Vehicle Lane Changing Decision Method, Device, Electronic Device, and Storage Medium

Abstract

An embodiment of the present disclosure discloses a method, apparatus, electronic device, and storage medium for vehicle lane-changing decision-making. The method includes: obtaining an environmental scenario in which an autonomous vehicle is located; determining a driving strategy according to the environmental scenario and lane-changing encodings of various preset influencing factors; determining a utility coefficient according to the driving strategy; obtaining characteristic utilities of the preset influencing factors; determining a total lane-changing utility according to the utility coefficient and the characteristic utilities; determining the maximum value of the total lane-changing utility according to a lane-changing behavior, and determining the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the autonomous vehicle. The vehicle lane-changing decision-making in the embodiment of the present disclosure solves the problem of rule conflicts or algorithmic coverage difficulties in complex scenarios, designs a scheme for encoding combinations of lane-changing encodings of preset influencing factors to associate with environmental scenarios, efficiently reflects the expectations of different scenarios for lane-changing decision-making behaviors, and is easy to expand and customize development.

Description

Technical Field

[0001] The present disclosure relates to the field of autonomous driving technology, and in particular, to a vehicle lane-changing decision method, device, electronic device, and storage medium. Background Art

[0002] As the application of driverless technology in more extensive and complex road environments, traditional lane-changing decision algorithms based on rules or knowledge reasoning are difficult to cover a wide range of scenarios, and it is also difficult to meet different expectations of human beings for the lane-changing behavior of driverless vehicles in different scenarios. Moreover, deep learning methods based on reinforcement learning or data-driven lack engineering reliability and rely on a large amount of high-quality human driving data, making it difficult to stably apply in engineering. Therefore, the present invention provides a vehicle lane-changing decision method that associates environmental scenarios with the total lane-changing utility, solving the problems that traditional lane-changing decision methods cannot match multiple scenarios and rule conflicts or lane-changing algorithms are difficult to cover in complex scenarios. Summary of the Invention

[0003] To solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide a vehicle lane-changing decision method, device, electronic device, and storage medium, which solve the problems of rule conflicts or algorithm coverage difficulties in complex scenarios, design a scheme for encoding combinations of lane-changing encodings of preset influencing factors to associate with environmental scenarios, efficiently reflect the expectations of different scenarios for lane-changing decision-making behaviors, and are easy to expand and customize development.

[0004] In a first aspect, embodiments of the present disclosure provide a vehicle lane-changing decision method, which includes:

[0005] Obtain the environmental scenario where the driverless vehicle is located;

[0006] Determine a driving strategy according to the environmental scenario and the lane-changing encodings of each preset influencing factor;

[0007] Determine a utility coefficient according to the driving strategy;

[0008] Obtain the characteristic utility of the preset influencing factor;

[0009] Determine the total lane-changing utility according to the utility coefficient and the characteristic utility;

[0010] According to the lane-changing behavior, determine the maximum value of the total lane-changing utility, and determine the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle.

[0011] In a second aspect, embodiments of the present disclosure further provide a vehicle lane-changing decision device, which includes:

[0012] A first acquisition module, configured to obtain the environmental scenario where the driverless vehicle is located;

[0013] A first determination module, configured to determine a driving strategy according to the environmental scenario and the lane-changing codes of each preset influencing factor;

[0014] A second determination module, configured to determine a utility coefficient according to the driving strategy;

[0015] A second acquisition module, configured to acquire the characteristic utility of the preset influencing factor;

[0016] A third determination module, configured to determine the total lane-changing utility according to the utility coefficient and the characteristic utility;

[0017] A fourth determination module, configured to determine the maximum value of the total lane-changing utility according to the lane-changing behavior, and determine the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle.

[0018] In a third aspect, an embodiment of the present disclosure further provides an electronic device, where the electronic device includes: one or more processors; a storage device, configured to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle lane-changing decision method as described above.

[0019] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the vehicle lane-changing decision method as described above is implemented.

[0020] A vehicle lane-changing decision method provided by an embodiment of the present disclosure includes: acquiring an environmental scenario where a driverless vehicle is located; determining a driving strategy according to the environmental scenario and the lane-changing codes of each preset influencing factor; determining a utility coefficient according to the driving strategy; acquiring the characteristic utility of the preset influencing factor; determining the total lane-changing utility according to the utility coefficient and the characteristic utility; determining the maximum value of the total lane-changing utility according to the lane-changing behavior, and determining the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle. The technical solution of the present disclosure solves the problem of rule conflicts or algorithm coverage difficulties in complex scenarios, designs a scheme for encoding combinations of lane-changing codes of preset influencing factors to associate with environmental scenarios, efficiently reflects the expectations of different scenarios for lane-changing decision-making behaviors, and is easy to expand and customize. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the embodiments of the present disclosure will become more obvious. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the original elements and elements are not necessarily drawn to scale.

[0022] Figure 1Flow chart of a vehicle lane - changing decision - making method in an embodiment of the present disclosure;

[0023] Figure 2 Structural schematic diagram of an off - ramp scenario in an embodiment of the present disclosure;

[0024] Figure 3 Structural schematic diagram of a vehicle lane - changing decision - making device in an embodiment of the present disclosure;

[0025] Figure 4 Structural schematic diagram of an electronic device in an embodiment of the present disclosure. Detailed implementation manners

[0026] Embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0027] It should be noted that concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0028] The names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and do not limit the scope of these messages or information.

