Method and device for controlling acceleration of autonomous vehicle, vehicle and medium

Abstract

The application discloses a kind of automatic driving vehicle's throttle control method, device, vehicle and medium, the method includes: obtaining the environmental information around vehicle, and according to the environmental information, determine the driving environment of the vehicle from the driving environment database of the dataset created in advance;Determine the first control strategy corresponding to the driving environment in the first function logic set of the dataset;Based on the first control strategy, control the vehicle runs, and obtain the first vehicle data after the vehicle executes the first control strategy;The first vehicle data is input to the set machine learning framework for training, and the second function logic set is obtained;Determine the second control strategy corresponding to the driving environment in the second function logic set;According to the first control strategy and the second control strategy, determine the target control strategy for vehicle throttle control.The method can solve the problem of inaccurate throttle control, and ensure driving safety.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a throttle control method, device, vehicle, and medium for an autonomous vehicle. Background Technology

[0002] In related technologies, throttle control of autonomous vehicles is a very important part of autonomous driving systems, mainly involving three aspects: environmental perception, decision-making and planning, and control. In terms of decision-making and planning, deep learning-based target detection and classification algorithms, as well as machine learning-based path planning algorithms, can automatically judge road conditions, select a verified driving route, and generate corresponding throttle control commands in different driving scenarios, thereby realizing automated intelligent throttle adjustment. However, in complex traffic scenarios, it is difficult to achieve precise control of the vehicle's throttle, thus making it difficult to ensure the smoothness and safety of vehicle driving. Summary of the Invention

[0003] Therefore, it is necessary to provide a throttle control method, device, vehicle, and medium for autonomous vehicles that can precisely control the throttle of a vehicle, addressing the aforementioned technical problems.

[0004] A throttle control method for an autonomous vehicle includes the following steps:

[0005] Acquire environmental information about the vehicle's surroundings, and determine the vehicle's driving environment from a pre-created dataset's driving environment database based on that environmental information;

[0006] A first control strategy corresponding to the driving environment is determined in the first functional logic set of the dataset;

[0007] The vehicle is controlled to operate based on the first control strategy, and first vehicle data is obtained after the first control strategy is executed.

[0008] The first vehicle data is input into a set machine learning framework for training to obtain the second functional logic set;

[0009] The second control strategy corresponding to the driving environment is determined in the second functional logic set;

[0010] Based on the first control strategy and the second control strategy, a target control strategy for vehicle throttle control is determined.

[0011] A throttle control device for an autonomous vehicle, comprising:

[0012] The first determining module is used to acquire environmental information around the vehicle and determine the driving environment of the vehicle from a pre-created driving environment database based on the environmental information.

[0013] The second determining module is used to determine the first control strategy corresponding to the driving environment in the first functional logic set of the dataset;

[0014] The acquisition module is used to control the operation of the vehicle based on the first control strategy and acquire the first vehicle data after the first control strategy is executed.

[0015] The training module is used to input the first vehicle data into a set machine learning framework for training to obtain the second functional logic set.

[0016] The third determining module is used to determine the second control strategy corresponding to the driving environment in the second functional logic set;

[0017] The fourth determining module is used to determine a target control strategy for vehicle throttle control based on the first control strategy and the second control strategy.

[0018] An autonomous vehicle includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the throttle control method for the autonomous vehicle.

[0019] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the aforementioned throttle control method for an autonomous vehicle.

[0020] The aforementioned throttle control method, device, vehicle, and medium for autonomous vehicles determine the vehicle's driving environment through environmental information surrounding the vehicle. Based on the vehicle's driving environment and a first functional logic set, a corresponding first control strategy is determined. Then, the first vehicle data executing the first control strategy is learned to obtain a second functional logic set. The second functional logic set is used to determine a second control strategy. Thus, the throttle of the vehicle is controlled through the first and second control strategies. This allows the throttle control strategy to be generated effectively and safely based on the current driving conditions, improving the efficiency and accuracy of throttle control. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the system architecture for throttle control of an autonomous vehicle in one embodiment;

[0022] Figure 2 This is a flowchart illustrating the throttle control method for an autonomous vehicle in one embodiment.

