Overtaking prompt control method, electronic device, readable storage medium and vehicle

Abstract

The present invention discloses an overtaking prompt control method, an electronic device, a readable storage medium and a vehicle, and relates to the field of advanced driver assistance technology. Specifically, a first algorithm is used to infer whether the current driving condition meets the driver's overtaking intention by combining multi-dimensional traffic factors, and then the overtaking prompt strategy is activated after the overtaking intention is judged to be met: first, several candidate overtaking schemes are obtained, and then the second algorithm is used to predict the success rate of each candidate overtaking path relative to the current driver's driving experience in combination with the current traffic variables, vehicle driving conditions, etc., and the optimal scheme is locked therefrom; finally, based on the driver's actual overtaking action, the aforementioned algorithm tool is updated to make it more in line with the driver's habits. The present invention provides personalized and intelligent overtaking decision prompts through human-machine collaboration, and provides optimized overtaking suggestions according to the behavior patterns and environmental conditions of different drivers, thereby improving the driver's operating safety and driving efficiency.

Description

Technical Field

[0001] The present invention relates to the field of advanced driver assistance technology, and in particular to an overtaking prompt control method, an electronic device, a readable storage medium and a vehicle. Background Art

[0002] In today's complex and changeable road conditions, drivers usually perform same-lane overtaking operations according to their own driving needs (same-lane overtaking means that the vehicle overtakes the vehicle in the same lane and in front of the vehicle, and usually returns to the original lane in front of the front vehicle after overtaking. This process generally involves two lane change operations, which is referred to as "overtaking" in the present invention. For lane change overtaking that does not return to the original lane after overtaking, the technical concept proposed in the present invention can be used as a reference). However, due to improper operation, misjudgment of the vehicle and the environment, and lack of driving experience, overtaking failures occur from time to time, and such mistakes often lead to road disputes or even traffic accidents. Therefore, how to provide assistance to the driver and accurately judge the feasibility of overtaking in combination with his personal driving habits and current road scenes has become one of the important challenges facing modern advanced driver assistance systems (ADAS).

[0003] Regarding the driver's overtaking safety, the existing overtaking behavior guidelines around the world, such as the Euro NCAP test protocol, require that the distance between vehicles, speed and lane occupancy should be considered when overtaking. In particular, in the field of assisted driving systems (ADAS), the norms of overtaking behavior are gradually taking shape, and multiple international standards and studies have conducted in-depth discussions on this. Euro NCAP has established evaluation standards for overtaking behavior for autonomous driving and assisted driving systems, requiring vehicles to effectively monitor the distance between vehicles in the surrounding environment and perform safe overtaking operations at an appropriate speed; in addition, NHTSA has also proposed multiple specifications related to overtaking and lane changing behaviors, such as Blind Spot Intervention and Adaptive Cruise Control, which directly affect the driver's overtaking behavior. However, existing assisted driving technologies still have deficiencies in judging overtaking behavior, cannot effectively interact with the driver, and cannot provide an accurate assessment of the probability of successful overtaking; at the same time, the assistance system cannot provide personalized overtaking suggestions based on the driver's personal driving habits.

[0004] Based on the above situation, even in vehicles currently equipped with assisted driving technology, most drivers still rely on rearview mirrors or vehicle charts for subjective judgment when overtaking. This judgment process has a large degree of uncertainty and is deeply affected by personal driving habits and technical level. Therefore, in order to further improve driving safety, it is urgent to develop an intelligent overtaking assistance system that can combine the driver's personalized behavior pattern and road environment to provide more accurate overtaking suggestions and judgments.

[0005] In addition, although there are studies in the industry that have proposed a variety of overtaking behavior evaluation methods and systems, and established an interactive framework and auxiliary system from the hardware level, providing a new development direction for the collaboration between the vehicle system and various modules. However, these systems generally still have shortcomings in some key aspects. For example: the strategy algorithm design is unclear: although the system mentions that each overtaking behavior is recorded through the strategy algorithm, it does not clearly explain how to design a specific overtaking strategy algorithm. The system lacks strategies to ensure overtaking safety and efficiency in different scenarios, and does not make in-depth use of the driver's historical overtaking behavior. This leads to the lack of dynamic adaptability of the algorithm, and it cannot be adjusted according to the personalized needs of different drivers, making it difficult to provide an accurate and personalized overtaking strategy framework; the data utilization method is vague: although the system mentions that the probability of overtaking success is calculated through the collected environmental data, it does not explain in detail how to use this data to derive an accurate calculation formula. This ambiguity has become the main bottleneck for the system to provide accurate overtaking suggestions, limiting the system's sensitivity to environmental changes and the accuracy of overtaking decisions. Summary of the invention

[0006] In view of the above, the present invention aims to provide an overtaking prompt control method, an electronic device, a readable storage medium and a vehicle to solve the above-mentioned technical problems.

