WO2023284333A1

WO2023284333A1 - Method and apparatus for determining confidence of automatic driving strategy

Info

Publication number: WO2023284333A1
Application number: PCT/CN2022/083508
Authority: WO
Inventors: 陈昆盛; 贾思博; 夏炎; 高继扬; 朱望江; 单乐; 韩永根; 张驰; 冉旭
Original assignee: 魔门塔(苏州)科技有限公司
Priority date: 2021-07-14
Filing date: 2022-03-28
Publication date: 2023-01-19
Also published as: CN115700204A

Abstract

Provided are a method and apparatus for determining the confidence of an automatic driving strategy. The method comprises: acquiring external environment information and vehicle state information when a vehicle drives in an automatic driving mode according to a target automatic driving strategy (S101); determining first confidence according to the external environment information (S102), the first confidence being used for representing the degree of influence of the external environment information on the automatic driving strategy; determining second confidence according to the vehicle state information (S103), the second confidence being used for representing the degree of influence of the vehicle state information on the automatic driving strategy; determining third confidence according to the external environment information and the vehicle state information (S104), the third confidence being used for representing to the degree of influence of the driving state of the vehicle on the automatic driving strategy; determining fourth confidence according to the first confidence, the second confidence, and the third confidence (S105), the fourth confidence being used for representing the credibility of the automatic driving strategy; and determining the credibility level of the automatic driving strategy according to the fourth confidence and the fifth confidence of the driving mode of the vehicle (S106). The safety of the automatic driving strategy can be quantitatively evaluated more accurately and effectively, and both the safety and flexibility of automatic driving can be achieved.

Description

Method and device for determining confidence degree of automatic driving strategy

Cross References to Related Applications

This application claims the priority of the Chinese patent application with the application number 202110796378.7 and the application title "Method and device for determining confidence level of automatic driving strategy" submitted to the China Patent Office on July 14, 2021, the entire contents of which are incorporated herein by reference. Applying.

technical field

The present application relates to the field of automatic driving, and in particular to a method and device for determining the confidence level of an automatic driving strategy.

Background technique

With the continuous development of automotive technology and electronic technology, more and more self-driving vehicles with automatic driving functions have gradually entered the market. Among them, this type of vehicle not only allows the driver to operate the vehicle, but also can actively control the vehicle's driving by the built-in automatic driving module without the driver's operation, thus enriching the driving mode of the vehicle and making the vehicle more intelligent . Among them, in the automatic driving mode, the vehicle can obtain the road condition information of the road ahead of the vehicle in the driving direction through sensors such as radars and cameras installed on the vehicle, and combine the map data of the navigation, etc., to finally determine the automatic driving system used to control the automatic driving of the vehicle. Strategy.

In the existing technology, even though the automatic driving vehicle can calculate the automatic driving strategy faster and more accurately for subsequent driving through the machine learning model, it still cannot make a decision on the driving strategy in some emergency situations, and the driver still needs to take over the vehicle Driving work to ensure the safety of vehicles and personnel. Therefore, how to measure the safety of the automatic driving strategy, so that when the safety is low, prompt the driver in time or switch to the manual driving mode in time, so as to improve the flexibility and safety of automatic driving, is an urgent problem in this field. technical problem.

Contents of the invention

The present application provides a method and device for determining the confidence level of an automatic driving strategy, so as to measure the safety of the automatic driving strategy and improve the flexibility and safety of the automatic driving.

The first aspect of the present application provides a method for determining the confidence level of an automatic driving strategy, including: acquiring external environment information and vehicle state information when the vehicle is driving according to the target automatic driving strategy in automatic driving mode; determining the first Confidence degree; wherein, the first degree of confidence is used to characterize the degree of influence of external environmental information on the automatic driving strategy; the second degree of confidence is determined according to the vehicle state information; wherein, the second degree of confidence is used to characterize the influence of the vehicle state information on the automatic driving strategy degree of influence; determine the third degree of confidence according to the external environment information and vehicle state information; wherein, the third degree of confidence is used to characterize the degree of influence of the driving state of the vehicle on the automatic driving strategy; according to the first degree of confidence, the second degree of confidence and the degree of Three confidence levels, determine the fourth confidence level; wherein, the fourth confidence level is used to characterize the credibility of the automatic driving strategy; according to the fourth confidence level and the fifth confidence level of the vehicle driving mode, determine the credibility of the automatic driving strategy degree level.

In an embodiment of the first aspect of the present application, the confidence level of the automatic driving strategy includes: the first level, used to indicate that the vehicle cannot continue driving with the automatic driving strategy; the second level, used to indicate that the vehicle can continue driving with the automatic driving strategy Continue to drive and instruct the driver of the vehicle to assist in the vehicle's automatic driving process; the third level is used to indicate that the vehicle can continue to drive with an automatic driving strategy.

In an embodiment of the first aspect of the present application, determining the first confidence level according to the external environment information includes: determining the first confidence level according to obstacle targets existing in the external environment.

In an embodiment of the first aspect of the present application, determining the second confidence level according to the vehicle state information includes: determining the second confidence level according to the current position, attitude, and motion parameters of the vehicle.

