Driving assistance device, system and method for vehicle

Abstract

The present invention provides a driving assistance device, system and method for a vehicle. The device is provided with a plurality of information categories and a traffic regulation database containing traffic regulation information, and the driving assistance device comprises: an acquisition module configured to acquire vehicle status information and environmental information around the vehicle; a decision module configured to make a driving behavior decision for a vehicle in an automatic driving mode based on the acquired information; a preprocessing module configured to classify the acquired information into corresponding information categories among the plurality of information categories; and a judgment module configured to acquire one or more traffic regulations related to the preprocessed information from the traffic regulation database; judge whether the driving behavior decided violates at least one of the acquired traffic regulations; if it is judged to violate at least one traffic regulation, further judge the legal penalty risk, property penalty risk and personal injury risk of the driving behavior decided.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of vehicle driving assistance, and in particular to a driving assistance device for a vehicle, including a driving assistance system of the device, a vehicle provided with the driving assistance device or system, and a corresponding driving assistance method. Background Art

[0002] Driving assistance technology is a research hotspot and an important development direction for the automotive industry. In existing technologies, there are many aspects of driving assistance. When a vehicle with driving assistance function participates in traffic in an autonomous driving mode, it needs to abide by traffic rules.

[0003] In this regard, in one existing solution, it is detected whether the autonomous driving vehicle violates traffic regulations, and the event of the detected autonomous driving vehicle violating traffic regulations is sent to the background server or traffic management department. In another existing solution, the detected vehicle driving behavior is compared with the traffic regulations, and when the driving behavior deviates from the traffic regulations, the vehicle control is intervened or a warning message is output to the driver and the violation event is recorded. It can be seen that the existing solutions are mainly concerned with determining whether the vehicle's autonomous driving behavior violates traffic regulations, and issuing warnings, interventions and recordings when it is determined to violate traffic regulations.

[0004] However, for an autonomous driving vehicle, when it is detected that it may violate traffic regulations, how to assist the vehicle in making optimal behavioral decisions so that the autonomous driving vehicle can avoid risks to the greatest extent in the current scenario, there is no better solution in the existing technology. Summary of the invention

[0005] A brief summary of one or more aspects is given below to provide a basic understanding of such aspects. This summary is not a general overview of all aspects, and is not intended to identify the key or important elements of all aspects, nor is it intended to describe the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a preface to the more detailed description given later.

[0006] According to one aspect of the present invention, there is provided a driving assistance device for a vehicle, which is provided with a plurality of information categories and a traffic regulations database containing traffic rules information, the driving assistance device comprising: an acquisition module, configured to acquire vehicle status information and environmental information around the vehicle; a decision module, configured to make driving behavior decisions for a vehicle in an automatic driving mode based on the acquired information; a preprocessing module, configured to classify the acquired information into corresponding information categories among the plurality of information categories; and a judgment module, configured to acquire one or more traffic regulations related to the classified information from the traffic regulations database; judge whether the decided driving behavior violates at least one of the acquired traffic regulations; if it is determined to violate at least one traffic regulation, further judge the legal penalty risk, property penalty risk and personal injury risk of the decided driving behavior.

[0007] According to one embodiment, the driving assistance device includes a closed-loop control with the judgment result as a feedback parameter, so that the decision module outputs a stable driving behavior decision.

[0008] According to one embodiment, the closed-loop control comprises:

[0009] Step (a): the judgment module makes a judgment based on the received decision and outputs the judgment result to the decision module;

[0010] Step (b): the decision module makes a latest decision based on the judgment result, and outputs the latest decision to the judgment module; and

[0011] The above steps (a) and (b) are executed repeatedly until the decision output by the decision module does not change.

[0012] According to one embodiment, the traffic regulation information in the traffic regulation database is processed in an external device and includes at least one of the following:

[0013] (1) a label used to identify one or more of the following: vehicle behavior, scene, weather, object status, and penalty criteria;

[0014] (2) various combinations of vehicle behavior and vehicle surroundings and penalty criteria for each combination; and

[0015] (3) Logically processed traffic rules items include conditional elements and result elements. The conditional elements include the vehicle behavior and the environment in which the behavior occurs, and the result elements include the penalty criteria for the vehicle behavior.

[0016] According to one embodiment, the plurality of information categories at least partially correspond to labels of the traffic rule information, and the determination module matches the classified information with the labels to obtain relevant traffic rules.

[0017] According to one embodiment, the plurality of information categories at least partially correspond to the plurality of combinations, and the determination module matches the classified information with the combinations to obtain relevant traffic regulations.

[0018] According to one embodiment, the judgment module is configured to obtain the relevant one or more traffic rules by means of a traffic model; and use the classified information as a model input, and output the relevant one or more traffic rules after being processed by the traffic model.

[0019] According to one embodiment, the plurality of information categories include one or more of a scene category, a weather and environment category, a vehicle status category, and an object status category.

[0020] According to one embodiment, some or all of the multiple categories each include subcategories.

[0021] According to one embodiment, the traffic rule information in the traffic rule database includes:

[0022] Traffic regulations information that can be obtained from an external device, including traffic regulations applicable to the vehicle's driving location and traffic regulations updated over time;

[0023] Traffic regulation information that is instantly recognized by the vehicle-mounted sensor and is inconsistent with or missing from the traffic regulation applicable to the vehicle's driving position; and

[0024] Traffic regulation information that is identified by the roadside facility and sent to the vehicle and is inconsistent with or missing from the traffic regulations applicable to the vehicle's driving position.

[0025] According to one embodiment, the pre-processing module is configured to:

[0026] classifying the same type of information obtained by different sensors into the same corresponding information category among the plurality of information categories; and

[0027] The information classified into the same information category is fused and calculated so that the fusion calculation results can be used for subsequent processing.

[0028] According to one embodiment, the judgment module is configured to determine the degree of legal liability that needs to be borne for violating traffic regulations in the decided driving behavior as the legal penalty risk.

[0029] According to one embodiment, the degree of legal liability is expressed as a percentage between 0% and 100% or as a number of levels.

[0030] According to one embodiment, the judgment module is configured to determine the degree of property penalty for the driving behavior that has been decided to violate traffic regulations as the property penalty risk.

[0031] According to one embodiment, the degree of property penalty is expressed in several levels based on the penalty amount.

[0032] According to one embodiment, the judgment module is configured to determine the degree of personal injury caused by the driving behavior of the decision, including the number of injured persons and casualties, as the personal injury risk.

[0033] According to one embodiment, the degree of personal injury is expressed by a number of levels that combine the number of injured persons and casualties.

[0034] According to one embodiment, the decision module is further configured to decide whether to start the autonomous driving mode based on the acquired information before making the first autonomous driving behavior decision;

[0035] When the current state of the vehicle meets the requirements for autonomous driving and there is no risk of the vehicle colliding with surrounding objects, the decision is to turn on the autonomous driving mode; and

[0036] When the current state of the vehicle does not meet the requirements of autonomous driving or there is a risk of the vehicle colliding with surrounding objects, the decision is not to turn on the autonomous driving mode.