[0023] Figure 3 This is a flowchart illustrating the throttle control method for an autonomous vehicle in yet another embodiment;

[0024] Figure 4This is a flowchart illustrating the throttle control method for an autonomous vehicle in yet another embodiment;

[0025] Figure 5 This is a flowchart illustrating the throttle control method for an autonomous vehicle in yet another embodiment;

[0026] Figure 6 This is a flowchart illustrating the throttle control method for an autonomous vehicle in yet another embodiment;

[0027] Figure 7 This is a flowchart illustrating the throttle control method for an autonomous vehicle in yet another embodiment;

[0028] Figure 8 This is a structural block diagram of the throttle control device for an autonomous vehicle in one embodiment. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0030] Before providing a detailed description of the embodiments of this application, a brief explanation of the throttle control strategy in the related art will be given.

[0031] The accelerator is the most basic means of controlling vehicle power, determining the amount of torque output by the engine. During manual driving, the driver can adjust the accelerator pedal opening to change the accelerator signal according to power needs. The engine controller calculates the torque demand based on the accelerator signal and other necessary input signals, then determines the fuel injection quantity based on the torque demand and engine speed, ultimately controlling the power output.

[0032] For autonomous vehicles, the autonomous driving controller sends the target vehicle speed to the vehicle controller, which then sends a signal to the vehicle power system based on the target speed, and the engine controller controls the power output.

[0033] However, throttle control in autonomous vehicles still faces several challenges. One major issue is achieving precise throttle control in complex traffic scenarios. While throttle control is relatively easy to implement on simpler roads like highways, it becomes more complex in urban environments, requiring consideration of factors such as pedestrians, traffic lights, and parking spaces. The changes and uncertainties of these factors all impact throttle control. Another challenge is achieving efficient and safe throttle control. For autonomous vehicles, throttle control needs to make decisions and generate corresponding control commands within a short timeframe to ensure smooth and safe driving. Therefore, improving the efficiency and accuracy of throttle control while ensuring safety is a pressing issue that needs to be addressed.

[0034] Based on this, the embodiments provided in this application can control the vehicle's throttle in an effective and safe manner according to the current driving conditions, thereby overcoming the shortcomings of the related technologies.

[0035] like Figure 1 As shown, the throttle control method for autonomous vehicles provided in this application can be achieved through... Figure 1 The system architecture shown is implemented. Among them, the on-board computer 101, the environment recognition module 102, and the control module 103 constitute the throttle control system.

[0036] In this system, the environmental recognition module 102 collects environmental data around the vehicle. The environmental recognition module 102 is connected to the vehicle computer 101 via the wiring harness 21. The environmental data collected by the environmental recognition module 102 can be transmitted to the vehicle computer 101 via the wiring harness 21.

[0037] The onboard computer 101 can process environmental data, confirm the current driving environment of the autonomous vehicle, and generate a first control strategy for the driving environment using the first functional logic set.

[0038] The on-board computer 101 is connected to the control module 103 via the wiring harness 22, and transmits the first control strategy to the control module 103. The control module 103 controls the throttle of the autonomous vehicle according to the first control strategy.

[0039] The onboard computer 101 uses vehicle data generated by the execution of the first control strategy by the autonomous vehicle to learn within a set machine learning framework, generating a second logic set, and then uses the second logic set to generate a second control strategy for the driving environment. The first and second control strategies are analyzed, evaluated, and compared to generate the final target control strategy.

[0040] The onboard computer 101 transmits the target control strategy to the control module via the wiring harness 22, and the control module responds to the vehicle's throttle-by-wire system.

[0041] The implementation details of the technical solutions in the embodiments of this application are described in detail below.

[0042] In one embodiment, such as Figure 2 As shown, a throttle control method for an autonomous vehicle is provided, which is applied to... Figure 1 Taking the throttle control system of an autonomous vehicle as an example, the throttle control method of the autonomous vehicle may include the following steps:

[0043] Step S201: Obtain environmental information around the vehicle, and determine the vehicle's driving environment from a pre-created driving environment database based on the environmental information.

[0044] Here, there is a certain correlation between vehicle throttle control and the driving environment. Different driving environments will affect the changes in throttle control strategies. Therefore, it is necessary to determine the driving environment in which the vehicle is located. The driving environment in which the vehicle is located can be determined by collecting environmental information around the vehicle.

[0045] In practical applications, vehicle sensors and external devices can be used to collect and sense environmental information. Common environmental information can include the following aspects:

[0046] (1) Cameras. Cameras installed on vehicles can capture images of roads and the surrounding environment. Through image processing and computer vision technology, they can identify information such as road markings, traffic signs, intersections and other vehicles.