[0029] Figure 1 Flow chart of a vehicle lane - changing decision - making method in an embodiment of the present disclosure. This embodiment is applicable to the lane - changing decision - making of driverless vehicles in different environmental scenarios. This method can be executed by a vehicle lane - changing decision - making device, which can be implemented in software and / or hardware, and can be configured in an electronic device. As Figure 1 shown, the method can specifically include the following steps:

[0030] S110. Obtain the environmental scenario where the driverless vehicle is located.

[0031] According to the sensors of the driverless vehicle, such as radar or camera, identify the environmental scenario where the vehicle is located, or obtain the already stored environmental scenario of the vehicle from the memory; wherein, the environmental scenario can be a normal road driving scenario, a pulling - over or station - entering scenario, a traffic - light queuing scenario, a dead - end road or off - ramp scenario, a construction section scenario, etc.

[0032] S120. Determine the driving strategy according to the lane-changing codes of the environmental scenario and each preset influencing factor.

[0033] In one example, determining the driving strategy according to the lane-changing codes of the environmental scenario and each preset influencing factor includes: combining the lane-changing codes of each preset influencing factor according to the environmental scenario to determine the driving strategy.

[0034] Specifically, the preset influencing factors may be safety factors, trafficability factors, and rule factors. Among them, the safety factor is that the lane-changing behavior may collide with vehicles on the side or behind; the rule factor is that the lane-changing behavior may violate traffic rules, such as crossing the solid line; the trafficability factor is whether the lane-changing behavior brings benefits in terms of trafficability, such as overtaking to avoid obstacles. It can be understood that, for example, if there is nothing in front of the lane, the trafficability is good; if there is a slow vehicle in front, the trafficability is relatively poor; and if there is a static obstacle, it is even worse.

[0035] The preset influencing factors may also be the continuing route factor or the driving factor. The continuing route factor is whether the lane-changing behavior considers reaching the destination of the route, such as the lane-changing behavior at the on-ramp or off-ramp; the driving factor is that the driverless vehicle tends to drive in the right lane or the center lane.

[0036] The lane-changing codes of the preset influencing factors are preset. The lane-changing codes of the preset influencing factors may be the safety lane-changing code of the safety factor, the trafficability lane-changing code of the trafficability factor, the rule lane-changing code of the rule factor, the continuing route lane-changing code of the continuing route factor, or the driving lane-changing code of the driving factor.

[0037] It can be understood that the safety lane-changing code 00 means that the lane-changing decision does not consider the safety factor; the safety lane-changing code 01 means that the lane-changing decision slightly considers the safety factor; the safety lane-changing code 10 means that the lane-changing decision normally considers the safety factor; the safety lane-changing code 11 means that the lane-changing decision conservatively considers the safety factor;

[0038] The rule lane-changing code 00 means that the lane-changing decision does not consider the rule factor; the rule lane-changing codes 01, 10, and 11 mean that the lane-changing decision considers the rule factor;

[0039] The trafficability lane-changing code 00 means that the lane-changing decision does not consider the trafficability factor; the trafficability lane-changing code 01 means that the lane-changing decision slightly considers the trafficability factor; the trafficability lane-changing code 10 means that the lane-changing decision normally considers the trafficability factor; the trafficability lane-changing code 11 means that the lane-changing decision seriously considers the trafficability factor;

[0040] The continue route lane change code 00 indicates that the lane change decision does not consider the continue route factor. The continue route lane change code 01 indicates that the lane change decision slightly considers the continue route factor. The continue route lane change code 10 indicates that the lane change decision normally considers the continue route factor. The continue route lane change code 11 indicates that the lane change decision seriously considers the continue route factor;

[0041] The driving lane change code 00 indicates that the lane change decision does not consider the driving factor. The driving lane change code 01 indicates that the lane change decision slightly considers the driving factor. The driving lane change code 10 indicates that the lane change decision normally considers the driving factor. The driving lane change code 11 indicates that the lane change decision seriously considers the driving factor.

[0042] Exemplarily, in the normal road driving scenario, safety factors, passability factors, rule factors, continue route factors, and driving factors are all normally considered. Therefore, the rule lane change code is 10, the safety lane change code is 10, the passability lane change code is 10, the continue route lane change code is 10, and the driving lane change code is 10. At this time, the code combination is 1010101010. The code combination is the driving strategy. For simplicity, decimal conversion is performed, and the driving strategy obtained is 682.

[0043] In the scenario of pulling over or entering a station, if there are pedestrians or obstacles in the station, it is necessary to carefully consider changing lanes to avoid obstacles. The passability factor should be considered more slightly, and the obligation to drive on the right should be considered more seriously. The remaining factors are normally considered. Therefore, the rule lane change code is 10, the safety lane change code is 10, the passability lane change code is 01, the continue route lane change code is 10, and the driving lane change code is 11. At this time, the code combination is 1010011011. The code combination is the driving strategy. For simplicity, decimal conversion is performed, and the driving strategy obtained is 667.

[0044] In the traffic light queuing scenario, the vehicle should not consider changing lanes to avoid obstacles because of the queuing vehicles. The passability factor should be ignored; there is no need to consider the obligation to drive on the right or in the middle. The remaining factors should be normally considered. Therefore, the rule lane change code is 10, the safety lane change code is 10, the passability lane change code is 00, the continue route lane change code is 10, and the driving lane change code is 00. At this time, the code combination is 1010001000. The code combination is the driving strategy. For simplicity, decimal conversion is performed, and the driving strategy obtained is 648.