[0037] According to another aspect of the present invention, there is provided a driving assistance system for a vehicle, comprising: a communication interface capable of wirelessly communicating with an external device, configured to receive vehicle status information, environmental information and traffic regulations information from the external device; on-board sensors, including environmental sensors and vehicle status sensors, configured to collect environmental information around the vehicle and status information of the vehicle; and a driving assistance device as described above, communicatively connected to the communication interface and the on-board sensors respectively, configured to receive information from the communication interface and the on-board sensors, and perform driving assistance for the vehicle based on the received information.

[0038] According to another aspect of the present invention, a driving assistance method for a vehicle is provided. Optionally, the method is performed by the device and / or the system as described above, and the method includes:

[0039] Obtain vehicle status information and environmental information around the vehicle;

[0040] Make driving behavior decisions for the vehicle in autonomous driving mode based on the acquired information;

[0041] Classifying the acquired information into a corresponding information category among a plurality of pre-set information categories;

[0042] Obtaining one or more traffic regulations related to the pre-classified information from a stored traffic regulations database;

[0043] Determine whether the driving behavior decided violates at least one of the acquired traffic regulations; and

[0044] If it is determined that at least one traffic regulation is violated, the legal penalty risk, property penalty risk and personal injury risk of the decided driving behavior will be further judged.

[0045] According to yet another aspect of the present invention, a machine-readable storage medium is provided, which stores executable instructions. When the instructions are executed, a processor is enabled to perform the method described above.

[0046] It is understood that one or more of the above embodiments include the features described in detail below and specifically pointed out in the claims. The following description and drawings set forth in detail certain illustrative features of the one or more aspects. These features are merely indicative of the various ways in which the principles of various aspects may be implemented, and the present disclosure is intended to include all such aspects and their equivalents. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1A An exemplary traffic environment is schematically illustrated in which some implementations of the present invention may be implemented.

[0048] Figure 1B Another exemplary traffic environment is schematically illustrated in which some implementations of the present invention may be implemented.

[0049] Figure 2 A vehicle provided with a driving assistance system according to an embodiment of the present invention is shown.

[0050] Figure 3 A driving assistance device according to an embodiment of the present invention is schematically shown.

[0051] Figure 4 The figure schematically shows a process of determining whether to start the automatic driving mode according to an embodiment of the present invention.

[0052] Figure 5A Schematically shows Figure 3 The working principle of the driving assistance equipment in the vehicle to perform the driving assistance process.

[0053] Figure 5B The driving assistance process according to one embodiment of the present invention is schematically shown.

[0054] Figure 6 is a flowchart of a driving assistance method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] Figure 1A An exemplary traffic environment in which some implementations of the present invention may be implemented is schematically shown. Figure 1A , vehicle V is in the autonomous driving mode in the lane L_ V When driving on the road, a pedestrian suddenly appears in front of the vehicle V. At this time, the vehicle V has three choices, namely, ① turn left, which will lead to a double solid line violation; ② brake to avoid hitting the pedestrian, which will lead to a collision with the vehicle V1 behind the vehicle V; ③ turn right, which will occupy the bus lane L_ BUS .

[0056] Figure 1B Another exemplary traffic environment in which some implementations of the present invention may be implemented is schematically illustrated. Figure 1B , vehicle V is in the autonomous driving mode in the lane L_ V When driving on the road, a group of people suddenly appeared in front of vehicle V. At this time, vehicle V has three choices, namely, ① turn left, vehicle V will enter the left reverse lane and may collide with the oncoming truck T; ② turn right, vehicle V will enter the right non-motorized vehicle lane L_ BIKE And may collide with cyclists on the non-motorized vehicle lane; ③ Emergency braking, but it is still likely to collide with pedestrians.

[0057] When the vehicle is in autonomous driving mode and faces an Figure 1A and Figure 1B In such a traffic environment, with the help of the technical solution of the embodiment of the present invention, an optimized automatic driving strategy can be provided for the vehicle. Specifically, the technical solution of the embodiment of the present invention can provide such a driving assistance strategy for the vehicle, that is, after comprehensively considering whether the vehicle behavior violates traffic rules and the legal penalty risk, property penalty risk and personal injury risk in the case of violation, an optimized automatic driving behavior decision is made. In contrast, the solutions in the prior art only detect vehicle behavior and determine whether the vehicle behavior violates the rules, and there is no driving assistance solution like the present invention.

[0058] Moreover, according to the technical solution of the embodiment of the present invention, a closed-loop control is formed between the judgment of the above four aspects (i.e., whether the vehicle behavior violates traffic regulations and the legal penalty risk, property penalty risk and personal injury risk in the case of violation) for the autonomous driving behavior decision and the decision based on the judgment result, so as to output a stable driving behavior decision. It can be seen that the technical solution of the embodiment of the present invention improves the vehicle's ability to recognize and judge traffic regulations, and helps to assist the vehicle in making optimized driving behavior decisions. In the case where the driving behavior cannot avoid violating traffic regulations, the degree of violation of traffic regulations by the vehicle is minimized.

[0059] Below, some terms in the present invention are explained and defined.

[0060] In the present invention, "traffic rules" or "traffic rule information" or "traffic rules" or "traffic rule information" include such information: regulations restricting vehicle travel (vehicle behavior); penalty standards for vehicles violating regulations; and dynamic updates of the regulations and penalty standards. The regulations include, for example, traffic lights, speed limits, lane attributes (passable direction, bus-only, passable time zone, etc.), lane boundary attributes (whether lane change is allowed, whether overtaking is allowed, etc.), etc. The penalty standards include, for example, liability determination and financial compensation for vehicles violating regulations. The dynamic update may include regulations that are updated as the geographical location of the vehicle changes; it may also include regulations and penalty standards that change over time.

[0061] "Traffic rules" or "traffic rule information" or "traffic rules" or "traffic rule information" may include traffic laws and regulations for human drivers. When an autonomous driving vehicle participates in traffic involving vehicles driven by human drivers, "traffic rules" or "traffic rule information" or "traffic rules" according to an embodiment of the present invention should include traffic laws and regulations for human drivers.

[0062] “Traffic rules” or “traffic rule information” or “traffic rules” or “traffic rule information” may include traffic laws and regulations formulated for autonomous vehicles. Traffic laws and regulations formulated for autonomous vehicles may include traffic laws and regulations customized for different stages of autonomous driving (i.e., driver assistance, semi-autonomous driving, highly autonomous driving, and fully autonomous driving).

[0063] Taking into account that different countries have their own traffic laws and regulations, special traffic laws and regulations may also be customized for different road sections. "Traffic rules" or "traffic rule information" or "traffic rules" or "traffic rules" may include traffic laws and regulations that are in line with the country or road section where the vehicle is currently traveling.