In an embodiment of the first aspect of the present application, the third confidence level is determined according to external environment information and vehicle state information, including: according to the number of sensors, coverage area, collection quality, map and location used by the target automatic driving strategy, and driving For other vehicles passing ahead of the path, a third confidence level is determined.

In an embodiment of the first aspect of the present application, according to the first confidence degree, the second confidence degree and the third confidence degree, determining the fourth confidence degree includes: combining the first confidence degree, the second confidence degree and the third confidence degree A weighted decision is made to obtain a fourth degree of confidence.

In an embodiment of the first aspect of the present application, according to the fourth confidence level and the fifth confidence level of the vehicle driving mode, determining the confidence level of the automatic driving strategy includes: combining the fourth confidence level and the fifth confidence level Add up to get the confidence level of the automatic driving strategy.

In an embodiment of the first aspect of the present application, after determining the confidence level of the automatic driving strategy, it also includes: prompting the driver of the vehicle the confidence level of the automatic driving strategy through a multi-modal reminder; wherein, multiple Modalities include: one or more of visual, auditory, tactile, and olfactory.

In an embodiment of the first aspect of the present application, it also includes: sending a prompt to the driver when driver fatigue is detected, when the vehicle breaks down, when the automatic driving strategy exits abnormally, and/or when the automatic driving is completed.

The second aspect of the present application provides a device for determining the confidence level of an automatic driving strategy, which can be used to implement the method for determining the confidence level of an automatic driving strategy as provided in the first aspect of the present application. In the automatic driving mode, the external environment information and vehicle state information when driving according to the target automatic driving strategy; the first confidence degree determination module determines the first confidence degree according to the external environment information; and sends the external environment information to the third confidence degree determination module. Module, sending the first confidence degree to the fourth confidence degree determination module; wherein, the first confidence degree is used to characterize the degree of influence of external environment information on the automatic driving strategy; the second confidence degree determination module determines the second degree of confidence according to the vehicle state information Confidence degree; and the vehicle state information is sent to the third confidence degree determination module, and the second confidence degree is sent to the fourth confidence degree determination module; wherein, the second confidence degree is used to characterize the degree of influence of the vehicle state information on the automatic driving strategy ; The third confidence degree determination module determines the third confidence degree according to the external environment information and the vehicle state information, and sends the third confidence degree to the fourth confidence degree determination module; wherein, the third confidence degree is used to characterize the driving state of the vehicle The degree of influence on the automatic driving strategy; the fourth confidence degree determination module determines the fourth confidence degree according to the first confidence degree, the second confidence degree and the third confidence degree; wherein, the fourth confidence degree is used to characterize the automatic driving strategy Credibility: The fifth confidence degree determining module determines the degree of confidence of the automatic driving strategy according to the fourth degree of confidence and the fifth degree of confidence in the vehicle driving mode.

In an embodiment of the second aspect of the present application, the confidence level of the automatic driving strategy includes: the first level, used to indicate that the vehicle cannot continue driving with the automatic driving strategy; the second level, used to indicate that the vehicle can continue driving with the automatic driving strategy Continue to drive and instruct the driver of the vehicle to assist in the vehicle's automatic driving process; the third level is used to indicate that the vehicle can continue to drive with an automatic driving strategy.

In an embodiment of the second aspect of the present application, the first confidence level determining module is specifically configured to determine the first confidence level according to obstacle targets existing in the external environment.

In an embodiment of the second aspect of the present application, the second confidence level determination module is specifically configured to determine the second confidence level according to the current position, attitude, and motion parameters of the vehicle.

In an embodiment of the second aspect of the present application, the third confidence determination module is specifically configured to use the map and location of the target automatic driving strategy according to the number of sensors, coverage area, acquisition quality, and other vehicles passing ahead of the driving path. , to determine the third confidence level.

In an embodiment of the second aspect of the present application, the fourth confidence level determination module is specifically configured to perform weighted decision-making on the first confidence level, the second confidence level, and the third confidence level to obtain the fourth confidence level.

In an embodiment of the second aspect of the present application, the fifth confidence degree determination module is specifically configured to add the fourth confidence degree and the fifth confidence degree to obtain the confidence level of the automatic driving strategy.

In an embodiment of the second aspect of the present application, the device further includes: an interaction module; wherein, the interaction module is specifically used to remind the driver of the vehicle of the confidence level of the automatic driving strategy through a multi-modal reminder; wherein, the multi-modal Modalities include: one or more of visual, auditory, tactile, and olfactory.

In an embodiment of the second aspect of the present application, the interaction module is also used to send a prompt to the driver when driver fatigue is detected, when the vehicle breaks down, when the automatic driving strategy exits abnormally, and/or when the automatic driving has been completed .

To sum up, the method and device for determining the confidence level of an automatic driving strategy provided by this application can determine multiple confidence levels based on the acquired external environment information and vehicle state information, and perform summation processing on the multiple confidence levels, and combine The confidence degree of the driving mode finally obtains the confidence degree that can be used to characterize the confidence level of the target automatic driving strategy, so as to quantify the safety of the used target automatic driving strategy through the confidence degree. The obtained confidence is triggered from multiple dimensions and summed and quantified, which can more accurately and effectively quantify the safety of the automatic driving strategy and improve the safety of the automatic driving system. In the future, when the safety represented by the confidence level is low, the driver can be prompted in time or switched to the manual driving mode in time to ensure the safety of the vehicle and personnel. focus on improving user experience while satisfying the safety and flexibility of autonomous driving.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present application. Those skilled in the art can also obtain other drawings based on these drawings without any creative effort.