[0045] In scenarios such as dead-end road scenarios, on-ramp scenarios, or off-ramp scenarios, the vehicle needs to change lanes to a drivable lane as much as possible, and the continue route factor should be seriously considered; when there are low-speed obstacles in the target lane, the vehicle should also change lanes as much as possible and adopt a following driving strategy, and the passability factor should be considered more lightly; in some cases, the vehicle needs to execute a driving mode of overtaking the vehicle behind in the target lane, and the safety factor should be considered more lightly. The remaining factors should be considered normally. Therefore, the rule lane change code is 10, the safety lane change code is 01, the passability lane change code is 01, the continue route lane change code is 11, and the driving lane change code is 10. At this time, the code combination is 1001011110. After decimal conversion, the driving strategy obtained is 606.

[0046] In the construction section scenario, the vehicle should avoid obstacles and pass as soon as possible, and the passability factor should be seriously considered, and the remaining factors should be considered normally. Therefore, the rule lane change code is 10, the safety lane change code is 10, the passability lane change code is 11, the continue route lane change code is 10, and the driving lane change code is 10. At this time, the code combination is 1010111010. After decimal conversion, the driving strategy obtained is 698.

[0047] The association relationship between the lane change code of each preset influencing factor and the utility coefficient in the code combination (driving strategy) can be preset. For example, as shown in Table 1:

[0048] Table 1

[0049]

[0050] Specifically, the utility coefficient corresponding to the influencing factor of the driving strategy can also be preset. For example, in the construction section scenario, the traffic regulation construction section scenario is 1, the safety utility coefficient is 0.3, the passability utility coefficient is 0.4, the continue route utility coefficient is 0.3, and the driving utility coefficient is 0.75.

[0051] It can be understood that Table 1 can be pre-stored in the planning and decision-making module of the autonomous vehicle, or in the cloud, etc. The same driving strategy can match different environmental scenarios. For example, the driving strategy 606 can match the dead-end road scenario, the on-ramp scenario, or the off-ramp scenario. When a new environmental scenario occurs and the existing driving strategies cannot meet the requirements, the number of preset influencing factors and / or the number of lane change codes can be adjusted to re-match the code combination (driving strategy) corresponding to the environmental scenario, and then the new environmental scenario code combination and the utility coefficient corresponding to the code combination are added to Table 1. The advantage of this processing is that there is no need to modify the top-level decision algorithm logic, making the decision algorithm easy to expand and maintain, being able to flexibly adapt to more complex road conditions, and obtaining reasonable decision results in different environmental scenarios.

[0052] In one embodiment, the utility coefficient is the rule utility coefficient of the rule factor, the safety utility coefficient of the safety factor, the trafficability utility coefficient of the trafficability factor, the continuation route utility coefficient of the continuation route factor, or the driving utility coefficient of the driving factor, wherein the sum of the safety utility coefficient, the trafficability utility coefficient, and the continuation route utility coefficient is 1.

[0053] As shown in Table 1, the utility coefficient is the rule utility coefficient γ1 of the rule factor, the safety utility coefficient γ2 of the safety factor, the trafficability utility coefficient γ3 of the trafficability factor, the continuation route utility coefficient γ4 of the continuation route factor, or the driving utility coefficient γ5 of the driving factor, wherein the sum of the safety utility coefficient γ2, the trafficability utility coefficient γ3, and the continuation route utility coefficient γ4 is 1. Among them, γ1 and γ5 are independent, γ2, γ3, γ4 satisfy the relationship γ2 + γ3 + γ4 = 1, and the utility coefficient is taken to two decimal places.

[0054] S130. Determine the utility coefficient according to the driving strategy;

[0055] Based on the above embodiments, after identifying the current environmental scenario of the vehicle, match the driving strategy suitable for the current scenario. After the driving strategy (coding combination), obtain the utility coefficient corresponding to the lane-changing code of the preset influencing factor in the coding combination. Among them, for different environmental scenarios, the utility coefficients corresponding to the lane-changing codes of the same preset influencing factor are different.

[0056] Exemplarily, according to Table 1, when it is identified that the driving strategy is 628, that is, in the normal road driving scenario, the rule utility coefficient γ1 corresponding to the rule lane-changing code 10 of the rule factor in Table 1 can be obtained as 1, the safety utility coefficient γ2 corresponding to the safety lane-changing code 10 of the safety factor is 0.33, the trafficability utility coefficient γ3 corresponding to the trafficability lane-changing code 10 of the trafficability factor is 0.33, the continuation route utility coefficient γ4 corresponding to the continuation route lane-changing code 10 of the continuation route factor is 0.33, and the driving utility coefficient γ5 corresponding to the driving lane-changing code 10 of the driving factor is 0.75.

[0057] S140. Obtain the characteristic utility of the preset influencing factor.

[0058] In one embodiment, the characteristic utility of the preset influencing factor is the safety characteristic utility of the safety factor, the trafficability characteristic utility of the trafficability factor, the rule characteristic utility of the rule factor, the continuation route characteristic utility of the continuation route factor, or the characteristic utility of the driving factor.