[0064] In addition, “traffic rules” or “traffic rule information” or “traffic rules” or “traffic rule information” may include dynamically updated traffic laws and regulations, such as laws and regulations specifically formulated and effective for autonomous driving as research progresses in the fields of artificial intelligence and law.

[0065] In the present invention, “autonomous driving mode” or “autonomous driving” includes partial autonomous driving or fully autonomous driving with the aid of a driving assistance function of the vehicle.

[0066] Figure 2A vehicle 100 is shown in which a driving assistance system 10 according to an embodiment of the present invention is provided. The vehicle 100 has an assisted driving or an automatic driving function. For example, the vehicle 100 can switch between a human driver driving mode and an automatic driving mode. The automatic driving can include partial automatic driving or full automatic driving. The vehicle 100 can be a car, an engineering vehicle, a bus, a truck, a van, a van or a motorcycle, etc.

[0067] The driving assistance system 10 and its components are described below.

[0068] As shown in FIG1 , the driving assistance system 10 is disposed on a vehicle 100 , and may also be referred to as a vehicle-mounted system 10 . The driving assistance system 10 mainly includes a communication interface 11 , a sensor 12 , and a driving assistance device 13 .

[0069] The communication interface 11 can provide a wireless communication connection between the vehicle 100 and an external device. In other words, the vehicle 100 can exchange information with an external device via the communication interface 11. Any suitable wireless communication method (e.g., 3G / 4G / 5G, C-V2X, DSRC, Wi-Fi, Bluetooth) can be used between the vehicle 100 and the external device to exchange information. An external device refers to a device other than the vehicle 100. The external device is, for example, a cloud server (e.g., a cloud server communicates with the vehicle 100 via a 5G cellular network), an edge server (e.g., an edge server communicates with the vehicle 100 via a 5G cellular network), a roadside facility (e.g., a roadside facility communicates with the vehicle 100 via C-V2X), other vehicles that can communicate wirelessly with the vehicle 100 (e.g., other vehicles communicate with the vehicle 100 via C-V2X), and electronic devices that can communicate wirelessly with the vehicle 100 (e.g., smartphones carried by pedestrians around the vehicle 100).

[0070] The information received from the external device via the communication interface 11 is transmitted to the driving assistance device 13 so as to participate in information processing in the driving assistance device 13. The information received from the external device via the communication interface 11 may include (1) environmental information around the vehicle 100, (2) status information of the vehicle 100, (3) map and vehicle navigation information, and (4) traffic regulation information.

[0071] (1) Environmental information around vehicle 100

[0072] The environmental information may include information about objects around the vehicle 100, especially information about potential collision objects of the vehicle 100. The object information may include the type of object (e.g., obstacles, pedestrians, pets, other vehicles), state (e.g., stationary or moving, moving speed, moving trend), position (e.g., absolute position of the object or relative position to a reference object), and size.

[0073] The environmental information may include climate and weather information. The climate and weather information may include weather conditions and parameters characterizing the weather conditions, such as fog, hail, cloudy, rain, visibility, illumination, rainfall, and the like.

[0074] The environmental information may include scene information. The scene information may be represented by elements representing the scene, such as a highway, an intersection, a one-way street, etc. The elements representing the scene may be obtained by collecting environmental information through semantic recognition (e.g., intelligent semantic recognition using AI technology).

[0075] (2) Status information of vehicle 100

[0076] The status information of the vehicle 100 may include: the driving direction, speed and acceleration of the vehicle 100; the current lane of the vehicle 100 and the position relative to the lane; the movement trend of the vehicle 100, and so on.

[0077] (3) Maps and vehicle navigation information

[0078] The map and vehicle navigation information may include map data and navigation routes of the driving section of the vehicle 100. The map and vehicle navigation information may be automatically sent to the vehicle 100 based on the positioning information of the vehicle 100 by the external device, or may be sent to the vehicle 100 based on a request from the vehicle 100 to the external device.

[0079] The above-mentioned environmental information, vehicle status information, and map and vehicle navigation information can be collected by sensors of external devices, and processed in the processor of the external device to obtain the information required by the driving assistance system 10, and then the processed information is sent to the communication interface 11 of the vehicle 100 via the wireless communication unit at the external device. In this way, the amount of calculation in the driving assistance system 10 of the vehicle 100 can be reduced, saving the computing power in the vehicle 100.

[0080] (4) Traffic rules information

[0081] Traffic rule information may include traffic rules for human drivers, traffic rules for autonomous vehicles, and penalty standards for violating traffic rules.

[0082] Traffic regulation information may include “static traffic regulation information” and “dynamic traffic regulation information”.

[0083] "Static traffic rules information" refers to information obtained from external devices. For example, "static traffic rules information" may include traffic rules that have been formulated and recorded in traffic rules manuals / traffic rules guidelines, as well as traffic rules that are updated with the advancement of technology and laws (these updated traffic rules can be stored in the cloud server and sent to the vehicle end).

[0084] In contrast to "static traffic regulation information", "dynamic traffic regulation information" may include traffic regulation information that is instantly recognized by the on-board sensors and is inconsistent with the traffic regulations applicable to the current geographic location of the vehicle in the static traffic regulations or is missing in the traffic regulations applicable to the current geographic location of the vehicle. For example, the current section of road is under construction, roadblocks are set up and speed limits are set. This speed limit information belongs to "dynamic traffic regulation information". In this case, the vehicle 100 can transmit the recognized "dynamic traffic regulation information" to an external device (for example, upload it to a cloud server) so that it can be integrated and updated in the external device for reference by other vehicles that pass through the location later.

[0085] "Dynamic traffic regulation information" may also include traffic regulation information that is identified by a sensing device (e.g., a camera) at a roadside facility and is inconsistent with or missing from the traffic regulation applicable to the current geographic location of the vehicle. Such traffic regulation information is, for example, temporary roadblocks or speed limit information. When the vehicle 100 or other vehicles travels near the roadside facility, the identified traffic regulation information is sent to the vehicle 100 or other vehicles via a wireless communication unit at the roadside facility.

[0086] Thus, the vehicle 100 can obtain traffic regulation information including traffic regulation information obtained by the vehicle 100 from an external device and information instantly recognized by the vehicle 100. In this way, the traffic regulation information involved in subsequent processing in the driving assistance device 13 can fully comply with the traffic regulations required by the current road section where the vehicle is traveling or the current position where the vehicle is located.

[0087] It can be understood that the above description about traffic rule information is also applicable here.

[0088] In one embodiment, traffic rule information stored in an external device (e.g., a cloud server) is processed in the external device, and then the processed traffic rule information is sent from the external device to the vehicle 100, for example, to a communication interface of the vehicle 100.

