Automatic driving judgment system, automatic driving control system and vehicle

Abstract

The present disclosure provides an automatic driving judgment system, a control system and a vehicle. The judgment system includes: a data unit, which obtains environmental data items and generates attention weights to generate at least two data samples, wherein each data sample includes at least two environmental data items and corresponding attention weights; a decision unit, which generates at least two decision results for automatic driving of the vehicle based on the at least two data samples; and a judgment unit, which scores the at least two decision results based on the environmental data items, and selects the decision result with the highest score as the judgment result for automatic driving of the vehicle. The automatic driving judgment system of the present disclosure can improve the automatic driving performance of the vehicle.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and more specifically, to an autonomous driving judgment system, an autonomous driving control system and a vehicle. Background Art

[0002] Providing the vehicle with a path plan with target acceleration and speed is an important part of the vehicle's autonomous driving system, and it is also the decision-making part that links the environmental perception of the autonomous driving system with the execution mode. At present, the basis for autonomous driving vehicles to make decisions based on the perceived environment is mainly based on the comparison between traffic rules and predefined thresholds. For example, lane changes cannot be made through solid lines, and lane changes are not allowed when the rear vehicle is within a certain distance of the vehicle. Therefore, the current autonomous driving decision-making part cannot meet the needs of users well. Even if some vehicles allow users to choose between "soothing" and "aggressive" driving modes, their essence is still to adjust the threshold. Users can clearly perceive that this is machine behavior, which is obviously different from human driving behavior, and cannot make decisions similar to human drivers, resulting in a poor user experience. When the decision output of the autonomous driving system fluctuates near the threshold or suddenly fluctuates violently, the system will not be able to operate stably, which is likely to cause user discomfort or even danger.

[0003] In addition, the current autonomous driving decision-making part cannot adapt to changes in traffic rules in different regions and cannot improve itself during operation.

[0004] Therefore, how to provide a new way of making decisions on autonomous driving so that users can experience the feeling of a human driver controlling the operation of the vehicle when using the vehicle's autonomous driving mode and provide driving behaviors that are more in line with human driving habits is a technical problem that needs to be solved. Summary of the invention

[0005] An object of the present invention is to provide a new technical solution for an automatic driving judgment system.

[0006] According to a first aspect of the present invention, there is provided an automatic driving judgment system, the system comprising:

[0007] A data unit, acquiring environmental data items and generating attention weights to generate at least two data samples, wherein each data sample includes at least two environmental data items and corresponding attention weights;

[0008] A decision unit, generating at least two decision results for automatic driving of the vehicle based on the at least two data samples; and

[0009] A judgment unit is configured to score the at least two decision results based on the environmental data item, and select the decision result with the highest score as the judgment result for automatic driving of the vehicle.

[0010] Optionally, the decision result in the judgment system includes at least one autonomous driving control plan including speed, acceleration and direction of autonomous driving.

[0011] Optionally, the data unit in the judgment system obtains an environmental data group including at least two environmental data items, generates at least two groups of attention weights, and combines the environmental data group with the at least two groups of attention weights respectively to generate at least two data samples.

[0012] Optionally, the decision unit in the judgment system generates a decision result based on a first neural network, and the judgment unit performs scoring based on a second neural network.

[0013] Optionally, the judgment system further includes a setting unit and an evaluation unit.

[0014] Wherein, the setting unit can set a first training mode,

[0015] In which, in the first training mode, the parameters of the second neural network are fixed, the judgment unit generates a judgment result according to the decision result of the decision unit, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a first evaluation result, and the decision unit adjusts the parameters of the first neural network based on the first evaluation result.

[0016] Optionally, the setting unit in the judgment system can set a second training mode.

[0017] In which, in the second training mode, the parameters of the first neural network are fixed, the judgment unit generates a judgment result according to the decision result of the decision unit, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a second evaluation result, and the judgment unit adjusts the parameters of the second neural network based on the second evaluation result.

[0018] Optionally, the data unit in the judgment system includes an environmental data group of at least two environmental data items, generating at least two groups of attention weights based on a third neural network, and combining the environmental data group with the at least two groups of attention weights respectively to generate at least two data samples,

[0019] Wherein, the setting unit can set a third training mode,

[0020] In which, in the third training mode, the parameters of the first neural network and the second neural network are fixed, the judgment unit generates a judgment result according to the decision result of the decision unit, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a third evaluation result, and the data unit adjusts the parameters of the third neural network based on the third evaluation result.