[0059] In one embodiment, obtaining the characteristic utility of the preset influencing factors includes: obtaining the safety characteristic utility of the safety factor according to the interaction relationship between the lane-changing behavior trajectory and the vehicle behind, where the expression of the safety characteristic utility is:

[0060]

[0061] Where U safe (a ot ) represents the safety characteristic utility, a ot represents the optimal acceleration for cutting in, a ot = 100 represents the acceleration when cutting in is not possible, [a lb , a ub represents the acceleration interval of the sigmoid function, [s ub , s lb represents the interval where the original normalized value is mapped by the sigmoid function;

[0062] Obtaining the rule characteristic utility of the rule factor according to the prediction of whether the lane-changing behavior trajectory exits the road network or crosses the solid line, where the expression of the rule characteristic utility is:

[0063]

[0064] Where U rule (traj) represents the rule characteristic utility, traj obey rule means that the lane-changing behavior trajectory does not exit the road network or does not cross the solid line, otherwise means that the lane-changing behavior trajectory exits the road network or crosses the solid line;

[0065] Obtaining the passability characteristic utility of the passability factor according to the static collision distance, collision time, and obstacle speed of the lane-changing behavior trajectory colliding with the obstacle in front, where the expression of the passability characteristic utility is:

[0066]

[0067] Where U passable (sd, v, ttc) represents the passability characteristic utility, sd represents the static collision distance between the lane-changing behavior trajectory and the obstacle in front, v represents the obstacle speed, ttc represents the collision time of the lane-changing behavior trajectory colliding with the obstacle in front, [v lb1 , v ub1 , [v lb2 , v ub2 , [v lb3 , v ub3 respectively represent the static collision distance interval, obstacle speed interval, and collision time interval of the sigmoid function; [s ub , slb represents the interval where the original normalized value is mapped by the sigmoid function;

[0068] According to the lane-changing behavior trajectory and the effective distance, obtain the continuing route feature utility of the continuing route factor, where the expression of the continuing route feature utility is:

[0069]

[0070] Among them, U continu (D) represents the continuing route feature utility, D represents the effective distance, that is, the distance from the current position of the driverless vehicle to the lane fork, [v ub , v lb represents the normalized interval of the effective distance, [s ub , s lb represents the interval where the original normalized value is mapped by the sigmoid function;

[0071] According to the result that the lane-changing behavior trajectory is driving in the right lane or the middle lane, obtain the driving feature utility of the driving factor, where the expression of the driving feature utility is:

[0072]

[0073] Among them, U priority (lane) represents the driving feature utility, lane is prioritized means that the lane-changing behavior trajectory and the current lane are both driving in the right lane or the middle lane, otherwise means that the lane-changing behavior trajectory and the current lane are not both driving in the right lane or the middle lane.

[0074] Exemplarily, Figure 2 is a schematic structural diagram of a ramp scenario in an embodiment of the present disclosure. Refer to Figure 2 As shown, in the determination of each feature utility related quantity in the ramp scenario, the effective distance D is the straight-line distance from the current position of the driverless vehicle to the lane fork. The speed of the host vehicle ve, the speed of the obstacle (social vehicle) v, the distance d between the lane-changing behavior trajectory of the driverless vehicle and the social vehicle (obstacle) in front at the current position, and the collision time between the lane-changing behavior trajectory of the driverless vehicle and the social vehicle (obstacle) in front at the current position is the collision time (ttc), that is, ttc = d / (ve - v). s ub can be 6, s lb can be -6, v ub1 can be 70, v lb1 can be 10, v ub2 can be 0.1 * the speed limit of the environmental road network, v lb2 can be 0.8 * the speed limit of the environmental road network, vub3 can be 15, v lb3 can be 3, v ub can be 150, v lb can be 50, the static collision distance is the distance sd of the collision between the lane-changing behavior trajectory and the static obstacle ahead, and the collision time can be the time of the collision between the lane-changing behavior trajectory and the static obstacle ahead, or the time of the collision between the lane-changing behavior trajectory and the dynamic obstacle ahead.

[0075] S150. Determine the total lane-changing utility according to the utility coefficient and the characteristic utility.

[0076] In one embodiment, determining the total lane-changing utility according to the utility coefficient and the characteristic utility includes: determining the product of the sum of the product of the safety utility coefficient and the safety characteristic utility, the product of the trafficability utility coefficient and the trafficability characteristic utility, the product of the continue-route utility coefficient and the continue-route characteristic utility, and the product of the driving utility coefficient and the driving characteristic utility and the product of the rule utility coefficient and the rule characteristic utility as the total lane-changing utility.

[0077] It can be understood that, on the basis of the above embodiments, the product of the safety utility coefficient γ2 and the safety characteristic utility U safe the product of the trafficability utility coefficient γ3 and the trafficability characteristic utility U passable the product of the continue-route utility coefficient γ4 and the continue-route characteristic utility U Continue the product of the driving utility coefficient γ5 and the driving characteristic utility U priority the product of the rule utility coefficient γ1 and the rule characteristic utility U rule is determined as the total lane-changing utility U total , where the expression of the total lane-changing utility U total is:

[0078] U total = γ1U rule ×(γ2U safe + γ3U passable + γ4U continue + γ5U priority )

[0079] In the above formula, according to the driving strategy determined by the above environmental scenario, the total lane-changing utility is determined according to the utility coefficient and the characteristic utility corresponding to the lane-changing code in the driving strategy.

[0080] S160. Determine the maximum value of the total lane-changing utility according to the lane-changing behavior, and determine the current lane-changing decision of the driverless vehicle as the lane-changing behavior corresponding to the maximum value.