Figure 1 is a schematic diagram of the application scenario of the present application;

FIG. 2 is a schematic flowchart of an embodiment of a method for determining the confidence level of an automatic driving strategy provided by the present application;

FIG. 3 is a schematic structural diagram of a confidence determination device for an automatic driving strategy provided by the present application;

Fig. 4 is a schematic structural diagram of an embodiment of an interaction module provided by the present application.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and not necessarily Used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the application described herein, for example, can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

Before formally introducing the embodiments of the present application, the scenarios where the present application is applied and the problems existing in the prior art will be described with reference to the accompanying drawings. Wherein, FIG. 1 is a schematic diagram of the application scenario of the present application. The vehicle in the scenario shown in FIG. 1 has an automatic driving function, and the automatic driving function may be called adaptive cruise control (Adaptive Cruise Control, ACC for short). ) mode, automatic driving mode, etc. In the automatic driving mode, the automatic driving system installed in the vehicle can detect the road conditions of the road ahead in the driving direction through sensors such as radar and camera. At the same time, the automatic driving system can obtain or store high-precision Map, so as to determine the automatic driving strategy in time according to the information provided by the high-precision map of the road condition information of the road ahead, and then adjust the driving parameters of the vehicle according to the automatic driving strategy, and realize the automatic driving of the vehicle without driver intervention.

For example, in the example shown in FIG. 1 , when the vehicle is driving forward, the radar sensor installed on the vehicle continuously sends out radar signals in the direction marked ① in the figure. When a pedestrian appears in front of the vehicle, the radar will receive the radar signal reflected by the pedestrian in the direction marked ② in the figure. Subsequently, in step ③, the vehicle’s automatic driving system calculates the distance between the vehicle and the pedestrian based on the received radar signal reflected by the pedestrian, and there may be a lane for lane change on the right side of the current lane determined by the high-precision map. And when the sensor determines that there is no vehicle in the right lane, it can control the vehicle to change lanes to the right lane, and prevent the vehicle from colliding with pedestrians while maintaining the normal driving of the vehicle.

In some embodiments, in the automatic driving system of the vehicle, an automatic driving strategy is generated through a machine learning model, information acquired by sensors and map data are input into the machine learning model, and the automatic driving strategy is output. However, even if the self-driving vehicle can calculate the self-driving strategy faster and more accurately for subsequent driving through the machine learning model, it still cannot make decisions on the driving strategy in some emergency situations, and the driver still needs to take over the driving work of the vehicle To ensure the safety of vehicles and personnel.

Therefore, this application provides a method and device for determining the confidence level of an automatic driving strategy, which can be applied to the vehicle shown in Figure 1. When the vehicle is in the automatic driving mode, and the automatic driving system determines the automatic When the driving strategy is driving, the confidence level can be determined to quantify the safety of the automatic driving strategy. Subsequently, when the automatic driving system can determine the safety level of the automatic driving strategy, it can promptly prompt the driver or switch to the manual driving mode in time to ensure the safety of vehicles and personnel when the safety is low; when the safety is high, Reduce the driver's attention to the automatic driving process to a greater extent, improve user experience, and meet the safety and flexibility of automatic driving.

The technical solution of the present application will be described in detail below with specific embodiments. The following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

FIG. 2 is a schematic flowchart of an embodiment of a method for determining the confidence level of an automatic driving strategy provided by the present application. The method shown in FIG. 2 can be applied to the scene shown in FIG. 1 and executed by the automatic driving system in the vehicle. Specifically, the method includes:

S101: Obtain external environment information and vehicle state information when the vehicle is driving in an automatic driving mode according to a target automatic driving strategy.

Specifically, in this embodiment, the confidence degree of the target automatic driving strategy can be determined to indicate the safety of the automatic driving strategy through the confidence degree, then the automatic driving system can obtain the real-time information of the vehicle when using the target automatic driving strategy to control the automatic driving of the vehicle. external environment information and vehicle state information, so as to evaluate the current automatic driving strategy; Automated driving strategies used for evaluation.

In some embodiments, the sensors installed on the vehicle can be used to collect the external environment information of the environment where the vehicle is located. For example, the vehicle is equipped with various types of sensors such as cameras, infrared sensors, and radars, and all the sensors can be used together to obtain external environment information. information. The obtained external environment information includes lanes, curbs, other vehicles and pedestrians, buildings, obstacles and so on.

In some embodiments, sensors installed inside the vehicle can be used to collect vehicle status information, such as speedometers, thermometers, spirit levels and the like. The obtained vehicle state information includes: the vehicle's driving direction, speed, horizontal state, tire pressure, engine temperature, etc.