[0047] (2) LiDAR. LiDAR can generate a three-dimensional point cloud map of the environment around a vehicle by emitting a laser beam and measuring its return time. By analyzing the point cloud data, it can detect and track surrounding vehicles, pedestrians, obstacles, etc.

[0048] (3) Radar. Radar can detect the position and speed of surrounding objects by emitting radio waves and measuring their return time. Radar on vehicles can be used to detect and track other vehicles, pedestrians, obstacles, etc.

[0049] (4) Ultrasonic sensor. Ultrasonic sensors can measure the distance and position of objects around a vehicle and are used to detect obstacles approaching the vehicle.

[0050] (5) Positioning system. The positioning system can provide the vehicle's location information. By combining it with map data, the road and location of the vehicle can be determined.

[0051] (6) Weather sensors. Weather sensors on vehicles can detect information such as latitude, humidity, air pressure, and precipitation. This information can be used to determine weather and road conditions, which has an important impact on driving decisions.

[0052] (7) Vehicle status sensors. Vehicle status sensors on a vehicle can detect information such as vehicle speed, acceleration, steering angle, and braking status. This information can be used to determine the vehicle's dynamic state and driving behavior.

[0053] By comparing and matching the acquired environmental information with a pre-created dataset of driving environments, the driving environment of the vehicle can be determined.

[0054] It should be noted that the driving environment database is a dataset pre-collected and organized based on factors such as different geographical locations, weather conditions, and road types. It contains the characteristics and patterns of various driving environments. By comparing environmental information with the driving environment database, the driving environment in which the vehicle is located can be determined. In practical applications, the relevant data in the driving environment database can consist of previously collected data and continuously expanding machine learning data, and may include information such as traffic signs, traffic lights, and lane information.

[0055] In one embodiment, Figure 3 A flowchart illustrating the throttle control method for an autonomous vehicle is shown.

[0056] Step S301: Compare the environmental information with the driving environment stored in the driving environment database to determine the corresponding probability value.

[0057] Step S302: If the probability value is greater than the second preset threshold, the corresponding driving environment is determined as the vehicle's driving environment.

[0058] In determining the driving environment of a vehicle, the collected environmental information can be compared with the features corresponding to the driving environment stored in the driving environment database. This allows us to obtain the probability value of the collected environmental information compared with each driving environment. This probability value can be used to describe the matching probability between the collected environmental information and the compared driving environment. The higher the probability value, the higher the probability that the collected environmental information belongs to the compared driving environment.

[0059] In this embodiment, the probability value can be compared with a preset threshold. When the probability value is higher than the preset threshold, it is determined that the vehicle is in the driving environment being compared, thereby completing the determination of the vehicle's driving environment.

[0060] In practical applications, when multiple probability values ​​are greater than a preset threshold, the driving environment corresponding to the highest probability value can be determined as the vehicle's driving environment.

[0061] Step S202: Determine the first control strategy corresponding to the driving environment in the first functional logic set of the dataset.

[0062] Here, the first functional logic set can be understood as the logic that issues control commands corresponding to defined driving conditions. In this embodiment, the first functional logic set uses vehicle sensor data and other vehicle data as inputs and control variables, which are then learned to generate corresponding control strategies. The first functional logic set is obtained through learning with a sufficiently large amount of data and undergoes appropriate verification.

[0063] After determining the vehicle's driving environment, the first control strategy can be determined based on the different driving environments within the first functional logic set of the dataset.

[0064] It's important to note that the primary control strategy is not a simple digital or analog control command, such as accelerating the vehicle from its current speed of 30 km / h to 60 km / h. Instead, it's a detailed control strategy, such as a throttle control strategy incorporating lateral and longitudinal components, which assists vehicle driving over a period of time. The lateral and longitudinal components represent the two main directions of vehicle movement. The lateral component refers to the vehicle's movement in the horizontal direction, such as steering and lane changing; the longitudinal component refers to the vehicle's movement in the vertical direction, such as acceleration and braking. Control strategies for these two directions can be generated based on the vehicle's driving environment.

[0065] Step S203: Control the vehicle operation based on the first control strategy and obtain the first vehicle data after the first control strategy is executed.

[0066] After determining the first control strategy, the vehicle is controlled based on this strategy. In other words, the vehicle executes the first control strategy to control the throttle. During this execution, first vehicle data is acquired. This first vehicle data reflects the execution status of the first control strategy and may include pedal position signals, target vehicle speed, target acceleration, target following distance, etc.