[0089] In the external device, the traffic regulation information can be processed in this way, that is, the traffic regulation items contained in the traffic regulation information are semantically recognized (for example, natural semantic recognition), so as to extract the elements of interest contained therein (for example, vehicle behavior, scene, weather, object state and penalty standard). The semantic recognition and information extraction described here can be implemented by artificial intelligence (for example, a trained machine learning model). The elements of interest can be marked on the traffic regulation information by "labeling".

[0090] By extracting elements of interest from a large amount of traffic regulation information, it is possible to quickly and without omission extract desired traffic regulation items and penalty standards from a large amount of traffic regulation information without traversing all traffic regulation items.

[0091] Elements of interest can include one or more of the following.

[0092] (1) Extract keywords for vehicle driving behavior categories, scene categories, weather and climate categories, object categories, and violation penalty categories, and use the extracted keywords as elements of interest and attach "labels" to the corresponding traffic regulation items in each category. In this way, the processed traffic regulation information can contain labels for representing one or more of the vehicle behavior, scene, weather, object status, and penalty criteria.

[0093] (2) Combining the vehicle behavior category with one or more of the scene category, weather and climate category, and object category, and determining a penalty standard for each combination. In this way, the processed traffic regulation information may include multiple combinations of vehicle behavior and vehicle surroundings and a penalty standard for each combination.

[0094] (3) Logically process the traffic regulation items in the traffic regulation information in an external device to obtain a logicalized traffic regulation item. The logicalized traffic regulation item contains two elements, namely, a condition element and a result element. The condition element contains the vehicle behavior and the scenario in which the behavior occurs. The result element contains the penalty standard for the behavior. The condition element is, for example, "if a certain vehicle behavior is triggered (occurred), and the vehicle behavior occurs in a certain scenario", and the result element is, for example, "then according to the corresponding traffic regulations, it should be judged as illegal, the vehicle bears 100% responsibility and is fined a certain amount of money". In an example of a logicalized traffic regulation item, it includes "If the vehicle is traveling in a geographic fence area and the speed exceeds 80km / h, it violates the traffic regulations and is fined."

[0095] It is understood that the scene element may include content representing scenes such as highways, tunnels, and elevated roads. The weather and climate element may include content representing climate and weather such as visibility, lighting, and rainfall. The vehicle behavior element may include content representing vehicle behavior such as vehicle speed. The object element may include content representing objects such as the type of potential collision objects of the vehicle and their behavior.

[0096] In addition, these element types can be further divided into subcategories. For example, the scene type elements can be further divided into subcategories such as highways, one-way streets, and tunnels. By further dividing the subcategories, a more detailed classification of traffic regulation information can be achieved, thereby greatly reducing the operation time and improving the operation efficiency when extracting the required traffic regulation items from the traffic regulation information classified in this way.

[0097] The sensor 12 is disposed on the vehicle 100 , and thus can be referred to as a vehicle-mounted sensor. The information collected by the sensor 12 is transmitted to the driving assistance device 13 via a vehicle bus, so as to participate in information processing in the driving assistance device 13 .

[0098] The sensor 12 may include an environmental sensor for capturing environmental information around the vehicle. The environmental sensor may include a camera (e.g., single target, multi-target, surround view) and / or a radar (e.g., a laser radar, an ultrasonic radar, a millimeter wave radar). The camera may obtain environmental information around the vehicle 100 through image or video analysis, for example, the relative distance of the vehicle from the roadside or an obstacle may be obtained. The radar may obtain the relative distance of the vehicle from the roadside or an obstacle by analyzing the point cloud. The environmental sensor may include a plurality of environmental sensors arranged around the vehicle body, and the arrangement takes into account safety redundancy, that is, ensuring that the environmental conditions around the vehicle can be fully collected.

[0099] The sensor 12 may also include a vehicle state sensor for capturing vehicle state information. The vehicle state sensor may include a sensor that directly or indirectly measures vehicle state parameters, such as a wheel speed sensor, a suspension displacement sensor, an acceleration sensor, a steering angle sensor, and the like.

[0100] The driving assistance device 13 is connected to the communication interface 11 and the sensor 12 for communication respectively. A traffic regulation database and a plurality of information categories are provided in the driving assistance device 13. The processed traffic regulation information received via the communication interface 11 is stored in the traffic regulation database. The information received via the communication interface 11 (information other than traffic regulation information) and the information received via the vehicle bus (information received by the sensor 12 and via the vehicle bus) are classified in the driving assistance device 13 based on the set information categories, and then the classified information is used for subsequent information processing.

[0101] The driving assistance device 13 includes logic / strategies for driving assistance. The strategies include: obtaining vehicle status information, environmental information around the vehicle, and traffic rules information; making driving behavior decisions for the vehicle in the automatic driving mode based on the obtained information; classifying the obtained information into corresponding information categories in a plurality of pre-set information categories; and extracting one or more traffic rules related to the pre-classified information from the stored traffic rules database; judging whether the driving behavior decided violates at least one of the extracted traffic rules; if it is judged to violate at least one traffic rule, further judging the legal penalty risk, property penalty risk, and personal injury risk of the driving behavior decided.

[0102] The strategy also includes: closed-loop control using the judgment result as a feedback parameter to help the decision module output a stable driving behavior decision. The closed-loop control includes: cyclic execution of (a) the judgment module makes a judgment based on the received decision and outputs the judgment result to the decision module; (b) the decision module makes a latest decision based on the judgment result and outputs the latest decision to the judgment module until the decision output by the decision module no longer changes.

[0103] The driving assistance device 13 may be provided in the ECU of the vehicle 100 or in the domain controller of the vehicle 100 .

[0104] The driving assistance device 13 can be implemented in hardware or software or a combination of software and hardware. For the hardware-implemented part, it can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, electronic units designed to perform their functions, or a combination thereof. For the software-implemented part, it can be implemented with the aid of microcodes, program codes, or code segments, and they can also be stored in a machine-readable storage medium such as a storage component.

[0105] In one embodiment, the driving assistance device 13 may be implemented as a memory and a processor. The memory stores a traffic regulation database, multiple information categories, and instructions, which, when executed by the processor, enable the processor to execute the driving assistance strategy / method according to the embodiment of the present invention.

[0106] Figure 3 FIG. 2 shows a schematic block diagram of a driving assistance device according to an embodiment of the present invention. Figure 3 As shown, the driving assistance device 13 includes an acquisition module 131, a preprocessing module 132, a decision module 133 and a judgment module 134. An information category is set in the preprocessing module 132 so that the information acquired by the acquisition module can be classified in the preprocessing module. A traffic regulation database is set in the judgment module, in which the processed traffic regulation information as described above is stored, so as to extract the traffic regulation that matches the classified information from the traffic regulation database, that is, the traffic regulation required for this judgment.