[0021] According to the second aspect of the present invention, there is also provided an automatic driving control system, which comprises any automatic driving judgment system according to the first aspect of the present invention and a control unit, wherein the control unit controls the driving state of the vehicle based on the judgment result of the automatic driving judgment system.

[0022] Optionally, in the automatic driving control system, when the control unit detects that the vehicle is manually operated by a user, the control unit sets the automatic driving judgment system in a training mode through the setting unit of the automatic driving judgment system to train the corresponding units in the automatic driving judgment system, and wherein, when the control unit detects that the vehicle is in an automatic driving state, the control unit controls the driving state of the vehicle based on the judgment result of the automatic driving judgment system.

[0023] According to a third aspect of the present invention, there is also provided a vehicle, comprising any automatic driving control system described in the second aspect of the present invention.

[0024] By using the autonomous driving judgment system provided in one embodiment of the present disclosure, the performance of the autonomous driving system can be improved.

[0025] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0027] Figure 1 is a block diagram showing an example of a hardware configuration of an automatic driving system that can be used to implement an embodiment of the present invention.

[0028] Figure 2 A schematic diagram of the structure of an automatic driving judgment system provided by an embodiment of the present disclosure is shown;

[0029] Figure 3 A schematic diagram of the structure of an automatic driving control system provided by an embodiment of the present disclosure is shown;

[0030] Figure 4 A schematic diagram of the hardware structure of a vehicle provided in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0031] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless otherwise specifically stated.

[0032] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0033] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0034] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0035] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0036] <Implementation Environment and Hardware Configuration>

[0037] Figure 1 is a block diagram showing an example of a hardware configuration of an automatic driving system 100 that can be used to implement an embodiment of the present invention.

[0038] like Figure 1 As shown, the autonomous driving system 100 includes a vehicle 1000, a server 2000, and a network 3000. It should be appreciated that the overall architecture, settings, and operations, as well as the individual components of the system as shown here are well known in the art. Therefore, the following paragraphs only provide an overview of such an exemplary environment, and other systems that contain or adopt the autonomous driving system architecture shown in this implementation environment or have the relevant functions described herein can also be used as the implementation environment of the present technical solution.

[0039] The vehicle 1000 may be, for example, various types of automobiles, multi-purpose vehicles (MPVs), sport / suburban utility vehicles (SUVs), crossover utility vehicles (CUVs), recreational vehicles (RVs), autonomous vehicles (AVs), trucks, other mobile machines for transporting people or goods, etc. In many cases, the power supply mode of the vehicle 1000 may be, for example, powered by an internal combustion engine. The vehicle 1000 may also be a hybrid electric vehicle (HEV) powered by both an internal combustion engine and one or more electric motors, such as a series hybrid electric vehicle (SHEV), a parallel hybrid electric vehicle (PHEV), a parallel and series hybrid electric vehicle (PSHEV), etc. The type and power supply mode of the vehicle 1000 may be in any form, and the aforementioned examples are not limiting.

[0040] The vehicle 1000 may be provided with an electronic system, which may include, for example, a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, an output device 1500, an input device 1600, a navigation device 1700, and the like. The processor 1100 may be a microprocessor MCU, and the like. The memory 1200 may include, for example, a ROM (read-only memory), a RAM (random access memory), a non-volatile memory such as a hard disk, and the like. The interface device 1300 may include, for example, a USB interface, a headphone interface, and the like. The communication device 1400 may, for example, be capable of wired or wireless communication, and may, for example, be capable of short-distance and long-distance communication.

[0041] The output device 1500 may be, for example, a device for outputting signals, and may be a display device, such as a liquid crystal display, a touch display, a speaker, etc. The input device 1600 may include, for example, a touch screen, a button, a knob, a keyboard, a microphone, a camera, etc. When the input device 1600 is a microphone, it may provide a means for the driver or other vehicle occupants to input verbal or other auditory commands, and it may be matched with an embedded voice processing unit using human-machine interaction (HMI) technology known in the art, or it may be an independent component. The input device 1600 may also include one or more buttons or control components to enable the vehicle occupants to brake, or an electronic push button to initiate voice communication with the server 2000, etc.