[0081] Lane-changing behavior can be divided into three types: maintaining the current lane, changing lanes to the right, and changing lanes to the left. The total utility calculation for each lane-changing behavior will be based on the characteristic attributes of the target lane of the lane-changing behavior. For example, when maintaining the current lane, the characteristic information of the current lane (target lane) will be extracted for utility calculation. When changing lanes to the right, the characteristic information within the lane on the right side of the current lane (target lane) will be extracted for utility calculation. When changing lanes to the left, the characteristic information within the lane on the left side of the current lane (target lane) will be extracted for utility calculation. In the foregoing description, we divide the vehicle lane-changing behavior into five preset influencing factors. Therefore, the characteristic information of the lane is also extracted according to these five preset influencing factors, and the corresponding characteristic utility is calculated.

[0082] In one example, the lane-changing behavior includes maintaining the current lane, changing lanes to the right, or changing lanes to the left. According to the lane-changing behavior, determining the maximum value of the total lane-changing utility, and determining the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle, includes:

[0083] Determining a first total lane-changing utility according to maintaining the current lane and the total lane-changing utility;

[0084] Determining a second total lane-changing utility according to changing lanes to the right and the total lane-changing utility;

[0085] Determining a third total lane-changing utility according to changing lanes to the left and the total lane-changing utility;

[0086] Determining the lane-changing behavior corresponding to the maximum value among the first total lane-changing utility, the second total lane-changing utility, and the third total lane-changing utility as the current lane-changing decision of the driverless vehicle.

[0087] Exemplarily, for example, the current lane is the center lane, there is a vehicle with a speed of 50 km / h and a distance of 100 m ahead, the speed of the vehicle itself is 70 km / h, the road network speed limit is 80 km / h, the road network is all dotted lines, and normally considering traffic rules factors, there are no obstacles on the adjacent right lane.

[0088] The lane-changing behavior is to maintain the current lane: the traffic rules lane-changing code is 10, the safety lane-changing code is 10, the trafficability lane-changing code is 10, the continue route lane-changing code is 10, the driving lane-changing code is 10; the code combination (driving strategy 1010101010), correspondingly, the rule utility coefficient is 1, the safety utility coefficient is 0.33, the trafficability utility coefficient, the continue route utility coefficient is 0.33, the driving utility coefficient is 0.75; the current lane rule characteristic utility is 1, the safety characteristic utility is 1, the trafficability characteristic utility is 0.84, the continue route characteristic utility is 1, the driving characteristic utility is 0.2. The total utility of maintaining the current lane (the first total lane-changing utility) is 1.09.

[0089] The lane-changing behavior is a right lane change: the traffic rule lane-changing code is 10, the safety lane-changing code is 10, the trafficability lane-changing code is 10, the continuing route lane-changing code is 10, and the driving lane-changing code is 10; the code combination (driving strategy 1010101010), correspondingly, the rule utility coefficient is 1, the safety utility coefficient is 0.33, the trafficability utility coefficient is 0.33, the continuing route utility coefficient is 0.33, and the driving utility coefficient is 0.75; the right lane rule feature utility is 1, the safety feature utility is 1, the trafficability feature utility is 1, the continuing route feature utility is 1, and the driving feature utility is 0. The total utility of the right lane change (the total utility of the second lane change) is 1.0.

[0090] The lane-changing behavior is a left lane change: the traffic rule lane-changing code is 10, the safety lane-changing code is 10, the trafficability lane-changing code is 10, the continuing route lane-changing code is 10, and the driving lane-changing code is 10; the code combination (driving strategy 1010101010), correspondingly, the rule utility coefficient is 1, the safety utility coefficient is 0.33, the trafficability utility coefficient is 0.33, the continuing route utility coefficient is 0.33, and the driving utility coefficient is 0.75; the rule feature utility is 1, the safety feature utility is 1, the trafficability feature utility is 1, the continuing route feature utility is 1, and the driving feature utility is 0. The total utility of the left lane change (the total utility of the third lane change) is 1.0.

[0091] After calculating the total lane-changing utility of each lane-changing behavior, the lane-changing behavior with the highest total lane-changing utility is taken as the optimal lane-changing decision for the current frame. It can be understood that in the above formula, the total utility of keeping the current lane is 1.09, the total utility of changing to the right lane is 1.0, and the total utility of changing to the left lane is 1.0. The lane-changing behavior with the highest total lane-changing utility is to keep the current lane, and keeping the current lane is taken as the optimal lane-changing decision for the current frame.

[0092] However, in engineering applications, to avoid numerical fluctuations caused by the environment and perception noise, the vehicle will not immediately execute the lane-changing behavior. Only when the utility of the optimal lane-changing behavior at the current moment is greater than a certain threshold of the utility of the lane-changing behavior selected at the previous moment and after a certain period of observation cycle, the vehicle begins to take this optimal lane-changing behavior as the current lane-changing decision and execute the lane-changing behavior.