[0007] The technical solution adopted by the present invention is as follows:

[0008] In a first aspect, the present invention provides an overtaking prompt control method, comprising:

[0009] During vehicle driving, the real-time multi-dimensional data is combined with the pre-trained driver overtaking enthusiasm model to decide whether to start the overtaking prompt strategy; the multi-dimensional data includes: the front vehicle following status monitoring, the front gap monitoring and the set low speed monitoring;

[0010] After determining to start the overtaking prompt strategy, at least a plurality of potential overtaking paths are planned using the perception module of the vehicle, and the relatively optimal target overtaking path and the corresponding success rate among the potential overtaking paths are obtained in combination with the pre-trained driver overtaking proficiency model;

[0011] Based on the driver's response to the target overtaking path and the execution result, the driver's overtaking enthusiasm model and / or the driver's overtaking proficiency model is dynamically updated, wherein the response and the execution result include one of the following: giving up overtaking, executing overtaking and successfully overtaking, and executing overtaking and failing to overtake.

[0012] In this embodiment, the first algorithm is used to infer whether the current driving condition meets the driver's consistent overtaking intention by combining multi-dimensional traffic variable factors, and after judging from the technical level that the current situation meets the overtaking intention, the overtaking prompt strategy is activated; specifically, the path planning technology is used to first obtain several candidate overtaking plans, and then the second algorithm is used to predict the success rate of each candidate overtaking path relative to the current driver's driving experience by combining the current traffic variables, vehicle driving conditions, etc., and the optimal plan is locked; finally, based on the driver's actual overtaking action, the algorithm tools used in the above calculation process are optimized to better fit the driver's personal driving habits. Therefore, through human-machine collaboration, personalized and intelligent overtaking decision prompts can be provided to the driver, and optimized overtaking suggestions can be provided according to the behavior patterns and road environment conditions of different drivers, thereby improving the driver's operating safety and driving efficiency.

[0013] In at least one possible implementation manner, the control method further includes making a decision based on a comparison result of the success rate with a preset success rate threshold:

[0014] whether to display the target overtaking path and its success rate to the driver or whether to cancel the displayed target overtaking path;

[0015] Alternatively, whether to output overtaking action reminder information to the driver while displaying the target overtaking path and its success rate.

[0016] In this embodiment, it is considered that in actual application, it may be necessary to provide the driver with the overtaking path and success rate. However, no matter whether audio or image display methods are used, it may have a certain impact on the driver's concentration. Based on this thinking, it is proposed that when the system background calculates the optimal path and its success rate or after the path and its success rate have been displayed, it can be compared with the established success rate threshold (it can be understood that the success rate threshold does not mean that overtaking is difficult to succeed, but is set for the perspective of the impact on the driver that this embodiment focuses on). If it is lower than the success rate threshold, the background calculation result may not be output or the path and its success rate that have been visualized may be hidden (preferably, other prompting means can be used at this time to inform the driver that overtaking is not recommended for the time being); if it is equal to or higher than the success rate threshold, the displayed path and its success rate can be combined with reminders to assist the driver in performing overtaking actions, such as when, in which direction, at what speed, etc. to overtake, that is, to further complete safe and reliable overtaking actions by human-machine cooperation.

[0017] In at least one possible implementation, the control method further includes: after displaying the target overtaking path and its success rate, dynamically adjusting the displayed success rate value in combination with the current driving conditions and traffic variables detected in real time; and when the success rate is lower than a preset success rate threshold, canceling the displayed target overtaking path and its success rate, and prompting the driver with information indicating that overtaking is not recommended at this time.

[0018] Generally speaking, after showing the driver the optimal route and its success rate, the display content can be kept so that the driver can make human judgments and actions according to the display content. However, in this embodiment, considering that the real traffic scene is ever-changing and changing rapidly, it is not required to keep the displayed content, but to dynamically adjust the success rate value. On the one hand, only adjusting the success rate without adjusting the optimal path is to avoid changes in the two contents, which will confuse the driver or even make the opposite action, which brings risks; on the other hand, only adjusting the numbers is relatively easier for the driver to accept from the perspective of human attention and interpretation. In addition, it can be pointed out that the aforementioned embodiment only mentions the situation where the success rate value is lowered. In the real situation, the success rate can also be increased, and combined with reminders and prompts, the driver's overtaking confidence and decision-making ability can be significantly improved.

[0019] In at least one possible implementation, the control method further includes:

[0020] When the overtaking warning strategy is turned on and before planning a potential overtaking path, the real-time vehicle speed, surrounding vehicle speed, distance between the vehicle and surrounding vehicles, and current road conditions are combined to determine whether they are below the preset safety threshold;

[0021] If it is lower than the safety threshold, the overtaking prompt is terminated and a new decision is made whether to start the overtaking prompt strategy, and the turn signal and / or steering wheel signal are monitored simultaneously;

[0022] If it is detected that the current turn signal is on and / or the steering wheel is turned to a preset angle, an overtaking warning signal is output.

