and the predicted path information of the first obstacle within the first time length; Based on the driving type of the target vehicle within the first time period and the driving type of the first obstacle within the first time period, the right-of-way priorities of the target vehicle and the first obstacle within the first time period are obtained.

25. The device according to claim 23, wherein: The master-slave relationship determination module is used to determine the predicted intersection area based on the initial predicted motion parameters of the target vehicle at multiple predicted moments within a first time length and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length; determine the first master-slave relationship between the target vehicle and the first obstacle based on the road right priority of the target vehicle and the first obstacle within the first time length, the predicted intersection area and reference information, and the reference information includes at least one of the following: the driving type of the target vehicle, the driving type of the first obstacle, the motion parameters of the target vehicle at the current moment, the motion parameters of the first obstacle at the current moment, the initial predicted motion parameters of the target vehicle at multiple predicted moments within the first time length, and the initial predicted motion parameters of the first obstacle at multiple predicted moments within the first time length.

26. The device according to claim 14, wherein: Also includes: A decision instruction generation module is used to generate a decision instruction based on the first optimal driving information and an evaluation result corresponding to the first optimal driving information, wherein the decision instruction is used to control the target vehicle to perform a yielding process or an overtaking process.

27. 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 13.

28. 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 1-13.

29. 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 13.

30. An autonomous driving vehicle comprising the electronic device as claimed in claim 27.

Images