Subsequently, after the automatic driving system obtains the external environment information and vehicle state information, through S102-S104, respectively determine the first confidence level according to the external environment information, determine the second confidence level according to the vehicle state information, and determine the second confidence level according to the external environment information and vehicle state information. The information determines a third degree of confidence. The steps of S102-S104 will be described separately below, and the execution sequence of S102-S104 is not limited, or, S102-S104 can also be executed simultaneously.

S102: Determine a first confidence level according to the external environment information; wherein, the first confidence level is used to represent the degree of influence of the external environment information on the target automatic driving strategy.

Specifically, the first confidence level can be used to characterize the degree of influence of external environment information on automatic driving when the vehicle is driving using the target automatic driving strategy. For example, when the vehicle is driving forward under the control of the target automatic driving strategy, obstacles in front of the vehicle's driving direction can be considered to have a greater impact on the normal driving of the vehicle, and pedestrians behind the vehicle's driving direction can be considered to have a normal impact on the vehicle. The impact of driving is small, etc. Alternatively, the size, position, motion trajectory, and motion parameters of obstacles can all have different degrees of influence on the target automatic driving strategy, and the influence in different situations can be divided in advance and stored in the system in the form of a mapping relationship. After the external environment information is acquired, the corresponding first degree of confidence can be determined according to the external environment information.

In some embodiments, the first confidence level can be divided into at least the following three confidence levels. The first level is used to indicate that the external environment information will greatly affect the automatic driving of the vehicle, for example, it is directly ahead of the target driving strategy. A sudden obstacle is detected, and the obtained first degree of confidence corresponds to the first level; the second level: used to indicate that there are factors affecting the automatic driving of the vehicle in the external environment information, such as detection directly in front of the target driving strategy to a slow-moving object, the obtained first degree of confidence corresponds to the second level; the third level is used to indicate that the external environment information will not affect the automatic driving of the vehicle, and the obtained first degree of confidence corresponds to the third level .

In some embodiments, FIG. 3 is a schematic structural diagram of a device for determining the confidence level of an automatic driving strategy provided by the present application. The device shown in FIG. 3 can be used to execute the method shown in FIG. 2 , then in the device , the first confidence degree determination module (also called cognitive Precetion module, etc.), can be used to obtain sensor data, and after perception, obtain the first confidence degree by predicting the impact of obstacles on the current automatic driving strategy. The obtained first confidence degree is sent to the subsequent fourth confidence degree determination module for subsequent calculation, and at the same time, the first confidence degree determination module also sends the external environment information collected by the sensor to the third confidence degree determination module for subsequent calculation.

S103: Determine a second confidence level according to the vehicle state information; wherein, the second confidence level is used to characterize the degree of influence of the vehicle state information on the target automatic driving strategy.

Specifically, the second confidence level can be used to characterize the degree of influence of vehicle state information on automatic driving when the vehicle is driving using the target automatic driving strategy. For example, when the speed of the vehicle is too fast, the degree of influence on the automatic driving strategy is greater, and when the speed of the vehicle is slow, the influence on the automatic driving strategy is less. Alternatively, attitude information such as the position of the vehicle, remaining fuel, direction of motion, and level, as well as motion parameters such as speed and turning angle, etc., can all have different degrees of influence on the target automatic driving strategy, and divide different situations in advance. The influence of is stored in the system in the form of a mapping relationship. After the vehicle state information is obtained, the corresponding second confidence level can be determined according to the vehicle state information.

In some embodiments, the second degree of confidence can also be divided into the first level: used to indicate that the vehicle state information will greatly affect the automatic driving of the vehicle; factor and the third level: used to indicate that the vehicle status information will not affect the automatic driving of the vehicle.

In some embodiments, the second confidence determination module (also referred to as the Localization&Map module, etc.) in the device shown in FIG. degree, and send the second confidence degree to the fourth confidence degree determination module for subsequent calculation, and at the same time, the second confidence degree determination module also sends the collected vehicle state information to the third confidence degree determination module for subsequent calculation.

S104: Determine a third confidence degree according to the external environment information and the vehicle state information; wherein, the third confidence degree is used to characterize the influence degree of the driving state of the vehicle on the target automatic driving strategy.

Specifically, the third confidence level is used to characterize the degree of influence of the driving state of the vehicle on the target automatic driving strategy, which is jointly determined based on external environmental information and vehicle state information, and is used to measure the safety of the area that will pass ahead . For example, when the number of sensors in the front area where the vehicle is driving according to the target automatic driving strategy is large, the acquired coverage in the front area is wider, the blind area is smaller, and the quality of the acquisition is better, the accuracy of the target automatic driving strategy The degree of influence is low, and vice versa, the degree of influence is higher. When the lane information in the area where the vehicle passes can be provided by high-precision maps or sensed by sensors, it means that there are many sources of information, and when the information from multiple sources is consistent, the safety of autonomous driving can be guaranteed. The degree of influence on the target automatic driving strategy is low. For another example, when a vehicle passes through an area and sensors such as locators determine that other vehicles pass by, the degree of influence is low; the higher the accuracy of the map used for the designated target automatic driving strategy, the lower the degree of influence.