[0107] These modules of the driving assistance device 13 are, for example, software modules. The naming of each module should be understood as a logical (functional) description, and should not be understood as a limitation of the physical form or setting method. In other words, one or more of the acquisition module 131, the preprocessing module 132, the decision module 133 and the judgment module 134 can be implemented in the same chip or circuit, and they can also be respectively arranged in different chips or circuits, which is not limited by the present invention. Any of these modules can be further divided into sub-modules or combined together in terms of function.

[0108] Figure 4 FIG. 4 shows a process 400 for determining whether to turn on the automatic driving mode according to an embodiment of the present invention. The process 400 may be performed in the driving assistance device 13 according to an embodiment of the present invention. Figure 3 The process 400 is implemented in the driving assistance device 13 shown. Therefore, the above related description is also applicable here.

[0109] like Figure 4 As shown, in box 402, the acquisition module 131 acquires information, which includes traffic rule information, environmental information and vehicle status information. The acquired information includes information collected by the external device and transmitted to the driving assistance device 13 via the communication interface 11 and information collected by the vehicle sensor 12 and transmitted to the driving assistance device 13 via the vehicle bus. The present invention does not limit the order in which the information is acquired.

[0110] In box 404 , the judgment module 134 determines whether the current state of the vehicle 100 meets the requirements of autonomous driving based on the acquired information, and sends the judgment result to the decision module 133 .

[0111] In one embodiment, the judgment module 134 judges whether the vehicle 100 has at least one of the following situations based on the acquired information: the door or trunk lid is open; the vehicle head is not in the direction indicated by the navigation route; a vehicle sensor failure for autonomous driving is detected so that the vehicle safety cannot be guaranteed; a vehicle locator failure for autonomous driving is detected so that the vehicle navigation function fails; and the current area is not a legally permitted autonomous driving area. When it is determined that the vehicle 100 has one of the above situations, the judgment result is that the current state of the vehicle 100 does not meet the requirements of autonomous driving. When it is determined that the vehicle 100 does not have any of the above situations, the judgment result is that the current state of the vehicle 100 meets the requirements of autonomous driving.

[0112] In box 406, the decision module 133 makes a decision on whether to turn on the automatic driving based on the acquired information and the judgment result.

[0113] When the judgment result is that the current state of the vehicle 100 meets the requirements of the automatic driving, and the decision module 133 determines that the vehicle 100 does not have the risk of colliding with surrounding objects based on the acquired information, the decision module 133 makes a decision to turn on the automatic driving (Turn-On-AD). When the decision module 133 makes a decision to turn on the automatic driving, the driving assistance device 13 can implement the driving assistance process 500 described below.

[0114] When the judgment result is that the current state of the vehicle 100 does not meet the requirements of the automatic driving, or the decision module 133 determines that the vehicle 100 has a risk of colliding with surrounding objects based on the acquired information, the decision module 133 makes a decision not to start the automatic driving. In this case, the vehicle driver can be reminded to adjust the current state of the vehicle to meet the requirements of the automatic driving, and wait until the collision risk is eliminated before making a decision to start the automatic driving.

[0115] Figure 5A FIG. 3 shows the working principle of the driving assistance process implemented by the driving assistance device 13 . Figure 5B FIG. 5 shows a driving assistance process 500 according to an embodiment of the present invention. The process 500 may be executed in the driving assistance device 13. For example, the process 500 is implemented in the driving assistance device 13 as shown in FIG. 3. Therefore, the above related description is also applicable to this.

[0116] See also Figure 5B In box 502, similar to box 402, the acquisition module 131 acquires information, which includes traffic rule information, environmental information and vehicle status information. The acquired information includes information collected by the external device and transmitted to the driving assistance device 13 via the self-communication interface 11, and information collected by the vehicle sensor 12 and transmitted to the driving assistance device 13 via the vehicle bus. The present invention does not limit the order in which the information is acquired.

[0117] In block 504, the decision module 133 makes a decision (i.e., the output of the decision module) on the autonomous driving behavior of the vehicle 100 based on its input (e.g., the acquired information and the judgment result of the judgment module). The autonomous driving behavior of the decision includes various driving behaviors when the vehicle is in an automatic or semi-automatic driving mode, such as steering with a calculated steering angle, accelerating with a calculated acceleration, braking with a calculated braking force, and the like.

[0118] In box 506 , the preprocessing module 132 preprocesses the acquired information, including information classification (box 5061 ) and information fusion (box 5062 ).

[0119] A plurality of predetermined information categories are stored in the preprocessing module 132. These information categories correspond at least partially to the labels or element combinations of the traffic regulation information. In other words, the two processes of setting the information categories and labeling the traffic regulation elements or setting the element combinations are related. In this way, when the subsequent processing is performed based on the traffic regulation information and the classified information, the efficiency of information processing can be greatly improved because there is a correlation factor between the two information.

[0120] In one embodiment, in block 5061, the acquired information is classified into a corresponding category among a plurality of pre-set information categories.

[0121] See also Figure 5A , multiple information categories are set in the preprocessing module 132. The multiple information categories may include a scene class, a weather and environment class, a vehicle status class, and an object class. Some or all of the multiple categories may include multiple subcategories, respectively. For example, the scene class includes a highway subcategory, an intersection subcategory, a one-way street subcategory, and the like. The weather and environment class may include a visibility subcategory, an illumination subcategory, a weather subcategory, and the like. The vehicle status class may include a lane type subcategory, a driving trend subcategory, a motion parameter subcategory, and the like. The object class may include an object type subcategory, an object position subcategory, an object motion speed subcategory, and the like.

[0122] Understandably, Figure 5A Only some examples of information categories and their subcategories are shown. According to an embodiment of the present invention, an information category may include more information categories or information categories divided in other ways. A subcategory may also include more information categories or information categories divided in other ways.

[0123] In one embodiment, for a plurality of acquired information, the same type of information acquired via different sensors (for example, the different sensors include a communication interface for sending and receiving information and an onboard sensor for sensing information) is classified into the same corresponding information category in multiple information categories. For example, if the acquired information includes illumination information received via the communication interface and illumination information sensed by the onboard sensor, then both illumination information are classified into the weather and environment category. In the case where the weather and environment category includes an illumination subcategory, both illumination information are classified into the illumination subcategory.

[0124] In one embodiment, in block 5062, information acquired by different sensors and classified into the same information category is fused and calculated so as to participate in subsequent information processing with the result of the fusion calculation. For example, the illumination information received by the communication interface and the illumination information sensed by the vehicle-mounted sensor are fused and calculated, and the illumination information as the fusion result is used in subsequent information processing.

[0125] It is advantageous to provide multiple information categories in the driving assistance device 13 and classify the acquired information according to the corresponding categories, and then use the classified information to participate in subsequent information processing, because there may be redundancy in the acquired information, that is, the same type of information is acquired through multiple sensors, and a streamlined information set can be obtained by classifying them. Furthermore, by fusing the same type of information, an optimized information set can be obtained. The streamlined and optimized information set obtained by such information processing participates in subsequent information processing, which can greatly reduce the workload of subsequent information processing and improve work efficiency. For example, when the judgment module uses the classified and fused information set for information processing, the workload of the judgment module can be greatly reduced and the operating efficiency of the judgment module can be improved.