[0023] In this embodiment, the state changes in the real scene are also taken into consideration. On the one hand, the subsequent overtaking prompt calculation processing can be terminated in time to save unnecessary resource consumption. On the other hand, safe driving operation can be ensured so that the driver is unaware of the overtaking calculation processing in the background. In particular, this embodiment mentions that while terminating the current overtaking prompt function, it is necessary to determine whether a "quasi-overtaking action" has actually been made from a human perspective. This is because when this scheme is executed to the link of starting the overtaking prompt strategy, it indicates that the first algorithm has determined that several conditions at this time meet the driver's previous overtaking intention. Even if the system terminates the subsequent overtaking prompt due to an emergency, it cannot be ensured whether the driver as the actor can also terminate his intention. There is a certain probability that the driver is still preparing to perform the overtaking action at this time. Therefore, through the detection objects such as the turn signal and the steering wheel angle, the risk situation caused by the subjective initiative of the person can be determined at this time. Once the detection object indicates that the driver intends to overtake, a higher-level warning signal can be immediately triggered to alert the driver and cancel the overtaking action; of course, in actual operation, this embodiment can also be combined with lane control technology to force torque increase or constrain the tire's swing angle to prevent the vehicle from being taken out of the lane by the driver's operation and thus causing danger.

[0024] In at least one possible implementation, the method of dynamically updating the model includes:

[0025] According to the driver's decision result of giving up overtaking or executing overtaking, the driver's overtaking enthusiasm model is updated;

[0026] According to the execution result of the driver's overtaking, the driver's overtaking proficiency model is updated.

[0027] In this embodiment, it can be understood that since the two algorithm tools perform their respective functions, the sources of materials for further updating and learning are different. The enthusiasm model mainly considers whether the driver has actually performed the overtaking action; the proficiency model mainly focuses on whether the overtaking behavior is successful after the overtaking action is performed. The definition of success or failure will be specifically introduced later and will not be elaborated here.

[0028] In at least one possible implementation, the control method further includes:

[0029] In the process of processing each potential overtaking path by the driver overtaking proficiency model, the road condition is predicted in combination with the real-time feedback of the vehicle perception module; and the posterior probability of the corresponding potential overtaking path is adjusted according to the road condition prediction result.

[0030] The focus of this embodiment is that before the optimal path is predicted, the background processing process is still a real-time dynamic process. This is because the vehicle is in a driving state, so the road conditions are not static. During the calculation and processing, the posterior probability of the candidate path can be adaptively adjusted based on the transient working conditions, thereby ensuring that the final optimal path is more in line with the actual driving conditions.

[0031] In at least one possible implementation, the control method further includes:

[0032] Determine whether the vehicle is traveling on a straight road;

[0033] After confirming that the vehicle is traveling on a straight road, monitor the driver's interaction content;

[0034] Based on the interactive content actively sent by the driver, a decision is made whether to activate the overtaking prompt strategy.

[0035] In this embodiment, in addition to using the aforementioned automatic calculation method to decide whether to turn on the overtaking warning strategy function, a more three-dimensional decision-making factor is further introduced, starting from multiple dimensions such as the current actual intention of the actor, to provide a comprehensive and more humane processing basis for the decision-making process of whether to turn on the function. The driver interaction mode involved in this link will be expanded and explained later, which will not be elaborated here.

[0036] In a second aspect, the present invention also provides an electronic device, comprising: one or more processors, a memory and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the electronic device, enable the electronic device to execute the above-mentioned overtaking prompt control method.

[0037] In a third aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above-mentioned overtaking prompt control method when executed by a processor.

[0038] In a fourth aspect, the present invention further provides a vehicle, wherein the vehicle is equipped with the above-mentioned electronic device or the above-mentioned computer-readable storage medium.

[0039] It should be understood that the second to fourth aspects of the embodiments of the present invention are consistent with the technical solutions of the first aspect of the embodiments of the present invention, and the beneficial effects achieved by each aspect and the corresponding possible implementation methods are similar and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:

[0041] Figure 1 A schematic diagram of an overtaking prompt control method provided by an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of planning a potential overtaking path provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0044] The present invention proposes an embodiment of an overtaking prompt control method, specifically, as follows Figure 1 shown, including:

[0045] Step S1: during vehicle driving, combining the real-time monitored multi-dimensional data with the pre-trained driver overtaking enthusiasm model to decide whether to start the overtaking prompt strategy;

[0046] To elaborate, during the driving process of the vehicle, the overtaking prompt strategy system of the present invention is closed by default to reduce unnecessary computing load and avoid interfering with normal driving. Therefore, a method for turning on the overtaking prompt strategy is provided here: during the driving process of the vehicle, the key vehicle driving state data can be monitored in real time through the vehicle hardware module (including the vehicle's perception system, such as laser radar, camera, ultrasonic sensor, etc.). The monitoring objects required in this embodiment may include but are not limited to: front vehicle following state monitoring (i.e., the distance from the front vehicle, the speed difference between the vehicle and the front vehicle, etc.), front gap monitoring (detecting the gap between the vehicle and the front vehicle, and also including detecting the gap distance between the vehicle and the object in front of the vehicle, which takes into account the scenario of returning to the current lane after overtaking) and vehicle speed monitoring (monitoring the vehicle in a preset low-speed driving state, of course, the speed detection of the vehicle and the surrounding vehicles mentioned above can also be included).