[0042] The navigation device 1700, for example, includes a receiver of a global navigation satellite system GNSS (Global Navigation Satellite System), a global positioning system GPS (Global Positioning System) or a BeiDou Navigation Satellite System BDS (BeiDou Navigation Satellite System), a navigation HMI (Human Machine Interface), and a path determination unit. The navigation device 1700 stores map information in a storage device such as an HDD (Hard Disk Drive) or a flash memory. The receiver determines the position of the vehicle 1000 based on the signal received from a satellite such as a GNSS. The position of the vehicle 1000 can also be determined or supplemented by using an INS (Inertial Navigation System) output by a vehicle sensor or the like. The path determination unit, for example, refers to the map information to determine the path from the position of the vehicle 1000 determined by a receiver such as a GNSS (or an input arbitrary position) to the destination input by the passenger using the navigation HMI. The map information is, for example, information on the shape of the road represented by the lines representing the road and the nodes connected by the lines. The map information may also include the curvature of the road, POI (Point Of Interest), geographic coordinate information of each position on the road, and driving road information. The navigation device 1700 can also be implemented by the functions of terminal devices such as smart phones and tablet terminals held by passengers. The navigation device 1700 can also send the current location and destination to the navigation server via the communication device 1400, and obtain a path equivalent to the path on the map from the navigation server. The vehicle electronic system can determine the geographical location of the vehicle and the planned driving path through the navigation device, and can also upload the geographical location data to the server.

[0043] Despite Figure 11000, but the present technical solution may only use some of the devices, for example, the vehicle 1000 only involves the input device 1600 and the output device 1500. Alternatively, it may also include Figure 1 Not shown are the lighting system controlled by the processor 1100 and the sensor device for detecting the vehicle's surrounding environment.

[0044] Server 2000 provides a business point for processing, database, communication facilities, etc. Server 2000 may include a monolithic server or a distributed server across multiple computers or computer data centers. The server may be of various types, such as but not limited to a network server, a news server, a mail server, a message server, an advertising server, a file server, an application server, an interactive server, a database server, or a proxy server. In some embodiments, each server may include hardware, software, or an embedded logic component or a combination of two or more such components for executing the appropriate functions supported or implemented by the server. For example, a server such as a blade server, a cloud server, etc., or may be a server group consisting of multiple servers, which may include one or more of the above-mentioned types of servers, etc.

[0045] In one embodiment, the server 2000 may be as follows Figure 1 As shown, the server 2000 includes a processor 2100, a memory 2200, an interface device 2300, a communication device 2400, a display device 2500, and an input device 2600. In another embodiment, the server 2000 may also include a speaker, a microphone, etc., which are not limited here.

[0046] The processor 2100 may be a dedicated server processor, or a desktop processor or a mobile processor that meets the performance requirements, and is not limited here. The memory 2200 includes, for example, ROM (read-only memory), RAM (random access memory), and non-volatile memory such as a hard disk. The interface device 2300 includes, for example, various bus interfaces, such as a serial bus interface (including a USB interface), a parallel bus interface, etc. The communication device 2400 is, for example, capable of wired or wireless communication. The display device 2500 is, for example, a liquid crystal display, a touch display, etc. The input device 2600 may include, for example, a touch screen, a keyboard, etc. Although in Figure 1 2 shows multiple devices of the server 2000 , but the present invention may only involve some of the devices, for example, the server 2000 only involves the memory 2200 and the processor 2100 .

[0047] The network 3000 may include not only wireless communication networks and wired communication networks, but also generally refers to all communication methods that can be used for communication, such as optical fiber communication, ordinary microwave communication, power line carrier communication, wired audio cable communication, ultra-high frequency radio station communication, wireless spread spectrum communication, infrared, Bluetooth, radio frequency identification (RFID), keyless entry and smart key, etc. Figure 1 In the illustrated automatic driving system 100, the communication method between the vehicle 1000 and the server 2000 may be wireless communication through a network 3000, and the communication method between the vehicle 1000 and other vehicles may be wireless communication or Bluetooth. In addition, the network 3000 based on which the vehicle 1000 and the server 2000, and the vehicle 1000 and other vehicles communicate may be the same or different.