[0126] In block 508 , the determination module 134 obtains relevant traffic regulations from the traffic regulations database based on the pre-processed information.

[0127] In one embodiment, the judgment module 134 matches the preprocessed information with the processed traffic regulation information so as to extract the traffic regulation items that are successfully matched. In the matching process, the labels of the preprocessed information categories and subcategories can be matched with the labels of the traffic regulation information, and the information with matching labels can be extracted. For example, the label "illuminance" of the preprocessed illumination information is matched with the label "illuminance" in the traffic regulation information, thereby extracting all traffic regulations involving illumination in the traffic regulation library.

[0128] In another embodiment, the judgment module 134 obtains the relevant traffic rules by means of a predetermined traffic model. The traffic model can be an AI model that has undergone deep learning. The preprocessed information (e.g., classified information or classified and fusion-calculated information) is used as a model input, and after being processed by the traffic model, the traffic rules related to the input information are output, thereby obtaining the relevant traffic rules.

[0129] It can be seen that, based on the processing of traffic regulations information and the pre-processing of the acquired information, in the process of extracting traffic regulations, there is no need to traverse all traffic regulations in the traffic regulations library, and relevant traffic regulations can be extracted quickly and without omission, which greatly reduces the workload and improves work efficiency.

[0130] In box 510, the judgment module 134 makes a judgment on the autonomous driving behavior decided by the decision module 133. The judgment module 134 first judges whether the driving behavior decided violates at least one of the extracted traffic rules. When the judgment module 134 judges that the driving behavior decided does not violate any of the extracted traffic rules, the conclusion of "the driving behavior decided does not violate the traffic rules" is output. When the judgment module 134 judges that the autonomous driving behavior decided violates one or more traffic rules, the judgment of the three aspects of legal penalty risk, property penalty risk and personal injury risk is further performed on the autonomous driving behavior decided.

[0131] The judgment module 134 determines the legal penalty risk based on the extracted traffic regulations and the driving behavior of the decision. The legal penalty risk can be expressed by the degree of legal responsibility that the autonomous driving behavior of the decision needs to bear in the case of violation (i.e., the degree of responsibility of the vehicle) according to the penalty standards stipulated in the traffic regulations. The degree of legal responsibility can be expressed as a percentage or as a level.

[0132] In one embodiment, the degree of legal responsibility is expressed as a percentage between 0% and 100% according to the penalty standard. The percentage changes accordingly as the legal responsibility of the vehicle 100 changes. 100% means that the vehicle 100 is fully responsible. 0% means that the vehicle 100 is not responsible. 50% means that the vehicle 100 bears half of the responsibility.

[0133] In another embodiment, the degree of legal responsibility is represented by several levels according to the penalty standard. For example, between levels 1-n (n is a natural number greater than 1), level 1 indicates that the vehicle 100 is fully responsible, and level n indicates that the vehicle 100 is not responsible. In this embodiment, the levels can be divided according to 100% of the responsibility assumed by the vehicle 100. For example, 100% corresponds to level 1, 80% to 100% (including 80% but not including 100%) corresponds to level 2, and so on, 0% to 20% corresponds to level 6 (including 20% ​​but not including 0%), and 0% corresponds to level 7.

[0134] The judgment module 134 determines the property penalty risk based on the acquired traffic regulations and the driving behavior of the decision. The property penalty risk can be expressed by the degree of property penalty imposed on the autonomous driving behavior of the decision in the case of violation according to the penalty standards stipulated in the traffic regulations. The degree can be expressed in grades.

[0135] In one embodiment, the degree of property penalty can be represented by several levels according to the penalty standard. For example, between levels 1-m (m is a natural number greater than 1), level 1 represents the most severe property penalty, and level m represents no property penalty. The level number can represent the severity of the property penalty. For example, the larger the level number, the smaller the fine or penalty imposed.

[0136] In this embodiment, the level of property penalty can correspond to the penalty amount. For example, according to traffic regulations, several levels are divided between the maximum amount and the minimum amount of the fine or penalty (the maximum amount and the minimum amount can refer to the maximum penalty amount and the minimum penalty amount in all traffic regulations; they can also refer to the maximum penalty amount and the minimum amount in the obtained traffic regulations), and the level corresponds to the amount of the penalty.

[0137] The judgment module 134 determines (predicts) the risk of personal injury based on the driving behavior of the decision and the information obtained. The risk of personal injury can be expressed in terms of the degree of personal injury that will be caused by the driving behavior of the decision according to the penalty standards stipulated in the traffic regulations. The degree of personal injury can be comprehensively determined based on the estimated number of injured people and the degree of casualties. For example, based on the driving behavior of the decision and the fact that the vehicle 100 will collide head-on with a pedestrian at a speed of 40 km / h, the degree of personal injury is determined by estimating the personal injury to pedestrians that will be caused in this case.

[0138] In one embodiment, the number of injured persons and the degree of casualties may be comprehensively assessed and divided into several levels, for example, level 1-t (t is a natural number greater than 1), where level 1 indicates the highest degree of personal injury and level t indicates no personal injury.

[0139] For clarity, some examples of judgments made by the judgment module 134 based on the preprocessed information and the driving behavior of the decision are shown in the following Table 1 by way of example. In the example of Table 1, the degree of personal injury is represented by the most serious level-the lightest level, that is, level 1-level 5. The degree of property penalty is represented by the most serious level-the lightest level, that is, level 1-level 5. Legality refers to whether the driving behavior of the vehicle 100 (i.e., the vehicle) executing the decision complies with the traffic regulations, wherein "yes" means that it complies with the traffic regulations (legal), and "no" means that it does not comply with the traffic regulations (illegal).

[0140] Table 1

[0141]

[0142] After completing the above-mentioned judgments, the judgment module 134 outputs the judgment result (including the above-mentioned judgment results) to the decision module 133 .

[0143] Next, the judgment result is used as a feedback parameter to perform closed-loop control so that the decision module outputs a stable driving behavior decision. The closed-loop control includes cyclic execution of (a) the judgment module makes a judgment based on the received decision and outputs the judgment result to the decision module; (b) the decision module makes a latest decision based on the judgment result and outputs the latest decision to the judgment module until the decision output by the decision module no longer changes.

[0144] The input of the decision module may include judgment result input and other inputs. "Judgment result input" refers to the judgment result received by the decision module and made by the judgment module. "Other input" refers to the input received by the decision module other than the judgment result. "Other input" is, for example, information input to the decision module and acquired by the acquisition module. "Other input" may include vehicle status information, object status information, and environmental information involved in the decision process.

[0145] In other words, the closed-loop control includes taking the judgment result as a feedback parameter, cyclically executing box 504 and box 510, the judgment module outputs the judgment result to the decision module, and the decision module makes a latest decision based on the judgment result and outputs the latest decision to the judgment module.