[0047] In addition, the driver overtaking enthusiasm model mentioned here should be pointed out that the name of the model comes from its purpose, and its essence is just an algorithm tool. Therefore, when the present invention is actually implemented, the algorithm architecture adopted may not be limited (the same applies to the proficiency model described below). A relatively good example is provided here, which can be pre-combined with the driver's historical overtaking decision data and the Bayesian algorithm to construct and pre-train the driver's overtaking enthusiasm model, and based on the prior distribution, the preset logical function and the traffic environment variables, the overtaking intention probability that represents the driver's overtaking enthusiasm is obtained.

[0048] Specifically, after the vehicle-computer perception system collects and transmits the above-mentioned multi-dimensional data of real-time monitoring, the central processing unit (CPU) of the vehicle-computer calls the driver's overtaking enthusiasm model that is pre-constructed and pre-trained by combining the driver's historical overtaking decisions and the Bayesian algorithm, and calculates the posterior of the driver's overtaking intention that characterizes the driver's overtaking enthusiasm based on the prior distribution and the established logical function (Logistic Function); then, based on the current overtaking intention, decides whether to turn on the overtaking prompt strategy function. For example, if the posterior probability output value obtained after model processing is greater than 0.3 (for illustration only), it can be considered that the current driver has a high willingness to overtake. At this time, the overtaking prompt strategy can be triggered to start, and prepare to provide the driver with subsequent overtaking suggestions and reminders.

[0049] As a supplementary explanation, it can also be pointed out that the model architecture involved in the present invention is preferably Bayesian Logistic Regression. Compared with the Bayesian network, the Bayesian Logistic Regression model architecture can better solve the problems of concern in the scenario of the present invention, and is more interpretable, more stable, and has lower computational complexity.

[0050] Continuing from the previous article, specifically, in the Bayesian model of overtaking enthusiasm, the logistic function of overtaking willingness can be defined as follows:

[0051]

[0052] in, : own speed, : Speed ​​of the left rear vehicle, :Right rear vehicle speed, : Speed ​​of the preceding vehicle, : Speed ​​of the left front vehicle, : Speed ​​of the right front vehicle, : Distance to the left rear vehicle, : Distance to the right rear vehicle, :Distance to the vehicle ahead, : Distance to the left front vehicle, : Distance to the right front vehicle. Here we first explain the parameter units: Indicates the influence of the distance and speed difference of the preceding vehicle, It represents the ratio of the distance between the vehicle ahead and the obstacle or road boundary ahead to the lane width, and is used to measure whether there is enough space to return to the lane safely after overtaking. is a low-speed monitoring item, which indicates the impact when the vehicle speed is lower than 20km / h (actually changeable), and is used to reflect the change in the willingness to overtake under such low-speed conditions. and are Bayesian parameters, and their prior distribution can be determined by the driver's historical overtaking data and the vehicle's performance parameters (such as size, acceleration response, dynamics, braking characteristics, etc.). Therefore, let the prior distribution of the above parameter unit be , then the probability of the driver's willingness to overtake can be calculated as:

[0053]

[0054] This integral represents the sum of the logistic functions 1 and the prior distribution of the parameters , integrate the parameter space, and finally get the posterior probability of the driver's willingness to overtake. In this way, the driver's willingness to overtake can be judged in a timely and accurate manner. That is, the system infers whether the current driver has the intention to overtake by combining the current driving conditions and the historical data of overtaking decisions, which can be understood as the function "switch" that automatically turns on the overtaking prompt strategy. In actual applications, the vehicle CPU can work together with the real-time operating system (RTOS), memory control unit (MCU), central gateway module, etc. to complete the above work to ensure fast calculation and data processing.

[0055] In addition to the above-mentioned automatic determination of whether the overtaking prompt is turned on or not, in other embodiments of the present invention, it is also considered that corresponding processing can be performed in combination with the driver's active behavior. For example, when the vehicle computer determines that the driver turns on the turn signal on a straight road, or when the driver actively requests the overtaking assistance function by interacting with the vehicle computer, then the overtaking prompt strategy can be considered to be activated. Regarding the interaction mode mentioned here, it can be expanded to include a specific interaction method: a combination of head movement and eye movement triggering in addition to conventional multi-modal interaction methods such as button triggering and voice triggering. This is because when the driver has the intention to overtake, he will subconsciously express it through methods such as looking up, turning his head, and exploring his line of sight. Therefore, in some preferred embodiments, the present invention proposes that when making a decision on whether to turn on the overtaking prompt strategy, the driver's current head and eye movement is combined with the existing head movement detection and line of sight detection technology as one of the influencing factors of the decision.