[0093] The vehicle lane-changing decision-making method provided by this embodiment does not need to rely on a neural network model trained with a large number of samples for outputting lane-changing decisions, reducing labor costs and improving computing power. Without modifying the logic of the top-level decision-making algorithm, the decision-making algorithm is easy to expand and maintain. Nor does it need a lane-changing decision-making method based on fixed rules. Instead, preset influencing factors and lane-changing encodings of the preset influencing factors are set in advance. By combining the lane-changing encodings, a dynamic driving strategy is determined. For each driving strategy, the corresponding feature utility is matched to the lane-changing encoding. The driving strategy and utility coefficient are dynamically determined according to the environmental scenario, and then the lane-changing decision is determined based on the lane-changing behavior, utility coefficient, and feature utility, solving the problem of rule conflicts or algorithm coverage difficulties in complex scenarios, efficiently reflecting the expectations of different scenarios for lane-changing decision-making behaviors, improving the accuracy of lane-changing decisions, and being easy to expand and customize development.

[0094] Figure 3 It is a schematic structural diagram of a vehicle lane-changing decision-making device in an embodiment of the present disclosure. As Figure 3 shown: The device includes:

[0095] A first acquisition module, configured to acquire the environmental scenario where the driverless vehicle is located;

[0096] A first determination module, configured to determine a driving strategy according to the environmental scenario and the lane-changing encodings of each preset influencing factor;

[0097] A second determination module, configured to determine a utility coefficient according to the driving strategy;

[0098] A second acquisition module, configured to acquire the feature utility of the preset influencing factor;

[0099] A third determination module, configured to determine the total lane-changing utility according to the utility coefficient and the feature utility;

[0100] A fourth determination module, configured to determine the maximum value of the total lane-changing utility according to the lane-changing behavior, and determine the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle.

[0101] In an example, the preset influencing factors are safety factors, trafficability factors, rule factors, continuing route factors, or driving factors;

[0102] The lane-changing encodings of the preset influencing factors are safety lane-changing encodings for safety factors, trafficability lane-changing encodings for trafficability factors, rule lane-changing encodings for rule factors, continuing route lane-changing encodings for continuing route factors, or driving lane-changing encodings for driving factors;

[0103] The characteristic utility of the preset influencing factor is the safety characteristic utility of the safety factor, the trafficability characteristic utility of the trafficability factor, the rule characteristic utility of the rule factor, the continuation route characteristic utility of the continuation route factor, or the characteristic utility of the driving factor;

[0104] The utility coefficient is the safety utility coefficient of the safety factor, the trafficability utility coefficient of the trafficability factor, the rule utility coefficient of the rule factor, the continuation route utility coefficient of the continuation route factor, or the driving utility coefficient of the driving factor, where the sum of the safety utility coefficient, the trafficability utility coefficient, and the continuation route utility coefficient is 1.

[0105] In one example, the second determination module is further configured to combine the lane change encodings of each preset influencing factor according to the environmental scenario to determine a driving strategy.

[0106] In one example, the second determination module is further configured to combine the rule lane change encoding, the safety lane change encoding, the trafficability lane change encoding, the continuation route lane change encoding, and the driving lane change encoding according to the environmental scenario to determine a driving strategy.

[0107] In one example, the second acquisition module is further configured to obtain the safety characteristic utility of the safety factor according to the interaction relationship between the lane change behavior trajectory and the vehicle behind, where the expression of the safety characteristic utility is:

[0108]

[0109] Among them, U safe (a ot ) represents the safety characteristic utility, a ot represents the optimal forced acceleration, a ot = 100 represents the acceleration when forced overtaking is not possible, [a lb , a ub represents the acceleration interval of the sigmoid function, [s ub , s lb represents the interval mapped by the original normalized value in the sigmoid function;

[0110] According to the prediction of getting out of the road network or crossing the solid line by the lane change behavior trajectory, obtain the rule characteristic utility of the rule factor, where the expression of the rule characteristic utility is:

[0111]

[0112] Among them, U rule(traj) represents the utility of the rule feature. traj obey rule means that the lane - changing behavior trajectory does not exit the road network or does not cross the solid line, and otherwise means that the lane - changing behavior trajectory exits the road network or crosses the solid line;

[0113] According to the static collision distance, collision time, and obstacle speed between the lane - changing behavior trajectory and the obstacle ahead, obtain the passing - ability feature utility of the passing - ability factor. Among them, the expression of the passing - ability feature utility is:

[0114]

[0115] Among them, U passable (sd, v, ttc) represents the passing - ability feature utility, sd represents the static collision distance between the lane - changing behavior trajectory and the obstacle ahead, v represents the obstacle speed, and ttc represents the collision time between the lane - changing behavior trajectory and the obstacle ahead. [v lb1 , v ub1 , [v lb2 , v ub2 , [v lb3 , v ub3 respectively represent the static collision distance interval, obstacle speed interval, and collision time interval of the sigmoid function; [s ub , s lb represents the interval where the original normalized value is mapped by the sigmoid function;

[0116] According to the lane - changing behavior trajectory and the effective distance, obtain the continue - route feature utility of the continue - route factor. Among them, the expression of the continue - route feature utility is:

[0117]

[0118] Among them, U continu (D) represents the continue - route feature utility, D represents the effective distance, that is, the distance from the current position of the driverless vehicle to the lane fork. [v ub , v lb represents the normalized interval of the effective distance, [s ub , s lb represents the interval where the original normalized value is mapped by the sigmoid function;

[0119] According to the result that the lane - changing behavior trajectory is driving in the right - hand lane or the middle lane, obtain the driving feature utility of the driving factor. Among them, the expression of the driving feature utility is:

[0120]

[0121] Among them, U priority(lane) represents the driving feature utility. lane is prioritized means that the lane-changing behavior trajectory and the current lane are both driving in the rightmost lane or the middle lane. Otherwise means that the lane-changing behavior trajectory and the current lane are not both driving in the rightmost lane or the middle lane.