[0067] In one embodiment, after acquiring the first vehicle data, the first vehicle data can also be uploaded to the server, so that the server can synchronously learn the first vehicle data uploaded by all vehicles, that is, integrate and analyze the first vehicle data of multiple vehicles.

[0068] In practical applications, vehicles can also periodically download data from the server to update the vehicle's dataset. Understandably, by learning from the uploaded vehicle data, the server can provide more accurate predictions and decisions, enabling the vehicle to more accurately determine its driving environment and primary control strategy based on the updated dataset, thereby improving driving safety.

[0069] Step S204: Input the first vehicle data into the set machine learning framework for training to obtain the second functional logic set.

[0070] After acquiring the first vehicle data, the first vehicle data is input into the set machine learning framework. The set learning framework can learn any function of the throttle control system through the input first vehicle data, thereby obtaining the second functional logic set.

[0071] The machine learning framework described here is a machine learning framework specifically designed for simulating and controlling driving behavior. It can learn and predict driver behavior through machine learning algorithms and adjust the throttle control strategy of the autonomous driving system based on the prediction results.

[0072] In this embodiment, the first vehicle data is used as input. By setting up a machine learning framework to process and analyze the input first vehicle data, features related to driving behavior can be extracted. Then, machine learning algorithms, such as deep learning, decision trees, and support vector machines, are used to train and model these features to predict the driver's behavior, and a second functional logic set is generated based on the prediction results.

[0073] It should be noted that the second functional logic set here is generated by learning and predicting based on the first vehicle data of the vehicle executing the first control strategy. Compared with the first functional logic set, the second functional logic set provides more granular and personalized throttle control for the vehicle and can be adjusted according to specific driving behavior needs.

[0074] Step S205: Determine the second control strategy corresponding to the driving environment in the second functional logic set.

[0075] Here, in the second functional logic set, a corresponding second control strategy can be determined based on the different driving environments of the vehicle. Similarly, the second control strategy here is not a simple digital or analog control command, but a detailed control strategy, such as a vehicle throttle control strategy that includes lateral and longitudinal components.

[0076] It should be noted that the second control strategy here is a control strategy obtained based on the prediction results of the set machine learning framework, while the first control strategy is a control strategy obtained based on rules or empirical knowledge.

[0077] In practical applications, the second functional logic set is generated by learning using a set machine learning framework. The obtained second functional logic is very reliable, and the second control strategy determined based on the second functional logic set is also very reliable. Thus, a reliable control strategy can be obtained by setting a machine learning framework based on the collected and parsed sensor data or raw vehicle driving state data.

[0078] Step S206: Determine the target control strategy for vehicle throttle control based on the first control strategy and the second control strategy.

[0079] Here, the first control strategy and the second control strategy are compared or combined to determine a target control strategy for vehicle throttle control. The determined target strategy can be transmitted to the vehicle's control module, which will then respond to the vehicle's throttle-by-wire system (including the accelerator pedal and displacement sensor, electronic control unit and data bus, electric motor and throttle valve, etc.).

[0080] In practical applications, the second control strategy can be understood as a more refined control of the throttle based on the vehicle's execution of the first control strategy. This makes the throttle control strategy more in line with the vehicle's driving scenario, significantly increasing the safety level of throttle control compared to relying solely on a single control strategy.

[0081] In this embodiment, two different functional control logics can be easily obtained, both of which are used to control the vehicle throttle. Thus, two control strategies corresponding to the current driving environment can be obtained based on these two different functional control logics. In this way, safety standards that meet the Automotive Safety Integrity Level (ASIL) can be achieved, as well as safety standards that meet the expected functions.

[0082] In one embodiment, Figure 4 A flowchart illustrating the throttle control method for an autonomous vehicle is shown.

[0083] Step S401: Compare the first control strategy with the second control strategy to determine the first evaluation score.

[0084] Here, the first control strategy is compared with the second control strategy to obtain a first evaluation score for the second control strategy. This first evaluation score is mainly used to describe the similarity between the first control strategy and the second control strategy, and to determine whether there are significant differences between the first control strategy and the second control strategy.

[0085] Step S402: If the first evaluation score is greater than the first preset threshold, combine the first control strategy and the second control strategy to obtain the target control strategy.