In some embodiments, the third degree of confidence can also be divided into the first level: used to indicate that the driving state of the vehicle will greatly affect the automatic driving of the vehicle; the second level: used to indicate that the driving state of the vehicle has an influence Factors and Level 3 for autonomous driving of the vehicle: used to indicate that the driving state of the vehicle will not affect the automatic driving of the vehicle.

In some embodiments, the third confidence degree determination module (also referred to as the world model Worldmodel module, etc.) in the device shown in FIG. 3 can be used to determine the third confidence degree, wherein the third confidence degree determination module specifically The driving direction obtained after the fusion of the external environment information provided by the first confidence determination module and the high-precision map provided by the second execution determination module The map information ahead and the vehicle state information provided by the second confidence degree determination module are combined to obtain the third confidence degree after information fusion, and the calculated third confidence degree is sent to the fourth confidence degree determination module for subsequent calculation. In some embodiments, the third confidence determination module can store different map information, vehicle state information and confidence in the system in the form of a mapping relationship. After obtaining the map information and vehicle state information, the third confidence determination module That is, the corresponding third confidence degree can be determined according to the mapping relationship.

S105: Determine a fourth confidence level according to the first confidence level, the second confidence level, and the third confidence level; wherein, the fourth confidence level is used to characterize the degree of credibility of the target automatic driving strategy.

Specifically, after the first confidence degree, the second confidence degree and the third confidence degree are calculated through S102-S104, in the embodiment of the present application, weighted decision processing is also performed on these three confidence degrees, so as to jointly obtain the The fourth confidence level of the three confidence levels. In a specific implementation, the fourth confidence degree can be determined through multi-decision arbitration, strategy arbitration, etc., or it can also assign different weights to each confidence degree, and determine the fourth degree of confidence through weighted summation. For the level corresponding to the confidence degree, for example, a weight of 0.25 may be assigned to the first confidence degree, a weight value of 0.25 may be assigned to the second confidence degree, and a weight value of 0.5 may be assigned to the third confidence degree and summed. It can be understood that, since the fourth confidence level comprehensively considers the first confidence level, the second confidence level and the third confidence level, the fourth confidence level can be determined more comprehensively from multiple aspects such as vehicle state, external environment, and driving state. Confidence, so that the fourth confidence itself has higher accuracy and reliability.

In some embodiments, the fourth confidence degree determination module (also called planning control (Planning and Control, referred to as: PNC) module, etc.) as shown in FIG. After the first confidence level, the second confidence level sent by the second confidence level determination module, and the third confidence level sent by the third confidence level determination module, the fourth confidence level is jointly obtained and output to the subsequent fifth confidence level determination module.

S106: According to the fourth degree of confidence and the fifth degree of confidence of the vehicle driving mode, determine the degree of confidence of the target automatic driving strategy.

Specifically, the fifth degree of confidence can be the decision made by the automatic driving system on the overall control, it can be the information added to the system, and it can also be divided into three levels, for example, the vehicle is about to leave the design operating area (Operational Design Domain, referred to as: ODD), the degree of influence on the current automatic driving strategy is relatively large, then the fifth level of confidence can correspond to the first level: it is used to indicate that the driving mode of the vehicle will greatly affect the automatic driving of the vehicle; When there is a preset distance to drive out of the ODD, the fifth level of confidence can correspond to the second level: it is used to indicate that the driving mode of the vehicle will affect the automatic driving of the vehicle; when the vehicle will not drive out of the ODD within the preset distance, the fifth confidence level Five levels of confidence correspond to the third level: used to indicate that the driving state of the vehicle will not affect the current automatic driving of the vehicle.

Then in S106, the fourth confidence degree and the fifth confidence degree determined in S105 can be weighted and summed to determine the final sixth confidence degree. For example, a weight can be assigned to the fifth confidence degree 0.5, assign a weight of 0.5 to the fourth weight and add and so on. The sixth degree of confidence can be used to indicate the level of confidence of the target automatic driving strategy, and can also be divided into three levels. Among them, the first level of the sixth degree of confidence is used to indicate that when the vehicle uses the target driving strategy for automatic driving, The safety is low, and the automatic driving mode needs to be exited and the driver is manually driven; the second level is used to indicate that when the vehicle uses the target driving strategy for automatic driving, there are certain risks, and the automatic driving mode can be maintained but driver assistance is required Participate in the process of automatic driving, or the driver can judge at any time and switch the vehicle from automatic driving mode to manual driving mode; the third level is used to instruct the vehicle to use the target driving strategy for automatic driving. Drivers can rely entirely on autonomous driving, and the driver can reduce their attention to the vehicle. In some embodiments, the confidence level used to represent the target automatic driving strategy may also be called confidence awareness (Confidence Aware, CA for short).

To sum up, through S101-S106 of the embodiment of the present application, the automatic driving system can determine multiple confidence levels based on the acquired external environment information and vehicle state information, and sum up the multiple confidence levels, and combine the driving mode The confidence level of the target automatic driving strategy is finally obtained, which can be used to characterize the confidence level of the target automatic driving strategy, so as to quantify the safety of the used target automatic driving strategy through the confidence level. Since the obtained confidence is triggered from multiple dimensions and summed and quantified, it can more accurately and effectively quantify the safety of the automatic driving strategy and improve the safety of the automatic driving system. In the future, when the safety represented by the confidence level is low, the driver can be prompted in time or switched to the manual driving mode in time to ensure the safety of the vehicle and personnel. focus on improving user experience while satisfying the safety and flexibility of autonomous driving.