[0056] Continuing from the above, step S2, after determining to start the overtaking prompt strategy, at least use the vehicle perception module to plan a number of potential overtaking paths, and combine the pre-trained driver overtaking proficiency model to provide the driver with a relatively optimal target overtaking path and a predicted success rate;

[0057] As a supplement, the following six points should be pointed out:

[0058] First, in the technical route of the present invention, it is necessary to first determine the intention to overtake. If and only if it is determined that there is an intention to overtake, the overtaking prompt strategy is activated, and then the overtaking path is planned. In other words, even if the overtaking conditions are met, if the overtaking intention does not meet the established requirements, the overtaking path planning process will not be executed;

[0059] Second, the potential overtaking path is not only the overtaking path of the left and right adjacent lanes mentioned in the present invention, but also the overtaking path of the non-adjacent lanes is also within the scope of the potential overtaking path;

[0060] Third, the processing process of the present invention is mostly completed in the background. In order to avoid affecting the driver's driving operation, only the result after processing can be displayed to the driver in this link. That is, in the driving scene, the driver only receives the current relatively optimal overtaking path and its success probability value output by the system in the form of vision and / or hearing;

[0061] Fourth, based on the third point, based on the judgment of the success rate value, overtaking reminders can be given synchronously through visual and / or auditory means. For example, if the system gives an overtaking probability of 98%, which is greater than the established threshold of 90%, an audio output can be used to remind the driver of the timing and suggestion of overtaking: "You can overtake the vehicle in front at present. Please overtake from the right adjacent lane at a speed greater than 50km / h within eight seconds and return to this lane. The success rate is 98%"; if it is lower than the threshold, only the target path and probability can be output without giving suggestions and reminders.

[0062] Fifth, as mentioned above, the overtaking intention judgment process can be combined with driving conditions to incorporate whether the overtaking conditions are met as a factor in the intention decision. However, in order to avoid safety risks, in some preferred embodiments of the present invention, the background can compare the success rate of the current optimal overtaking path with the established threshold again to determine whether to display the relatively optimal path and its probability to the driver. For example, the background calculates that the success rate of the optimal overtaking path is 80%, which is lower than the preset threshold of 90%. At this time, the system may not display any overtaking path to the driver to avoid the risk of misleading or causing the driver to make risky driving actions based on the display.

[0063] Sixth, based on this conception, in other preferred embodiments of the present invention, considering that after the target overtaking path and its success rate are displayed to the driver, special circumstances may arise due to the ever-changing traffic scene. Therefore, after the predicted success rate is output, the success rate can be dynamically adjusted by continuously combining the current driving condition perception detection and real-time calculation. When the success rate is lower than a certain predetermined threshold, the display of the optimal target path is canceled and a prompt can be given that overtaking is not recommended at present.

[0064] Continuing from the previous article, in actual operation, the system can call on the visual perception modules of the vehicle computer and the core graphics card computing power (such as GPU or dedicated graphics processing unit, etc.) to plan the overtaking path, and for different directions, it can generate multiple potential overtaking paths, such as the following Figure 2 As shown, those skilled in the art can understand that the potential overtaking path planning itself is not the focus of the present invention, but what can be specifically described is:

[0065] The driver overtaking proficiency model mentioned here can be constructed and pre-trained in advance by combining the driver's historical overtaking result data and the Bayesian algorithm, and based on the prior distribution, the preset logical function and the traffic environment variables, the overtaking success probability that represents the driver's overtaking proficiency is obtained.

[0066] Specifically, the system can simultaneously detect the speed, position, and even driving intention of surrounding vehicles through sensing devices such as LiDAR, and in conjunction with technologies such as the Internet of Vehicles and roadside units, and transmit these data to the central processing unit (CPU) of the vehicle computer in real time. The pre-trained driver overtaking proficiency model can then be called. Similar to the overall construction method of the overtaking enthusiasm model mentioned above, the main difference between the driver overtaking proficiency model mentioned here is that the Bayesian model used is trained accordingly based on the driver's historical performance in overtaking success or failure under certain overtaking conditions.

[0067] Then, for each potential overtaking path, the vehicle will calculate the driver's overtaking success probability on each path in real time based on the prior distribution g and logic function L2 of the Bayesian model of overtaking proficiency, combined with the different traffic environment variables and driving conditions input at this time. Among all potential overtaking paths, the system can select the path with the highest overtaking success rate and transmit it to the driver through the display module (such as the central control screen HMI or AR-HUD) and / or audio module to help him make the safest overtaking decision.

[0068] Specifically, the overtaking logic function L2 of each potential overtaking path can be defined according to the specific working conditions and safety of the overtaking path. For example, for overtaking from the right lane (the left lane is similar and will not be repeated here), the logic function can be defined as:

[0069]

[0070] in, are all Bayesian parameters, and their prior distributions are learned from data such as the driver's historical overtaking success and vehicle characteristics; Indicates the distance to the vehicle ahead The speed difference between the front vehicle and the vehicle itself The smaller the value, the better the overtaking condition. It reflects the safe gap in the lane after overtaking. The larger the value, the larger the available space. Reflects the speed of the vehicle in front of you The impact of the front vehicle on the right lane increases the difficulty of overtaking when the speed of the front vehicle is high; Reflects the speed of the vehicle behind the right The impact of the right rear vehicle approaching or driving at high speed increases the risk of overtaking; Reflects the distance difference between the right front vehicle and the right rear vehicle. The larger the difference, the more space there is in the right lane to overtake, and the safety is higher.