[0048] It should be understood that although Figure 1 Only one vehicle 1000, server 2000, and network 3000 are shown, but this does not mean that their respective numbers are limited. The autonomous driving system 100 may include multiple vehicles 1000, multiple servers 2000, and multiple networks 3000.

[0049] In the above description, the technician can design instructions according to the solution provided by the present disclosure. How instructions control the processor to operate is well known in the art, so it will not be described in detail here.

[0050] Figure 1 The computing system shown is merely illustrative and is in no way intended to limit the invention, its application, or uses.

[0051] <Automatic Driving Judgment System Example>

[0052] Figure 2 2 is a schematic diagram of a structure of an automatic driving judgment system according to an embodiment. The automatic driving judgment system 200 may include a data unit 210, a decision unit 220 and a judgment unit 230 which are electrically connected in sequence.

[0053] The automatic driving judgment system 200 obtains environmental data items and generates attention weights through the data unit 210 to generate at least two data samples, wherein each data sample includes at least two environmental data items and corresponding attention weights.

[0054] In this embodiment, the environmental data items acquired by the data unit 210 may refer to, for example, pedestrians, non-motor vehicles, cars, trucks, buses, emergency vehicles, lane rules, traffic lights, traffic signs, etc. The acquired environmental data items may be any item of environmental data within a certain range around the vehicle.

[0055] In this embodiment, the attention weight generated by the data unit 210 may be generated randomly, and the sum of the weights of all environmental data items is 1. The weight is used here to simulate the different attention levels of humans to different environmental data items.

[0056] In this embodiment, the environmental data items in each data sample are the same, and under the same environmental data items, corresponding attention weights are randomly generated.

[0057] For example, the vehicle's current environmental data items include pedestrians and buses, and the corresponding randomly generated attention weights are 0.6 for pedestrians and 0.4 for buses. This is a data sample. Under the same current environmental data items, in another data sample, the pedestrian weight is 0.3 and the bus weight is 0.7, that is, in the current environment, the environmental data items in the two data samples are divided into different attention weights. At this time, a minimum threshold can also be set for environmental data items that need to be divided into more attention weights. For example, the attention weight assigned to the environmental data item of the pedestrian can be set to not less than 0.3. Then, when generating the attention weight, the attention weight assigned to the pedestrian will be greater than or equal to 0.3.

[0058] The automatic driving judgment system 200 generates at least two decision results for automatic driving of the vehicle based on at least two data samples generated by the data unit 210 through the decision unit 220.

[0059] In this embodiment, the decision unit 220 generates a decision result for each data sample based on data samples having the same environmental data item and different attention weights, wherein the decision result is a control plan that can control the vehicle to perform operations such as acceleration, deceleration, lane change, etc.

[0060] In one embodiment, the decision result for vehicle autonomous driving generated by the autonomous driving judgment system 200 through the decision unit 220 may be at least one autonomous driving control plan including the speed, acceleration and direction of autonomous driving.

[0061] The automatic driving judgment system 200 scores at least two decision results generated by the decision unit 220 based on the environmental data items obtained by the data unit 210 through the judgment unit 230, and selects the decision result with the highest score as the judgment result for vehicle automatic driving.

[0062] In this embodiment, the judgment unit 230 may score each decision result by, for example, using a Sigmoid function so that the score of each decision result falls within the range of 0-1, wherein the closer the decision result score is to 1, the higher the score is, which means that the decision result is closer to the driving behavior of a human driver for the same environmental data item; the closer the decision result score is to 0, the lower the score is, which means that the decision result is more different from the driving behavior of a human driver for the same environmental data item.

[0063] In order to make the driving behavior of the vehicle in the automatic driving mode closer to human driving behavior and provide a better user experience, the judgment unit 230 selects the decision result with the highest score for execution by the vehicle automatic driving system.

[0064] In one embodiment, the autonomous driving judgment system 200 obtains an environmental data group including at least two environmental data items through the data unit 210, generates at least two sets of attention weights, and combines the environmental data group with the at least two sets of attention weights respectively to generate at least two data samples.