[0122] In one example, the third determination module is further configured to determine the total lane-changing utility as the product of the product of the safety utility coefficient and the safety feature utility, the product of the trafficability utility coefficient and the trafficability feature utility, the product of the continued route utility coefficient and the continued route feature utility, and the product of the driving utility coefficient and the driving feature utility, and the product of the rule utility coefficient and the rule feature utility.

[0123] In one example, the fourth determination module is further configured to determine the first total lane-changing utility according to maintaining the current lane and the total lane-changing utility;

[0124] Determine the second total lane-changing utility according to changing lanes to the right and the total lane-changing utility;

[0125] Determine the third total lane-changing utility according to changing lanes to the left and the total lane-changing utility;

[0126] Determine the lane-changing behavior corresponding to the maximum value among the first total lane-changing utility, the second total lane-changing utility, and the third total lane-changing utility as the current lane-changing decision of the driverless vehicle.

[0127] The vehicle lane-changing decision device provided by the embodiments of the present disclosure can execute the steps in the vehicle lane-changing decision method provided by the method embodiments of the present disclosure, and the implementation steps and beneficial effects are not described in detail here.

[0128] Figure 4 It is a schematic structural diagram of an electronic device in the embodiments of the present disclosure. Specifically refer to the following Figure 4 which shows a schematic structural diagram of the electronic device 500 suitable for implementing the embodiments of the present disclosure. Figure 4 The shown electronic device is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present disclosure.

[0129] Such as Figure 4As shown, the electronic device 500 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 501, which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage device 508 into the random access memory (RAM) 503 to implement the method of the embodiments as described in the present disclosure. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0130] In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the embodiments of the present disclosure include a computer program product, which includes a computer program carried on a non-transitory computer-readable storage medium, and the computer program contains program codes for executing the method shown in the flowchart, so as to implement the vehicle lane-changing decision-making method as described above. In such an embodiment, the computer program may be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiments of the present disclosure are executed.

[0131] It should be noted that the above-mentioned computer-readable storage medium in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wires, 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 above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable storage medium other than the computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0132] The above-mentioned computer-readable storage medium can be included in the above-mentioned electronic device; it can also exist separately without being assembled into the electronic device. The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device is caused to: obtain the environmental scene where the driverless vehicle is located; determine a driving strategy according to the environmental scene and the lane-changing codes of each preset influencing factor; determine a utility coefficient according to the driving strategy; obtain the characteristic utility of the preset influencing factor; determine the total lane-changing utility according to the utility coefficient and the characteristic utility; determine the maximum value of the total lane-changing utility according to the lane-changing behavior, and determine the current lane-changing decision of the driverless vehicle as the lane-changing behavior corresponding to the maximum value.

[0133] Optionally, when the above-mentioned one or more programs are executed by the electronic device, the electronic device can also perform the other steps described in the above embodiments.

[0134] In the context of the present disclosure, a computer-readable storage medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium may be a machine-readable signal medium or a machine-readable storage medium. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0135] The foregoing description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features having similar functions disclosed in the present disclosure.

Claims

1. A vehicle lane-changing decision-making method, characterized in that The method includes: Obtaining the environmental scene where the driverless vehicle is located; Determining a driving strategy according to the environmental scene and the lane-changing codes of each preset influencing factor; Determining a utility coefficient according to the driving strategy; Obtaining the characteristic utility of the preset influencing factor; Determining the total lane-changing utility according to the utility coefficient and the characteristic utility; According to the lane-changing behavior, determining the maximum value of the total lane-changing utility, and determining the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle; the preset influencing factors are safety factors, trafficability factors, rule factors, continuing route factors or driving factors; The lane-changing codes of the preset influencing factors are the safety lane-changing code of the safety factor, the trafficability lane-changing code of the trafficability factor, the rule lane-changing code of the rule factor, the continuing route lane-changing code of the continuing route factor or the driving lane-changing code of the driving factor; The characteristic utility of the preset influencing factor is the safety characteristic utility of the safety factor, the trafficability characteristic utility of the trafficability factor, the rule characteristic utility of the rule factor, the continuing route characteristic utility of the continuing route factor or the characteristic utility of the driving factor; The utility coefficient is the safety utility coefficient of the safety factor, the trafficability utility coefficient of the trafficability factor, the rule utility coefficient of the rule factor, the continuing route utility coefficient of the continuing route factor or the driving utility coefficient of the driving factor, where the sum of the safety utility coefficient, the trafficability utility coefficient and the continuing route utility coefficient is 1; The determining the total lane-changing utility according to the utility coefficient and the characteristic utility includes: Determining the total lane-changing utility as the product of the sum of the product of the safety utility coefficient and the safety characteristic utility, the product of the trafficability utility coefficient and the trafficability characteristic utility, the product of the continuing route utility coefficient and the continuing route characteristic utility, and the product of the driving utility coefficient and the driving characteristic utility and the product of the rule utility coefficient and the rule characteristic utility.

2. The method according to claim 1, wherein Determining the driving strategy according to the environmental scene and the lane-changing codes of each preset influencing factor includes: Combining the lane-changing codes of each preset influencing factor according to the environmental scene to determine the driving strategy.