[0071] Similarly, let the prior distribution of these parameters be , then the posterior probability calculation formula for the overtaking success rate from the right lane is:

[0072]

[0073] By combining the above overtaking logistic function with the prior distribution The posterior probability of the success rate of overtaking from the right lane is obtained by integration (i.e., P2 in the figure, and P1 in the figure on the left, which will not be described in detail in the present invention). Finally, the success probabilities of different overtaking plans are compared, and the plan with the highest overtaking success rate is selected, and the optimal overtaking plan and its corresponding success probability are pushed to the driver. This process ensures that the driver can make overtaking decisions based on the optimal plan, improving driving safety and operating efficiency.

[0074] In addition, it can be added that the radar detector can be used to predict the road conditions of the road ahead, the left front overtaking road and the right front overtaking road. If the road conditions of the section through which the potential overtaking path passes are poor: for example, there are roadblocks, potholes or irregular bumps, etc., then the posterior probability of the potential overtaking path is reduced. It can be added here that the reduced probability mentioned here is different from the "dynamic adjustment of the success rate" mentioned above. Here it means that when calculating the probability of each potential overtaking path, the overtaking success rate of each relevant path can be additionally reduced by a certain probability value; afterwards, as the reminder strategy mentioned above, the driver can also be promptly informed through the display module or voice prompt: "The current overtaking section has poor road conditions, please choose overtaking behavior with caution to ensure safe decision-making."

[0075] Step S3, dynamically updating the driver's overtaking enthusiasm model and / or the driver's overtaking proficiency model based on the driver's response to the target overtaking path and the execution result, wherein the response and the execution result include one of the following: giving up overtaking, executing overtaking and overtaking successfully, and executing overtaking and overtaking failing.

[0076] Specifically, after giving the driver the optimal overtaking path and its probability of overtaking success, the driver may make the following three choices: 1. The driver decides not to overtake; 2. The driver fails to overtake; 3. The driver succeeds in overtaking. It can be emphasized here that for the algorithm tool optimization link here, one of the key factors involved is how to define whether the driver has successfully performed the overtaking action. In some preferred embodiments, it can be defined as: the successful overtaking is characterized by the entire body of the vehicle or a body greater than a predetermined proportion (such as 90%) returning to the original lane after the overtaking is performed, otherwise it is defined as an overtaking failure. This is a further design made in combination with the actual driving situation. If most of the body of the vehicle does not return to the original lane after overtaking, it may be because of unexpected reasons such as the unexpected acceleration of the "front vehicle" before overtaking, resulting in the gap in front of the "front vehicle" not allowing the vehicle to return to the original lane. This is a relatively unsafe overtaking result, and is therefore defined as an overtaking failure. This situation is used as a learning sample to provide iterative material for the aforementioned proficiency model and the pre-deployed overtaking path planning algorithm.

[0077] Continuing from the previous article, after the driver makes a decision and completes the overtaking behavior, no matter which situation it is, it can be regarded as a new sample data. Based on the latest sample, the system will update the driver's overtaking enthusiasm model and overtaking proficiency model respectively. That is to say, it can be understood that after each overtaking opportunity, the results of whether the overtaking is executed or not, whether it is successful or not, and the corresponding driving scene-related data are used as samples to update the prior model. Through this continuous model update, it can better adapt to the driver's personalized driving behavior and continuously optimize the accuracy of overtaking strategies and suggestions.

[0078] To elaborate: (1) If the driver does not perform any overtaking operation (such as not turning the steering wheel or turning on the turn signal) within ten seconds of the prompt to overtake, it is determined that the driver has decided to give up overtaking. In this case, the initial state will be returned and the driver's overtaking enthusiasm model will be updated. According to the Bayesian formula, its posterior distribution satisfies:

[0079]

[0080] in, is a logical function, is the prior distribution representing the driver’s overtaking enthusiasm (willingness) in the current situation.

[0081] (2) If the driver chooses to overtake, the posterior distribution of the overtaking aggressiveness model satisfies:

[0082]

[0083] At this time, the central processing unit (CPU) of the vehicle computer can, but is not limited to, call the Markov chain Monte Carlo (MCMC) algorithm to sample and calculate the new posterior distribution, and store it in a storage module such as the non-volatile memory (NVM) or flash memory of the vehicle computer as the prior distribution of the overtaking aggressiveness model called next time.

[0084] (3) The overtaking proficiency model can be updated based on the success or failure of the overtaking result. If the overtaking is successful, the posterior distribution of the overtaking proficiency model satisfies:

[0085]

[0086] If overtaking fails, the posterior distribution satisfies:

[0087]

[0088] in, is the logistic function of overtaking proficiency, is the prior distribution. By sampling, calculating and storing the new posterior distribution, the system can gradually update the overtaking proficiency model. In this way, the system can dynamically update the driver's overtaking enthusiasm model to more closely reflect the driver's personalized behavior and continuously optimize the overtaking strategy. With the continuous feedback of the driver's overtaking decision and response results, the system will gradually grasp and adapt to the current driver's driving habits, and dynamically adjust the overtaking proficiency model to better match the driver's personal style and behavior pattern, thereby improving the personalization and accuracy of the system.