[0146] In one embodiment, referring to Examples 1 to 3 in Table 1 above, for the same scene composition and the same vehicle state, the driving behavior of the first decision is emergency braking and hitting a pedestrian. The decision is output to the judgment module. The judgment module makes four judgments as in Example 1 for the received decision. Then, the judgment result is output to the decision module. The driving behavior of the second decision made by the decision module based on the judgment result is emergency braking, changing lanes to the reverse lane, and hitting a truck. The decision is output to the judgment module, and the judgment module makes four judgments as in Example 2. Then, the judgment result is output to the decision module. The third decision behavior made by the decision module based on the received judgment result is emergency braking, turning right, and hitting the wall head-on with a collision surface of 30%. The decision is output to the judgment module, and the judgment module makes four judgments as in Example 3. Then, the judgment result is output to the decision module, and the decision module makes a decision based on the judgment result, and the decision made this time is the same as the decision made the third time, that is, the decision output by the decision module does not change, and a stable final decision is obtained. In this way, the closed-loop control ends and the decision module outputs a stable (no longer changing) behavior decision.

[0147] The process of determining whether a stable decision is obtained can be executed in the decision module 133 (box 512). For example, when the current scene does not change in the sense of traffic regulations, the decision module determines whether the decisions outputted by it twice in a row (for example, the latest decision and the previous decision) are the same. If the decisions outputted twice in a row are the same, the decision module determines that a stable decision is obtained, and outputs the stable decision as the final decision of this round of loop. If the decisions outputted twice in a row are different, the decision module determines that a stable decision is not obtained, and continues the above loop until a stable decision is obtained.

[0148] When a stable decision is output, the closed-loop system composed of the judgment module and the decision module is in a stable state.

[0149] It should be understood that in a cycle, the elements in the scene (e.g., the elements in the acquired vehicle state information and environment information) may change dynamically in a short period of time, but based on the definition of traffic regulations, the scene is unchanged. In other words, during a cycle, a slight change in the state of the elements in the scene does not change the scene in the sense of traffic regulations. For example, see Figure 1A and Figure 1B After sensing the pedestrian, the pedestrian's movement distance within 1 second does not exceed 1 meter and remains within the vehicle's lane; or it is also possible that the movement of a non-motor vehicle or oncoming vehicle in the lane next to the vehicle's lane changes slightly, but still does not allow the vehicle to obtain sufficient avoidance space. In this way, the vehicle still only has the original three options.

[0150] In addition, the decision output by the decision module remains unchanged, which means that the decision logic remains unchanged. How the vehicle adjusts its own movement to achieve the decision can be changed, that is, the body control to achieve the decision can be changed. For example, in a millisecond cycle, the decision to change lanes to the left is the final stable decision, and the body control to achieve the decision to change lanes to the left may change from turning left 10 degrees to turning left 11 degrees.

[0151] In summary, according to the embodiments of the present invention, it usually takes milliseconds for the decision module and the judgment module to execute one cycle until they reach a stable state. In such a short time, the scene composition (for example, the scene composition includes the combination of various elements in the vehicle's own state information and the environmental information) can generally be considered stable. In other words, during the time of one cycle, the closed-loop system composed of the decision module and the judgment module reaches stability before the scene composition changes.

[0152] According to the driving assistance solution of the embodiment of the present invention, when making autonomous driving behavior decisions, multiple aspects of traffic violation penalties are comprehensively considered, and the degree of violation of various risks is quantified in the form of grades or percentages, so that driving behavior decisions can be accurately measured to minimize the risk level of autonomous driving vehicles in the event of traffic violation.

[0153] According to the driving assistance scheme of the embodiment of the present invention, the driving behavior decision and the judgment of the driving behavior for the decision form a closed-loop control so as to output a stable driving behavior decision with high operating efficiency and ensure the output of optimized driving behavior decision.

[0154] According to the driving assistance solution of the embodiment of the present invention, before executing multiple aspects of judgment in the vehicle, the processing of traffic regulation information is completed in the external device, and the preprocessing of the obtained information is completed in the vehicle, and the two processings are associated in the form of related elements, which greatly reduces the workload of information processing in the judgment process, improves work efficiency, and saves computing power in the vehicle.

[0155] Figure 6 A driving assistance method 600 according to an embodiment of the present invention is shown. The method 600 can be executed in the above-mentioned driving assistance device, or in the above-mentioned driving assistance system. Therefore, the above-mentioned related description is also applicable here.

[0156] See also Figure 6 In step 602, vehicle status information, vehicle surrounding environment information and traffic rules information are obtained.

[0157] In step 604 , a driving behavior decision is made for the vehicle in the autonomous driving mode based on the acquired information.

[0158] In step 606, the acquired information is classified into a corresponding information category among a plurality of pre-set information categories.

[0159] In step 608 , one or more traffic regulations related to the pre-classified information are obtained from a stored traffic regulations database.

[0160] In step 610, it is determined whether the determined driving behavior violates at least one of the acquired traffic regulations;

[0161] If it is determined in step 610 that at least one traffic regulation is violated, step 612 is executed. In step 612, the legal penalty risk, property penalty risk and personal injury risk of the determined driving behavior are further determined.

[0162] If it is determined in step 610 that no traffic regulations are violated, step 614 is executed. In step 614, the vehicle 100 is controlled to execute the driving behavior of the decision.

[0163] The present invention also provides a machine-readable storage medium storing executable instructions, which, when executed, enable a processor to perform the above-mentioned driving assistance method 600.

[0164] It is to be understood that all operations in the method described above are merely exemplary, and the present invention is not limited to any operations in the method or the order of these operations, but should cover all other equivalent transformations under the same or similar concept.

[0165] It is to be understood that all modules described above can be implemented in various ways. These modules can be implemented as hardware, software, or a combination thereof. In addition, any module in these modules can be further divided into submodules or combined together in function.

[0166] It is to be understood that the processor can be implemented using electronic hardware, computer software or any combination thereof. Whether these processors are implemented as hardware or software will depend on specific application and the overall design constraints imposed on the system. As an example, the processor provided in the present invention, any part of the processor or any combination of the processor can be implemented as a microprocessor, a microcontroller, a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic device (PLD), a state machine, a gate logic, a discrete hardware circuit and other suitable processing components configured to perform the various functions described in the present disclosure. The function of the processor provided in the present invention, any part of the processor or any combination of the processor can be implemented as software performed by a microprocessor, a microcontroller, a DSP or other suitable platforms.

[0167] It is to be understood that software should be broadly considered to represent instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, running threads, processes, functions, etc. Software can reside in a computer-readable medium. A computer-readable medium can include, for example, a memory, which can be, for example, a magnetic storage device (e.g., a hard disk, a floppy disk, a magnetic stripe), an optical disk, a smart card, a flash memory device, a random access memory (RAM), a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), a register, or a removable disk. Although the memory is shown as being separated from the processor in the various aspects provided in the present disclosure, the memory can also be located inside the processor (e.g., a cache or a register).