In some embodiments, after the automatic driving system determines the different confidence levels of the target automatic driving strategy through the above method, the driver is prompted with the current confidence level by means of multi-modal reminders through the interaction module, Make the driver make corresponding operations according to the current reliability level. Wherein, the multimodality includes: one or more of vision, hearing, touch and smell.

Exemplarily, FIG. 4 is a schematic structural diagram of an embodiment of an interaction module provided by the present application, wherein the interaction module can provide an icon composed of three horizontal lines as shown in FIG. 4 on the human-computer interaction interface set inside the vehicle, And when it is determined that the reliability level of the automatic driving strategy is the third level, all three horizontal lines are lighted up. At this time, the driver can determine that the current reliability level is the third level according to the icon, and there is no need to drive If the driver provides too much attention to driving, it can improve the driver's driving experience. And when it is determined that the reliability level of the automatic driving strategy is the second level, the next two horizontal lines in the three horizontal lines are lighted, and the driver can determine the current reliability level as the second level according to the icon. At this time, although the automatic driving mode has not been exited, the driver still needs to pay attention to the driving situation in real time, and assist the driving when necessary, or take over the driving. When it is determined that the confidence level of the automatic driving strategy is the first level, light up the bottom one of the three horizontal lines, and the driver can determine the current reliability level as the first level according to this icon , At this time, the driver needs to drive manually. The automatic driving system can directly exit the automatic driving mode and prompt the driver to perform manual driving, or the automatic driving system can also give a prompt and then exit the automatic driving mode according to the driver's operation.

In some embodiments, the interaction module provided by the present application can also use other ways to prompt the driver, and these different ways can be implemented individually or jointly. For example, when the reliability level of the automatic driving strategy is determined to be the third level, the voice prompts "Don't be afraid, I'm watching" and other gentle sentences, reminding the driver that he does not need to be too nervous about the current automatic driving, which can ease the driving. member's nervousness. When the reliability level of the automatic driving strategy is determined to be the second level, the driver is reminded to take over the car at any time through voice prompts such as "Please get ready". When the reliability level of the automatic driving strategy is determined to be the first level, the driver is prompted to directly take over the vehicle for manual driving through voice prompts such as "please drive".

In some embodiments, when the interaction module prompts the driver, it can also use different prompting methods when the vehicle is in different states, for example, at night or when it detects that the vehicle is in a tunnel with low The mode of lighting up the icon as shown in Figure 4 is prompted to the driver; and in the daytime or when the vehicle is detected to be in the sun and the light is strong, the driver can be prompted by sound, vibration or other methods at this time. The prompts corresponding to different states can be stored in advance through the mapping relationship, and the corresponding prompts can be determined when prompts are needed, which can improve the driver’s experience, further improve the prompt efficiency of the interaction module, and improve the driver’s ability to receive prompts. probability, reducing the difficulty of receiving hints.

In some embodiments, the interaction module provided by this application can also take pictures of the driver through a camera and other equipment in manual driving mode, and play the Prompt information to allow the driver to concentrate on driving, or directly switch to the automatic driving mode instead of the driver to ensure the safety of vehicles and personnel to a greater extent.

In some embodiments, the interactive module provided by this application can also use the temperature, tire pressure and other sensors set in the vehicle to determine the failure of the vehicle's components, and then play a prompt message to remind the driver of the current vehicle failure; or, directly switch to automatic driving mode for hedging. For example, when a vehicle tire blowout is detected, it switches to the automatic driving mode and slows down and stops to the side of the road in time, thereby improving the safety of the vehicle and the automatic driving system.

In some embodiments, when the vehicle is in the automatic driving mode, if a certain function is not completed, the interaction module needs to prompt the driver in real time to inform the current state. For example, when the vehicle is changing lanes and suddenly detects that the speed of the vehicle coming from behind is increasing, the vehicle returns to the original lane. At this time, the interactive module prompts the driver by playing "the vehicle coming from behind, try it later", which can stabilize driving. The driver's emotions can prevent the safety hazards caused by the driver when troubleshooting problems while driving.

In some embodiments, when the vehicle is in the automatic driving mode and the automatic driving system exits abnormally, the reason can be explained to the driver actively, for example, by voice playing "There is no map, I will exit the automatic driving, it is safer to drive by yourself ", to prompt the driver to perform manual driving, making the prompts more humane and easier for the driver to accept.

In some embodiments, after the automatic driving system exits normally, the driver can be prompted to evaluate and give feedback on the automatic driving by voice playing "I have made progress today, let's see my performance score", so that the manufacturer receives After receiving these feedbacks, the relevant functions and algorithms of automatic driving can be adjusted in time, which is conducive to the subsequent continuous use and improves the driver's experience.