[0086] If the first evaluation score is greater than the first preset threshold, it indicates that the first control strategy and the second control strategy are similar. Since the second control strategy is obtained by using vehicle data generated during the execution of the first control strategy, the second control strategy can better determine the execution effect of the first control strategy and the complexity of the driving environment. In other words, the second control strategy is an optimization of the first control strategy being executed. Based on this, the first control strategy and the second control strategy are combined to obtain the target control strategy. By controlling the vehicle throttle through the target control strategy, a better control effect can be achieved.

[0087] In this embodiment, the combination of the first control strategy and the second control strategy can be selected according to specific circumstances and requirements. The following are some possible combinations:

[0088] (1) Fusion strategy. The first control strategy and the second control strategy are fused together to make full use of the advantages of the two control strategies. In practical applications, the outputs of the two control strategies can be combined by weighted averaging, logical operation or other methods to obtain the target control strategy.

[0089] (2) Hierarchical Strategy. The first control strategy is designated as the primary strategy, and the second control strategy as the secondary strategy. The primary strategy handles basic control tasks, while the secondary strategy provides additional support and adjustments as needed. The relationship between the primary and secondary control strategies can be managed through priority, weight, or other methods.

[0090] In practical applications, this is equivalent to using the first evaluation score to assess the effectiveness of the first control strategy, thereby fine-tuning the first control strategy executed by the vehicle based on the vehicle's driving environment, and thus controlling the vehicle's throttle more accurately and safely.

[0091] The following provides a specific implementation method for calculating the first assessment score.

[0092] Suppose we have a comparison matrix Y, which is used to evaluate control strategies. The key parameters and thresholds of the comparison matrix Y can be preset.

[0093] Suppose the comparison matrix Y has n parameters, and each parameter has n thresholds:

[0094]

[0095] Assuming each threshold score has a weight of w, then w is:

[0096]

[0097] Thus, the comparison factor can be obtained.

[0098]

[0099] Among them, w k It refers to the k-th weight, which satisfies w k ≥0, k=1,2,…,n1*n2. Comparison factors It represents the minimum deviation in the evaluation of the control strategy.

[0100] Assume the control parameters included in the first control strategy are:

[0101]

[0102] Assume the control parameters included in the second control strategy are:

[0103]

[0104] By comparing the similarity between the first control strategy and the second control strategy, a first evaluation score can be obtained.

[0105] vec(T)=vec(t 1 )-vec(t 2 )

[0106]

[0107]

[0108] when The first control strategy and the second control strategy can then be combined to obtain the target control strategy, wherein the second control strategy is an optimization of the first control strategy being executed by the vehicle.

[0109] In one embodiment, Figure 5 This is a flowchart illustrating the throttle control method for autonomous vehicles.

[0110] Step S501: Input the second vehicle data into the set machine learning framework for training to obtain the third functional logic set.

[0111] Step S502: Determine the third control strategy corresponding to the driving environment in the third functional logic set.

[0112] Step S503: Determine the target control strategy based on the first control strategy, the second control strategy, and the third control strategy.

[0113] In this embodiment, if the first evaluation score is less than or equal to a first preset threshold, it indicates a significant deviation between the first control strategy and the second control strategy, thus requiring further learning using second vehicle data. The second vehicle data represents data related to the second functional logic set, including data obtained during the generation of the second control strategy based on the second functional logic set, and data obtained during the comparison between the second and first control strategies.

[0114] The second vehicle data is input into a set machine learning framework, which then learns from the second vehicle data to generate a third functional logic set.

[0115] After obtaining the third functional logic set using machine learning, the third control strategy corresponding to the vehicle's driving conditions is determined using this set. This third control strategy is not a simple digital or analog control command, but a detailed control strategy, such as a vehicle throttle control strategy that includes lateral and longitudinal components.

[0116] After determining the third control strategy, the first, second, and third control strategies are analyzed, evaluated, and compared to ultimately generate the target control strategy. In this embodiment, the vehicle's throttle control safety is further improved by providing three sets of functional logic and three control strategies based on the same driving conditions.

[0117] In one embodiment, Figure 6 This is a flowchart illustrating the throttle control method for an automated vehicle.

[0118] Step S601: Compare the first control strategy, the second control strategy, and the third control strategy to determine the second evaluation score for the three items.

[0119] Step S602: If the second evaluation scores of at least two items are greater than the first preset threshold, combine the first control strategy, the second control strategy and the third control strategy to obtain the target control strategy.