[0089] Finally, it can be added that in order to ensure driving safety, when the overtaking prompt strategy is turned on and before planning a potential overtaking path, the overtaking safety threshold can also be determined: if the value of a key parameter is lower than the preset safety threshold (threshold), the overtaking probability will be directly judged as 0, and the driver can be informed in time through the display module and / or audio module that overtaking is not recommended at present, thereby ensuring safe decision-making.

[0090] It should be pointed out here that the previous article mentioned auxiliary suggestions and reminders based on the success rate threshold, decisions on whether to display the optimal path, and cancellation of the displayed optimal path. These preferred solutions all occur after the potential path planning is executed and the model is used to process the optimal result. It can be understood that the above-mentioned success rate threshold solution mentioned in the previous article has been determined to meet the requirements of safe overtaking at the system level, but it is only to minimize the impact on the driver; the safety threshold mentioned here occurs before the potential overtaking path is planned, that is, after the overtaking intention is met and the overtaking prompt strategy function is turned on, the system level inspection believes that the objective conditions at this time cannot meet the requirements of safe overtaking.

[0091] The safety threshold mentioned here can be determined based on the vehicle's own speed and the speed and distance of surrounding vehicles. In actual operation, the following parameters (the same as those involved in the previous example) will be used for comprehensive evaluation: : own speed, : Speed ​​of the left rear vehicle, :Right rear vehicle speed, : Speed ​​of the preceding vehicle, : Speed ​​of the left front vehicle, : Speed ​​of the right front vehicle, : Distance to the left rear vehicle, : Distance to the right rear vehicle, :Distance to the vehicle ahead, : Distance to the left front vehicle, : Distance to the right front vehicle.

[0092] Based on the above parameter examples, the following security functions can be defined for reference: , determine whether it is safe to overtake:

[0093]

[0094] in, is the weight coefficient, It is a scaling parameter used to measure the impact of speed difference on safety distance.

[0095] when When the vehicle is detected, it is considered safe to overtake and the subsequent steps such as overtaking planning and prompting are continued.

[0096] when , the overtaking success rate can be directly set to 0 (without subsequent processing and calculation, it is forcibly assigned to 0), and the driver can be informed that it is unsafe to overtake at this time. For example, if it is lower than the safety threshold, the overtaking prompt strategy is terminated or stopped and a decision is made again whether to start the overtaking prompt strategy, and the turn signal and / or steering wheel signal can be monitored synchronously at this time; if it is detected that the current turn signal is lit and / or the steering wheel is turned to a preset angle, indicating that an unconventional dangerous working condition has occurred, that is, although it meets the positive evaluation, due to a sudden change in traffic conditions, the driver still wants to overtake when the objective conditions are not met, then a higher level warning signal representing the danger of overtaking can be output at this time.

[0097] The above safety threshold concept takes into account the speed and distance of the vehicles in front, behind, left and right, to ensure that when overtaking, it can avoid conflicts with surrounding vehicles and ensure the safety of the driver. and All of these can be determined by the vehicle's own performance parameters, and can be attributed to the factory settings. These parameters can be fixed and set according to factors such as the vehicle's dynamics, size, and braking characteristics during vehicle production, and will not change with the adaptive learning process of the two Bayesian models mentioned above. This ensures that even under different drivers and driving habits, fixed safety standards are always followed to ensure the basic safety of overtaking behavior.

[0098] In summary, the main design concept of the present invention is to first combine multi-dimensional traffic variable factors, use the first algorithm to infer whether the current driving condition meets the driver's consistent overtaking intention, and after judging at the technical level that the current scenario meets the overtaking intention, start the overtaking prompt strategy; specifically, use the path planning technology to first obtain several candidate overtaking plans, and then combine the current traffic variables, vehicle driving conditions, etc., use the second algorithm to predict the success rate of each candidate overtaking path relative to the current driver's driving experience, and lock the optimal plan; finally, based on the driver's actual overtaking action, the algorithm tool used in the above calculation process is optimized to better fit the driver's personal driving habits. The present invention provides drivers with personalized and intelligent overtaking decision prompts, and can provide optimized overtaking suggestions based on different drivers' behavior patterns and road environment conditions, thereby improving driving safety and driving efficiency.

[0099] Based on the above control method scheme, the present invention also provides an embodiment of an electronic device, which may include, for example: one or more processors, a memory and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, and when the instructions are executed by the electronic device, the electronic device executes the above overtaking prompt control method.

[0100] Based on the above control method scheme, the present invention also provides an embodiment of a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above overtaking prompt control method is implemented.

[0101] Based on the same inventive concept and the above hardware carrier solution, the present invention also provides an embodiment of a vehicle, the main feature of which is that it is equipped with the above electronic device or has the above computer-readable storage medium. The selection of the vehicle can be determined based on the correct understanding and reasonable implementation of the above solution by those skilled in the art.