3. The method according to claim 1, characterized in that, Combining the lane-changing codes of each preset influencing factor according to the environmental scene to determine the driving strategy includes: Combining the rule lane-changing code, the safety lane-changing code, the trafficability lane-changing code, the continuing route lane-changing code and the driving lane-changing code according to the environmental scene to determine the driving strategy.

4. The method according to claim 1, characterized in that, Obtaining the characteristic utility of the preset influencing factor includes: Obtaining the safety characteristic utility of the safety factor according to the interaction relationship between the lane-changing behavior trajectory and the vehicle behind, where the expression of the safety characteristic utility is: ; Among them, represents the safety feature utility, represents the optimal cut-in acceleration, = 100 represents the non-cut-in acceleration, the acceleration interval of the sigmoid function, the interval where the original normalized value is mapped by the sigmoid function; Obtaining the rule characteristic utility of the rule factor according to the prediction of the lane-changing behavior trajectory leaving the road network or crossing the solid line, where the expression of the rule characteristic utility is: ; Among them, represents the utility of the rule feature, represents that the lane-changing behavior trajectory does not lead out of the road network or does not cross the solid line, represents that the lane-changing behavior trajectory leads out of the road network or crosses the solid line; Obtain the trafficability characteristic utility of the trafficability factor according to the static collision distance, collision time, and obstacle speed between the lane-changing behavior trajectory and the obstacle ahead, where the expression of the trafficability characteristic utility is: ); Among them, represents the utility of the trafficability feature, sd represents the static collision distance between the lane-changing behavior trajectory and the obstacle ahead, v represents the obstacle speed, and ttc represents the collision time of the lane-changing behavior trajectory with the obstacle ahead, , , respectively represent the static collision distance interval, the obstacle speed interval, and the collision time interval of the sigmoid function; The interval where the original normalized value is mapped by the sigmoid function; Obtain the continued route characteristic utility of the continued route factor according to the lane-changing behavior trajectory and the effective distance, where the expression of the continued route characteristic utility is: ; Among them, represents the utility of the continue route feature, and D represents the effective distance, that is, the distance from the current position of the driverless vehicle to the lane fork. The normalized interval of the effective distance The interval where the original normalized value is mapped by the sigmoid function; Obtain the driving characteristic utility of the driving factor according to the result that the lane-changing behavior trajectory is driving in the right lane or the middle lane, where the expression of the driving characteristic utility is: ; Among them, represents the driving characteristic utility, represents that the lane change behavior trajectory and the current lane are both driving in the rightmost lane or the middle lane, represents that the lane change behavior trajectory and the current lane are not both driving in the rightmost lane or the middle lane.

5. The method according to claim 1, wherein The lane-changing behavior includes maintaining the current lane, changing lanes to the right, or changing lanes to the left. According to the lane-changing behavior, determine the maximum value of the total lane-changing utility, and determine the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle, including: Determine the first total lane-changing utility according to maintaining the current lane and the total lane-changing utility; Determine the second total lane-changing utility according to changing lanes to the right and the total lane-changing utility; Determine the third total lane-changing utility according to changing lanes to the left and the total lane-changing utility; Determine the lane-changing behavior corresponding to the maximum value among the first total lane-changing utility, the second total lane-changing utility, and the third total lane-changing utility as the current lane-changing decision of the driverless vehicle.

6. A vehicle lane change decision-making device, characterized in that The device includes: A first acquisition module for acquiring the environmental scene where the driverless vehicle is located; A first determination module for determining a driving strategy according to the environmental scene and the lane-changing codes of each preset influencing factor; A second determination module for determining a utility coefficient according to the driving strategy; A second acquisition module for acquiring the characteristic utility of the preset influencing factor; A third determination module for determining the total lane-changing utility according to the utility coefficient and the characteristic utility; A fourth determination module for determining the maximum value of the total lane-changing utility according to the lane-changing behavior, and determining the lane-changing behavior corresponding to the maximum value as the current lane-changing decision of the driverless vehicle; the preset influencing factors are safety factors, trafficability factors, rule factors, continued route factors, or driving factors; The lane-changing codes of the preset influencing factors are the safety lane-changing codes of the safety factors, the trafficability lane-changing codes of the trafficability factors, the rule lane-changing codes of the rule factors, the continued route lane-changing codes of the continued route factors, or the driving lane-changing codes of the driving factors; The characteristic utilities of the preset influencing factors are the safety characteristic utilities of the safety factors, the trafficability characteristic utilities of the trafficability factors, the rule characteristic utilities of the rule factors, the continued route characteristic utilities of the continued route factors, or the characteristic utilities of the driving factors; The utility coefficient is the safety utility coefficient of the safety factor, the trafficability utility coefficient of the trafficability factor, the rule utility coefficient of the rule factor, the continued route utility coefficient of the continued route factor, or the driving utility coefficient of the driving factor, where the sum of the safety utility coefficient, the trafficability utility coefficient, and the continued route utility coefficient is 1; The determination of the total lane-changing utility according to the utility coefficient and the characteristic utility includes: Determine the total lane-changing utility as the product of the sum of the product of the safety utility coefficient and the safety feature utility, the product of the trafficability utility coefficient and the trafficability feature utility, the product of the continued route utility coefficient and the continued route feature utility, and the product of the driving utility coefficient and the driving feature utility, and the product of the rule utility coefficient and the rule feature utility.

7. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, the method according to any one of claims 1-5 is implemented.

Images