[0120] Here, by comparing and analyzing the first, second, and third control strategies, we can obtain the second evaluation scores for all three. and in,

[0121] Second assessment score The second evaluation score is obtained by comparing and analyzing the first and third control strategies. It can be used to describe the similarity between the first control strategy and the third control strategy.

[0122] Second assessment score The second evaluation score is obtained by comparing and analyzing the second and third control strategies. It can be used to describe the similarity between the second and third control strategies.

[0123] Second assessment score The second evaluation score is obtained by comparing and analyzing the first, second, and third control strategies. It can be used to describe the similarity between the first control strategy, the second control strategy, and the third control strategy.

[0124] Of course, the specific calculation method for the second assessment score can adopt the specific calculation method for the first assessment score mentioned above.

[0125] In determining the second assessment score and Then, the second assessment score is specified. and If at least two of the parameters are greater than the first set threshold, the first control strategy, the second control strategy, and the third control strategy can be combined to obtain the target control strategy.

[0126] The combination of the first control strategy, the second control strategy, and the third control strategy can also refer to the combination of the first control strategy and the second control strategy.

[0127] In one embodiment, Figure 7 This is a flowchart illustrating the throttle control method for an automated vehicle.

[0128] Step S701: Input the third vehicle data into the set machine learning framework for training to obtain the fourth functional logic set.

[0129] Step S702: Determine the fourth control strategy corresponding to the driving environment in the fourth functional logic set, and determine the fourth control strategy as the target control strategy.

[0130] In this embodiment, when a third control strategy is introduced, if at least two of the three second evaluation scores are not greater than the first preset threshold, then a machine learning framework is used to learn the third vehicle data to output a fourth functional logic set. The third vehicle data is data related to the third functional logic set, including data obtained in the process of generating the third control strategy based on the third functional logic set, and data obtained in the process of comparing the first control strategy, the second control strategy, and the third control strategy.

[0131] After outputting the fourth functional logic set through a machine learning framework, the fourth control strategy corresponding to the driving environment is determined using this fourth functional logic set. This fourth control strategy serves as the target control strategy. In practical applications, the fourth control strategy enables the vehicle to operate in a safe state; that is, the fourth control strategy is a safe form of driving control strategy.

[0132] In the above embodiments, a first control strategy corresponding to the driving environment is determined using a first functional logic set, and the vehicle data generated by executing the first control strategy is learned to obtain a second functional logic set. A second control strategy corresponding to the driving environment is determined using the second functional logic set. The target control strategy is determined by using the first control strategy and the second control strategy. Compared with a single control command, the safety level can be significantly increased, and the throttle of the vehicle can be controlled more accurately and safely.

[0133] In one embodiment, a throttle control device for an autonomous vehicle is provided, with reference to... Figure 8 As shown, the throttle control device 800 of the autonomous vehicle may include: a first determining module 801, a second determining module 802, an acquisition module 803, a training module 804, a third determining module 805, and a fourth determining module 806.

[0134] The system comprises the following modules: a first determining module 801, which acquires environmental information surrounding the vehicle and determines the vehicle's driving environment from a pre-created driving environment database based on the environmental information; a second determining module 802, which determines a first control strategy corresponding to the driving environment from a first functional logic set in the dataset; an acquiring module 803, which controls the vehicle's operation based on the first control strategy and acquires first vehicle data after the first control strategy is executed; a training module 804, which inputs the first vehicle data into a set machine learning framework for training to obtain a second functional logic set; a third determining module 805, which determines a second control strategy corresponding to the driving environment from the second functional logic set; and a fourth determining module 806, which determines a target control strategy for the vehicle's throttle control based on the first control strategy and the second control strategy.

[0135] In one embodiment, the fourth determining module 806 is specifically used to compare the first control strategy with the second control strategy to determine a first evaluation score; the first evaluation score characterizes the similarity between the first control strategy and the second control strategy; and when the first evaluation score is greater than a first preset threshold, combine the first control strategy and the second control strategy to obtain the target control strategy.

[0136] In one embodiment, when the first evaluation score is less than or equal to a first preset threshold, the fourth determining module 806 is specifically configured to: input the second vehicle data into the set machine learning framework for training to obtain a third functional logic set; the second vehicle data represents data related to the second functional logic set; determine a third control strategy corresponding to the driving environment in the third functional logic set; and determine the target control strategy based on the first control strategy, the second control strategy, and the third control strategy.