[0065] The environmental data group in this embodiment refers to a plurality of different environmental data items at the same time. For example, the environmental data group may have four environmental data items: pedestrians, buses, lane rules, and traffic lights. Among them, generating at least two sets of attention weights means generating two different sets of attention weights for this environmental data group. For example, in the first set of attention weights, the attention weight of pedestrians is 0.2, the attention weight of buses is 0.1, the attention weight of lane rules is 0.3, and the attention weight of traffic lights is 0.4; in the second set of attention weights, the attention weight of pedestrians is 0.3, the attention weight of buses is 0.2, the attention weight of lane rules is 0.1, and the attention weight of traffic lights is 0.4. Among them, the generation method of attention weights can be, for example, random generation, or it can be generated after setting a minimum or maximum threshold limit for a certain environmental data item. The combination of the at least two sets of attention weights and the environmental data group is used as a data sample for the decision unit 220 to generate a decision result.

[0066] In one embodiment, the decision unit 220 in the autonomous driving judgment system 200 may generate a decision result based on a first neural network, and the judgment unit 230 may perform scoring based on a second neural network.

[0067] In one embodiment, the automatic driving judgment system 200 may further include a setting unit and an evaluation unit.

[0068] The setting unit can set a first training mode. In the first training mode, the parameters of the second neural network are fixed, the judgment unit 230 generates a judgment result according to the decision result of the decision unit 220, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a first evaluation result, and the decision unit 220 adjusts the parameters of the first neural network based on the first evaluation result.

[0069] In this embodiment, the first training mode is to train the first neural network parameters of the decision unit 220. The input external result refers to the driving decision made by the conventional human driver input to the judgment unit 230 under the environmental data item or environmental data group based on which the judgment result is generated, that is, the decision made by the conventional human driver under the same environmental data.

[0070] In this embodiment, the first evaluation result refers to the degree of closeness between the judgment result generated by the judgment unit 230 and the driving decision made by a conventional human driver under the same environmental data. Based on the first evaluation result, the parameters for generating the decision result in the decision unit 220 are adjusted, and the decision unit 220 is continuously optimized so that the decision result generated by it is continuously close to the driving behavior of a human driver.

[0071] In one embodiment, the setting unit in the automatic driving judgment system 200 is also used to set a second training mode.

[0072] In the second training mode, the parameters of the first neural network are fixed, the judgment unit 230 generates a judgment result according to the decision result of the decision unit 220, the evaluation unit is used to evaluate the consistency between the judgment result and the input external result to generate a second evaluation result, and the judgment unit adjusts the parameters of the second neural network based on the second evaluation result.

[0073] In this embodiment, the second training mode is to train the parameters of the second neural network of the judgment unit 230. The input external result refers to the driving decision made by the conventional human driver input to the judgment unit 230 under the environmental data item or environmental data group based on which the judgment result is generated, that is, the decision made by the conventional human driver under the same environmental data.

[0074] In this embodiment, the second evaluation result refers to the degree of proximity between the judgment result generated by the judgment unit 230 and the driving behavior of a human driver under the same environmental data. Based on the second evaluation result, the parameters for generating the judgment result in the judgment unit 230 are adjusted, and the judgment unit 230 is continuously optimized so that it can accurately distinguish between the decision result and the human driving behavior.

[0075] In one embodiment, the data unit 210 in the automatic driving judgment system 200 may also include an environmental data group of at least two environmental data items. The data unit 210 generates at least two groups of attention weights based on the third neural network, and combines the environmental data group with the at least two groups of attention weights to generate at least two data samples.

[0076] The setting unit in the automatic driving judgment system 200 can also set a third training mode. In the third training mode, the parameters of the first neural network and the second neural network are fixed, the judgment unit 230 generates a judgment result according to the decision result of the decision unit 220, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a third evaluation result, and the data unit 210 adjusts the parameters of the third neural network based on the third evaluation result.

[0077] In this embodiment, the third training mode is to train the third neural network parameters of the data unit 210. The input external result refers to the driving decision made by the conventional human driver input to the judgment unit 230 under the environmental data item or environmental data group based on which the judgment result is generated, that is, the decision made by the conventional human driver under the same environmental data.