In the above-mentioned embodiments, the method for determining the confidence level of the automatic driving strategy provided by the embodiment of the present application is introduced. The degree determining device may include a hardware structure and/or a software module, and realize the above-mentioned functions in the form of a hardware structure, a software module, or a hardware structure plus a software module. For example, each module in Fig. 3, and whether a certain function among the above-mentioned functions is implemented in the form of a hardware structure, a software module, or a hardware structure plus a software module depends on the specific application and design constraints of the technical solution.

It should be noted that it should be understood that in the device for determining the confidence level of the automatic driving strategy described above in this embodiment, the division of each component and module is only a division of logical functions, and can be fully or partially integrated into a physical Physically, it can also be physically separated. For example, the first confidence degree determination module, the second confidence degree determination module, the third confidence degree determination module, the fourth confidence degree determination module, and the fifth confidence degree determination module can be independent entities, or any number of integrated On one entity, these modules can be implemented in the form of calling software through processing elements; they can also be implemented in the form of hardware; some modules can also be implemented in the form of calling software through processing elements, and some modules can be implemented in the form of hardware. It may be a separately established processing element, or it may be integrated into a certain chip of the above-mentioned device for implementation. In addition, it may also be stored in the memory of the above-mentioned device in the form of program code, which is called and executed by a certain processing element of the above-mentioned device. Determine the functionality of the module. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together, and can also be implemented independently. The main control component described here may be an integrated circuit, which has signal processing capabilities. In the implementation process, each step of the above-mentioned method or each of the above-mentioned modules can be completed by an integrated logic circuit of hardware in the main control component or an instruction in the form of software.

For example, the above components/modules may be one or more integrated circuits configured to implement the above method, for example: one or more specific integrated circuits (application specific integrated circuit, ASIC), or, one or more microprocessors (digital signal processor, DSP), or, one or more field programmable gate arrays (field programmable gate array, FPGA), etc. For another example, when one of the above modules is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (central processing unit, CPU) or other processors that can call program codes. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

In the above embodiments, all or part of the method steps performed by the device for determining the confidence level of the automatic driving strategy may be implemented by software, hardware, firmware or any combination thereof. When implemented in software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application will be generated in whole or in part. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website, computer, server or data center Transmission to another website site, computer, server, or data center by wired (eg, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center integrated with one or more available media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, or a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)), etc.

The present application also provides an electronic device, including: a processor and a memory; wherein, a computer program is stored in the memory, and when the processor executes the computer program, the processor can be used to execute any automatic driving strategy as in the foregoing embodiments of the present application Confidence determination method.

The present application also provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed, it can be used to perform the method for determining the confidence level of any automatic driving strategy in the foregoing embodiments of the present application.

The embodiment of the present application also provides a chip for running instructions, and the chip is used to execute the method for determining the confidence level of any automatic driving strategy mentioned above in the present application.

The embodiment of the present application also provides a program product, the program product includes a computer program, the computer program is stored in a storage medium, at least one processor can read the computer program from the storage medium, and the at least one When the processor executes the computer program, it can realize the method for determining the confidence level of any automatic driving strategy mentioned above in this application.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by program instructions and related hardware. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit it; although the application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: It is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the various embodiments of the present application. scope.

Claims

A method for determining the confidence level of an automatic driving strategy, comprising:

Obtain the external environment information and vehicle status information when the vehicle is driving in the automatic driving mode according to the target automatic driving strategy;

Determine a first confidence level according to the external environment information; wherein, the first confidence level is used to characterize the degree of influence of the external environment information on the automatic driving strategy;

Determining a second confidence degree according to the vehicle state information; wherein, the second confidence degree is used to characterize the degree of influence of the vehicle state information on the automatic driving strategy;

Determine a third confidence level according to the external environment information and the vehicle state information; wherein, the third confidence level is used to characterize the degree of influence of the driving state of the vehicle on the automatic driving strategy;

Determine a fourth confidence level according to the first confidence level, the second confidence level, and the third confidence level; wherein, the fourth confidence level is used to characterize the degree of credibility of the automatic driving strategy;

According to the fourth confidence level and the fifth confidence level of the vehicle driving mode, the confidence level of the automatic driving strategy is determined.
The method according to claim 1, wherein the confidence level of the automatic driving strategy comprises:

The first level is used to indicate that the vehicle cannot continue to drive with the automatic driving strategy;

The second level is used to indicate that the vehicle can continue to drive with the automatic driving strategy, and instruct the driver of the vehicle to assist in participating in the automatic driving process of the vehicle;

The third level is used to indicate that the vehicle can continue driving with the automatic driving strategy.
The method according to claim 1 or 2, wherein the determining the first confidence level according to the external environment information comprises:

The first confidence level is determined according to the obstacle target existing in the external environment.
The method according to claim 1 or 2, wherein said determining the second confidence level according to the vehicle state information comprises:

The second degree of confidence is determined according to the current position, attitude, and motion parameters of the vehicle.
The method according to claim 1 or 2, wherein said determining the third degree of confidence according to said external environment information and said vehicle state information comprises:

The third confidence level is determined according to the number, coverage, and collection quality of the sensors, the map and location used by the target automatic driving strategy, and other vehicles passing ahead of the driving path.
The method according to claim 1 or 2, wherein said determining a fourth confidence degree according to said first confidence degree, said second confidence degree and said third confidence degree comprises:

Making weighted decisions on the first confidence level, the second confidence level and the third confidence level to obtain the fourth confidence level.
The method according to claim 1 or 2, wherein the determining the confidence level of the automatic driving strategy according to the fourth confidence level and the fifth confidence level of the vehicle driving mode includes :

Adding the fourth confidence level and the fifth confidence level to obtain the confidence level of the automatic driving strategy.
The method according to any one of claims 1-7, wherein after the determination of the confidence level of the automatic driving strategy, further comprising:

The driver of the vehicle is reminded of the confidence level of the automatic driving strategy by means of a multi-modal reminder; wherein the multi-modality includes: one or more of vision, hearing, touch and smell.
The method according to claim 8, further comprising:

When the driver's fatigue is detected, when the vehicle breaks down, when the automatic driving strategy exits abnormally and/or when the automatic driving has been completed, a prompt is sent to the driver.
A confidence determination device for an automatic driving strategy, characterized in that it includes:

The information acquisition module is used to acquire the external environment information and vehicle state information when the vehicle is driving in the automatic driving mode according to the target automatic driving strategy;

The first confidence degree determination module determines a first confidence degree according to the external environment information; and sends the external environment information to a third confidence degree determination module, and sends the first confidence degree to a fourth confidence degree determination module ; Wherein, the first confidence degree is used to characterize the degree of influence of the external environment information on the automatic driving strategy;

The second confidence degree determination module determines a second confidence degree according to the vehicle state information; and sends the vehicle state information to a third confidence degree determination module, and sends the second confidence degree to a fourth confidence degree determination module ; Wherein, the second confidence degree is used to characterize the degree of influence of the vehicle state information on the automatic driving strategy;

A third confidence degree determination module, determining a third confidence degree according to the external environment information and the vehicle state information, and sending the third confidence degree to the fourth confidence degree determination module; wherein, the third The degree of confidence is used to characterize the degree of influence of the driving state of the vehicle on the automatic driving strategy;

A fourth confidence level determination module, determining a fourth confidence level according to the first confidence level, the second confidence level and the third confidence level; wherein, the fourth confidence level is used to characterize the automatic driving the credibility of the strategy;

The fifth confidence degree determining module is configured to determine the confidence level of the automatic driving strategy according to the fourth confidence degree and the fifth confidence degree of the vehicle driving mode.
The confidence degree determining device according to claim 10, wherein the confidence level of the automatic driving strategy comprises:

The first level is used to indicate that the vehicle cannot continue to drive with the automatic driving strategy;

The second level is used to indicate that the vehicle can continue to drive with the automatic driving strategy, and instruct the driver of the vehicle to assist in participating in the automatic driving process of the vehicle;

The third level is used to indicate that the vehicle can continue driving with the automatic driving strategy.
The confidence degree determination device according to claim 10 or 11, wherein the first confidence degree determination module determines the first confidence degree according to the external environment information, comprising:

The first confidence level determination module is specifically configured to determine a first confidence level according to obstacle targets existing in the external environment.
The confidence degree determination device according to claim 10 or 11, wherein the second confidence degree determination module determines the second confidence degree according to the vehicle state information, comprising:

The second degree of confidence is determined according to the current position, attitude, and motion parameters of the vehicle.
The confidence degree determination device according to claim 10 or 11, wherein the third confidence degree determination module determines the third confidence degree according to the external environment information and the vehicle state information, comprising:

The third confidence level is determined according to the number, coverage, and acquisition quality of the sensors, the map and location used by the target automatic driving strategy, and other vehicles passing ahead of the driving path.
The confidence degree determination device according to claim 10 or 11, wherein the fourth confidence degree determination module, according to the first confidence degree, the second confidence degree and the third confidence degree, determines Fourth degree of confidence, including:

Weighting the first confidence degree, the second confidence degree, and the third confidence degree to obtain the fourth confidence degree.
The confidence degree determining device according to claim 10 or 11, wherein the fifth confidence degree determining module determines the Confidence levels for autonomous driving policies, including:

Adding the fourth confidence level and the fifth confidence level to obtain the confidence level of the automatic driving strategy.
The confidence degree determination device according to any one of claims 10-16, wherein the confidence degree determination device further comprises: an interaction module;

The interaction module is used to remind the driver of the vehicle of the confidence level of the automatic driving strategy through a multi-modal reminder; wherein the multi-modal includes: visual, auditory, tactile and olfactory one or more.
The confidence degree determination device according to claim 17, wherein the interaction module is also used for:

When the driver's fatigue is detected, when the vehicle breaks down, when the automatic driving strategy exits abnormally and/or when the automatic driving has been completed, a prompt is sent to the driver.
An electronic device, characterized in that the electronic device includes: a processor and a memory; a computer program is stored in the memory, and when the processor executes the computer program, the electronic device performs the following steps: The method described in any one of 1-9.
A computer-readable storage medium, characterized in that instructions are stored in the computer storage medium, and when the instructions are run on a computer, the computer executes the method according to any one of claims 1-9.
A computer program product, characterized in that, when the computer program product runs on a computer, the computer executes the method according to any one of claims 1-9.