[0102] If the expressions expressing orientation are mentioned in the embodiments of the present invention, they are relative concepts based on the embodiments. In addition, "at least one" means one or more, and "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b, c can be single or multiple.

[0103] The above describes in detail the structure, features and effects of the present invention based on the embodiments shown in the drawings, but the above is only a preferred embodiment of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred methods can be reasonably combined and matched into a variety of equivalent schemes by those skilled in the art without departing from or changing the design ideas and technical effects of the present invention; therefore, the present invention is not limited to the scope of implementation shown in the drawings, and all changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which still do not exceed the spirit covered by the specification and drawings, should be within the protection scope of the present invention.

Claims

1. A method for controlling an overtaking prompt, characterized in that: include: During vehicle driving, the real-time monitoring multi-dimensional data is combined with the pre-trained driver overtaking enthusiasm model to decide whether to activate the overtaking prompt strategy; The multidimensional data includes: monitoring of the following state of the preceding vehicle, monitoring of the gap ahead, and monitoring of the set low vehicle speed; wherein, the method for constructing the driver's overtaking enthusiasm model includes: pre-combining the driver's historical overtaking decision data and the Bayesian algorithm, constructing and pre-training the driver's overtaking enthusiasm model, and obtaining the overtaking intention probability representing the driver's overtaking enthusiasm based on the prior distribution, the preset logical function, and the traffic environment variables; After determining to start the overtaking prompt strategy, at least a plurality of potential overtaking paths are planned using the perception module of the vehicle, and the relatively optimal target overtaking path and the corresponding success rate among the potential overtaking paths are obtained in combination with the pre-trained driver overtaking proficiency model; wherein the driver overtaking proficiency model is constructed in a manner including: combining the driver's historical overtaking result data and the Bayesian algorithm in advance, constructing and pre-training the driver overtaking proficiency model, and obtaining the overtaking success probability representing the driver's overtaking proficiency based on the prior distribution, the preset logical function and the traffic environment variables; Based on the driver's response to the target overtaking path and the execution result, the driver's overtaking enthusiasm model and / or the driver's overtaking proficiency model is dynamically updated, wherein the response and the execution result include one of the following: giving up overtaking, executing overtaking and successfully overtaking, and executing overtaking and failing to overtake.

2. The overtaking prompt control method according to claim 1, characterized in that: The control method further includes making a decision based on a comparison result between the success rate and a preset success rate threshold: whether to display the target overtaking path and its success rate to the driver or whether to cancel the displayed target overtaking path; Alternatively, whether to output overtaking action reminder information to the driver while displaying the target overtaking path and its success rate.

3. The overtaking prompt control method according to claim 1, characterized in that: The control method also includes: after displaying the target overtaking path and its success rate, dynamically adjusting the displayed success rate value in combination with the current driving conditions and traffic variables detected in real time; and when the success rate is lower than a preset success rate threshold, canceling the displayed target overtaking path and its success rate, and prompting the driver with information indicating that overtaking is not recommended at the moment.

4. The overtaking prompt control method according to claim 1, characterized in that: The control method further comprises: When the overtaking warning strategy is turned on and before planning a potential overtaking path, the real-time vehicle speed, surrounding vehicle speed, distance between the vehicle and surrounding vehicles, and current road conditions are combined to determine whether they are below the preset safety threshold; If it is lower than the safety threshold, the overtaking prompt is terminated and a new decision is made whether to start the overtaking prompt strategy, and the turn signal and / or steering wheel signal are monitored simultaneously; If it is detected that the current turn signal is on and / or the steering wheel is turned to a preset angle, an overtaking warning signal is output.

5. The overtaking prompt control method according to claim 1, characterized in that: Ways to dynamically update models include: According to the driver's decision result of giving up overtaking or executing overtaking, the driver's overtaking enthusiasm model is updated; According to the execution result of the driver's overtaking, the driver's overtaking proficiency model is updated.

6. The overtaking prompt control method according to claim 1, characterized in that: The control method further comprises: In the process of processing each potential overtaking path by the driver overtaking proficiency model, the road condition is predicted in combination with the real-time feedback of the vehicle perception module; and the posterior probability of the corresponding potential overtaking path is adjusted according to the road condition prediction result.

7. The overtaking prompt control method according to any one of claims 1 to 6, characterized in that: The control method further comprises: Determine whether the vehicle is traveling on a straight road; After confirming that the vehicle is traveling on a straight road, monitor the driver's interaction content; Based on the interactive content actively sent by the driver, a decision is made whether to activate the overtaking prompt strategy.

8. An electronic device, characterized in that: include: One or more processors, a memory and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, and when the instructions are executed by the electronic device, the electronic device executes the overtaking prompt control method described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the overtaking prompt control method according to any one of claims 1 to 7 is implemented.

10. A vehicle, characterized in that: The vehicle is provided with the electronic device of claim 8 or the computer-readable storage medium of claim 9.

Images