[0078] In this embodiment, the third evaluation result refers to the degree of closeness between the judgment result generated by the judgment unit 230 and the driving behavior of a human driver under the same environmental data. Based on the third evaluation result, the attention weight parameter generated in the data unit 210 is adjusted, and the data unit 210 is continuously optimized so that the attention weight parameters generated for different environmental data items are closer to the attention given by humans to different environmental data items.

[0079] Through the embodiments of the present disclosure, the automatic judgment system 200 can continuously train and learn human driving behavior, so that the behavior output by the vehicle during automatic driving is closer to that of humans, thereby improving the user experience of users using the automatic driving mode and increasing the user's comfort and sense of security.

[0080] <Automatic Driving Control System Example>

[0081] In this embodiment, an automatic driving control system 300 is provided. Figure 3 As shown, the control system 300 includes an automatic driving judgment system 200 provided according to any embodiment of the present invention and a control unit 310 electrically connected thereto.

[0082] Among them, the control unit 310 is used to control the driving state of the vehicle based on the judgment result of the automatic driving judgment system 200.

[0083] In one embodiment, when the control unit 310 in the automatic driving control system 300 detects that the vehicle is manually operated by the user, the control unit 310 sets the automatic driving judgment system 200 in the training mode through the setting unit in the automatic driving judgment system 200 to train the corresponding unit in the automatic driving judgment system 200; and when the control unit 310 detects that the vehicle is in the automatic driving state, the control unit 310 controls the driving state of the vehicle based on the judgment result of the automatic driving judgment system 200.

[0084] In this embodiment, the control unit 310 detects whether the vehicle is in manual operation mode or automatic driving mode. In manual operation mode, the automatic driving judgment system 200 enters the training mode under the control of its setting unit. In automatic mode, the vehicle operation is controlled according to the judgment result of the automatic driving judgment system 200.

[0085] The automatic driving judgment system 200 is a setting unit that sets the automatic driving judgment system 200 in a training mode to train the corresponding units in the automatic driving judgment system 200. For example, the data unit 210, the decision unit 220 and the judgment unit 230 in the automatic driving judgment system 200 may be trained sequentially, that is, the data sample in the data unit 210 is sent to the decision unit 220 to generate a decision result, and the judgment unit 230 then scores and determines the judgment result based on the environmental data item and the decision result, first based on the training data unit 210, then the decision unit 220 is trained in sequence, and finally the judgment unit 230. For another example, the data unit 210, the decision unit 220 and the judgment unit 230 in the automatic driving judgment system 200 may be trained simultaneously, and for another example, any two units may be selected for simultaneous training, etc. The training arrangement for the corresponding units in the automatic driving judgment system 200 may be various, which is not limited here.

[0086] By training the data unit 210, decision unit 220 and judgment unit 230 in the autonomous driving judgment system 200, the autonomous driving judgment system 200 can be continuously improved and perfected during the training process, so that the decision results are continuously approaching the driving behavior of human drivers, thereby providing users with a more comfortable vehicle operating state.

[0087] <Vehicle Embodiment>

[0088] In this embodiment, a vehicle 400 is also provided. The vehicle 400 may be Figure 1 The vehicle 1000 shown in the figure may also be a vehicle of other structures, which is not limited here.

[0089] like Figure 4 As shown, the vehicle 400 may include an automatic driving control system 300 according to any embodiment of the present invention.

[0090] Those skilled in the art will appreciate that data units, decision units, judgment units, setting units and evaluation units can be implemented in various ways. For example, data units, decision units, judgment units, setting units and evaluation units can be implemented by configuring a processor with instructions. For example, instructions can be stored in a ROM, and when the device is started, instructions are read from the ROM into a programmable device to implement data units, decision units, judgment units, setting units and evaluation units. For example, data units, decision units, judgment units, setting units and evaluation units can be solidified into a special device (such as an ASIC). Data units, decision units, judgment units, setting units and evaluation units can be divided into independent units, or they can be combined together to implement. Data units, decision units, judgment units, setting units and evaluation units can be implemented by one of the above-mentioned various implementations, or can be implemented by a combination of two or more of the above-mentioned various implementations.