[0168] Although some embodiments have been described above, these embodiments are given by way of example only and are not intended to limit the scope of the invention. The appended claims and their equivalents are intended to cover all modifications, substitutions and changes made within the scope and spirit of the invention.

Claims

1. A driving assistance device for a vehicle, provided with a plurality of information categories and a traffic regulation database containing traffic regulation information, the driving assistance device comprising: An acquisition module is configured to acquire vehicle status information and environment information around the vehicle; a decision module configured to make a driving behavior decision for the vehicle in the autonomous driving mode based on the acquired information; a preprocessing module configured to classify the acquired information into a corresponding information category among the plurality of information categories; as well as The judgment module is configured to obtain one or more traffic regulations related to the classified information from the traffic regulations database; judge whether the driving behavior decided violates at least one of the obtained traffic regulations; if it is determined that at least one traffic regulation is violated, further judge the legal penalty risk, property penalty risk and personal injury risk of the driving behavior decided, Wherein, the determination module is configured to perform at least one of the following: Determine the degree of legal liability that needs to be borne for the driving behavior that violates traffic regulations as the legal penalty risk; Determine the degree of property penalty for the driving behavior that violates traffic regulations as the property penalty risk; as well as The degree of personal injury caused by the driving behavior of the decision, including the number of injured persons and casualties, is determined as the personal injury risk.

2. The driving assistance device according to claim 1, wherein: The driving assistance device includes a closed-loop control with the judgment result as a feedback parameter, so that the decision module outputs a stable driving behavior decision.

3. The driving assistance device according to claim 2, wherein: The closed-loop control comprises: Step (a): the judgment module makes a judgment based on the received decision and outputs the judgment result to the decision module; Step (b): the decision module makes a latest decision based on the judgment result, and outputs the latest decision to the judgment module; The above steps (a) and (b) are executed repeatedly until the decision output by the decision module does not change.

4. The driving assistance device according to claim 1, wherein: The traffic regulation information in the traffic regulation database is processed in an external device and includes at least one of the following: (1) a label used to identify one or more of vehicle behavior, scene, weather, object status, and penalty criteria; (2) Various combinations of vehicle behavior and the vehicle's surrounding environment and the penalty criteria for each combination; and (3) Logically processed traffic rules items include conditional elements and result elements. The conditional elements include the vehicle behavior and the environment in which the behavior occurs, and the result elements include the penalty criteria for the vehicle behavior.

5. The driving assistance device according to claim 4, wherein: The plurality of information categories at least partially correspond to labels of the traffic rule information, and the determination module matches the classified information with the labels to obtain relevant traffic rules.

6. The driving assistance device according to claim 4, wherein: The plurality of information categories at least partially correspond to the plurality of combinations, and the determination module matches the classified information with the combinations to obtain relevant traffic regulations.

7. The driving assistance device according to claim 1, wherein: The judgment module is configured to obtain the one or more relevant traffic rules by means of the traffic rule model; and The classified information is used as a model input, and the related one or more traffic rules are output after being processed by the traffic model.

8. The driving assistance device according to claim 1, wherein: The plurality of information categories include one or more of a scene category, a weather and environment category, a vehicle status category, and an object status category.

9. The driving assistance device according to claim 1, wherein: Some or all of the multiple categories each contain subcategories.

10. The driving assistance device according to claim 1, wherein: The traffic rules information in the traffic rules database includes: Traffic regulations information that can be obtained from an external device, including traffic regulations applicable to the vehicle's driving location and traffic regulations updated over time; Traffic regulation information that is instantly recognized by the vehicle-mounted sensor and is inconsistent with or missing from the traffic regulation applicable to the vehicle's driving position; and Traffic regulation information that is identified by the roadside facility and sent to the vehicle and is inconsistent with or missing from the traffic regulations applicable to the vehicle's driving position.

11. The driving assistance device according to claim 1, wherein: The pre-processing module is configured as follows: classifying the same type of information obtained by different sensors into the same corresponding information category among the plurality of information categories; and The information classified into the same information category is fused and calculated so that the fusion calculation results can be used for subsequent processing.

12. The driving assistance device according to claim 1, wherein: The degree of legal liability is expressed as a percentage between 0% and 100% or in several levels.

13. The driving assistance device according to claim 1, wherein: The degree of property penalties is expressed in several levels based on the amount of the penalty.

14. The driving assistance device according to claim 1, wherein: The degree of personal injury is expressed in terms of several levels that combine the number of injured and casualties.

15. The driving assistance device according to claim 1, wherein: The decision module is further configured to decide whether to start the autonomous driving mode based on the acquired information before making the first autonomous driving behavior decision; When the current state of the vehicle meets the requirements of autonomous driving and there is no risk of the vehicle colliding with surrounding objects, the decision is to turn on the autonomous driving mode; and When the current state of the vehicle does not meet the requirements of autonomous driving or there is a risk of the vehicle colliding with surrounding objects, the decision is not to turn on the autonomous driving mode.

16. A driving assistance system for a vehicle, comprising: a communication interface capable of wirelessly communicating with an external device and configured to receive vehicle status information, environmental information, and traffic regulation information from the external device; The vehicle-mounted sensors include an environmental sensor and a vehicle status sensor, configured to collect environmental information around the vehicle and status information of the vehicle; as well as The driving assistance device as described in any one of claims 1-15 is communicatively connected to the communication interface and the vehicle-mounted sensor respectively, and is configured to receive information from the communication interface and the vehicle-mounted sensor, and perform driving assistance for the vehicle based on the received information.

17. A driving assistance method for a vehicle, the method being executed by the device according to any one of claims 1 to 15 and / or by the system according to claim 16, the method comprising: Obtain vehicle status information and environmental information around the vehicle; Make driving behavior decisions for the vehicle in autonomous driving mode based on the acquired information; Classifying the acquired information into a corresponding information category among a plurality of pre-set information categories; Obtaining one or more traffic regulations related to the pre-classified information from a stored traffic regulations database; Determine whether the driving behavior decided violates at least one of the acquired traffic regulations; as well as If it is determined that at least one traffic regulation is violated, the legal penalty risk, property penalty risk and personal injury risk of the driving behavior are further determined, wherein the further determination includes at least one of the following: Determine the degree of legal liability that needs to be borne for the driving behavior that violates traffic regulations as the legal penalty risk; Determining the degree of property penalty for the driving behavior of the decision that violates the traffic regulations as the property penalty risk; and The degree of personal injury caused by the driving behavior of the decision, including the number of injured persons and casualties, is determined as the personal injury risk.

18. A machine-readable storage medium storing executable instructions, which when executed cause a processor to perform the method of claim 17.

Images