[0091] The present invention may be a system, an apparatus and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0092] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0093] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0094] The computer program instructions for performing the operation of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present invention.

[0095] Various aspects of the present invention are described herein with reference to the flow charts and / or block diagrams of the methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each box of the flow chart and / or block diagram and the combination of each box in the flow chart and / or block diagram can be implemented by computer-readable program instructions.

[0096] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0097] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0098] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a part of a module, a program segment or an instruction, and a part of the module, a program segment or an instruction contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that it is equivalent to implement it by hardware, implement it by software, and implement it by combining software and hardware.

[0099] Embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the marketplace, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the present invention is defined by the appended claims.

Claims

1. An automatic driving judgment system, comprising: A data unit, acquiring environmental data items and generating attention weights to generate at least two data samples, wherein each data sample includes at least two environmental data items and corresponding attention weights, the environmental data item is any one of the environmental data within a certain range around the vehicle, the environmental data item in each data sample is the same, and the attention weight is randomly generated corresponding to the same environmental data item; A decision unit, generating at least two decision results for automatic driving of the vehicle based on the at least two data samples; and A judgment unit scores the at least two decision results based on the environmental data item, and selects the decision result with the highest score as the judgment result for the automatic driving of the vehicle, wherein the judgment unit calculates the score for each of the at least two decision results so that the score of each decision result falls within the interval of 0-1, wherein the closer the score of the decision result is to 1, the higher the score is, which means that the decision result is closer to the driving behavior made by a human driver for the same environmental data item; and the closer the score of the decision result is to 0, the lower the score is, which means that the decision result is more different from the driving behavior made by a human driver for the same environmental data item.

2. The automatic driving judgment system according to claim 1, wherein: The decision result is an autonomous driving control plan including at least one of the speed, acceleration and direction of autonomous driving.

3. The automatic driving judgment system according to claim 1, wherein: The data unit obtains an environmental data group including at least two environmental data items, generates at least two groups of attention weights, and combines the environmental data group with the at least two groups of attention weights respectively to generate at least two data samples.

4. The automatic driving judgment system according to claim 1, wherein: The decision unit generates a decision result based on the first neural network, and the judgment unit performs scoring based on the second neural network.

5. The automatic driving judgment system according to claim 4, further comprising a setting unit and an evaluation unit, in, The setting unit can set a first training mode, In which, in the first training mode, the parameters of the second neural network are fixed, the judgment unit generates a judgment result according to the decision result of the decision unit, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a first evaluation result, and the decision unit adjusts the parameters of the first neural network based on the first evaluation result.

6. The automatic driving judgment system according to claim 5, wherein: The setting unit can set the second training mode, In which, in the second training mode, the parameters of the first neural network are fixed, the judgment unit generates a judgment result according to the decision result of the decision unit, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a second evaluation result, and the judgment unit adjusts the parameters of the second neural network based on the second evaluation result.

7. The automatic driving judgment system according to claim 5 or 6, wherein: The data unit includes an environmental data group of at least two environmental data items, generating at least two groups of attention weights based on a third neural network, and combining the environmental data group with the at least two groups of attention weights respectively to generate at least two data samples, Wherein, the setting unit can set a third training mode, In which, in the third training mode, the parameters of the first neural network and the second neural network are fixed, the judgment unit generates a judgment result according to the decision result of the decision unit, the evaluation unit evaluates the consistency between the judgment result and the input external result to generate a third evaluation result, and the data unit adjusts the parameters of the third neural network based on the third evaluation result.

8. An automatic driving control system, comprising: The automatic driving judgment system according to any one of claims 1 to 7; The control unit controls the driving state of the vehicle based on the judgment result of the automatic driving judgment system.

9. The automatic driving control system according to claim 8, wherein: The control unit sets the automatic driving judgment system in a training mode through the setting unit of the automatic driving judgment system to train corresponding units in the automatic driving judgment system when detecting that the vehicle is manually operated by the user, and Among them, when the control unit detects that the vehicle is in an automatic driving state, it controls the driving state of the vehicle based on the judgment result of the automatic driving judgment system.

10. A vehicle comprising an automatic driving control system according to claim 8 or 9.

Images