[0137] In one embodiment, the fourth determining module 806 is specifically used to compare the first control strategy, the second control strategy, and the third control strategy to determine a second evaluation score for each of the three items; the second evaluation scores for the three items respectively include the similarity between the first control strategy and the third control strategy, the similarity between the second control strategy and the third control strategy, and the similarity between the first control strategy, the second control strategy, and the third control strategy; if the second evaluation scores of at least two items are greater than the first preset threshold, the first control strategy, the second control strategy, and the third control strategy are combined to obtain the target control strategy.

[0138] In one embodiment, if at least two of the second evaluation scores are less than or equal to the first preset threshold, the fourth determining module 806 is specifically configured to: input the third vehicle data into the set machine learning framework for training to obtain a fourth functional logic set; the third vehicle data represents data related to the third functional logic set; determine the fourth control strategy corresponding to the driving environment in the fourth functional logic set, and determine the fourth control strategy as the target control strategy; the fourth control strategy is used to control the vehicle to operate in a safe state.

[0139] In one embodiment, the first determining module 801 is specifically used to compare the environmental information with the driving environment stored in the driving environment database to determine the corresponding probability value; the probability value represents the matching probability between the environmental information and the corresponding driving environment; and when the probability value is greater than a second preset threshold, the corresponding driving environment is determined as the driving environment of the vehicle.

[0140] In one embodiment, the acquisition module 803 is further configured to upload the first vehicle data to a cloud server to learn from vehicle data uploaded by other vehicles.

[0141] Specific limitations regarding the throttle control device for autonomous vehicles can be found in the limitations on the throttle control method for autonomous vehicles described above, and will not be repeated here. Each module in the aforementioned throttle control device for autonomous vehicles can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0142] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0143] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0144] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0145] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0146] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A throttle control method for an autonomous vehicle, characterized in that, include: Acquire environmental information about the vehicle's surroundings, and determine the vehicle's driving environment from a pre-created dataset's driving environment database based on that environmental information; A first control strategy corresponding to the driving environment is determined in the first functional logic set of the dataset; The vehicle is controlled to operate based on the first control strategy, and first vehicle data is obtained after the first control strategy is executed. The first vehicle data is input into a set machine learning framework for training to obtain the second functional logic set; The second control strategy corresponding to the driving environment is determined in the second functional logic set; Based on the first control strategy and the second control strategy, a target control strategy for vehicle throttle control is determined, including: The first control strategy is compared with the second control strategy to determine a first evaluation score; the first evaluation score represents the similarity between the first control strategy and the second control strategy. If the first evaluation score is greater than the first preset threshold, the first control strategy and the second control strategy are combined to obtain the target control strategy. When the first evaluation score is less than or equal to a first preset threshold, the following applies: The second vehicle data is input into the set machine learning framework for training to obtain the third functional logic set; the second vehicle data represents data related to the second functional logic set, including data obtained in the process of generating the second control strategy based on the second functional logic set, and data obtained in the process of comparing the second control strategy and the first control strategy; The third control strategy corresponding to the driving environment is determined in the third functional logic set; Determining the target control strategy based on the first control strategy, the second control strategy, and the third control strategy includes: The first control strategy, the second control strategy, and the third control strategy are compared to determine the second evaluation score of the three items; the second evaluation score of the three items includes the similarity between the first control strategy and the third control strategy, the similarity between the second control strategy and the third control strategy, and the similarity between the first control strategy, the second control strategy, and the third control strategy. If the second evaluation score of at least two items is greater than the first preset threshold, the first control strategy, the second control strategy and the third control strategy are combined to obtain the target control strategy; If at least two of the second evaluation scores are less than or equal to the first preset threshold, the third vehicle data is input into the set machine learning framework for training to obtain a fourth functional logic set; the third vehicle data represents data related to the third functional logic set; a fourth control strategy corresponding to the driving environment is determined in the fourth functional logic set, and the fourth control strategy is determined as the target control strategy; the fourth control strategy is used to control the vehicle to operate in a safe state.

2. The method according to claim 1, characterized in that, Determining the vehicle's driving environment from a stored driving environment database based on the environmental information includes: The environmental information is compared with the driving environment stored in the driving environment database to determine the corresponding probability value; the probability value represents the matching probability between the environmental information and the corresponding driving environment. If the probability value is greater than the second preset threshold, the corresponding driving environment is determined as the driving environment of the vehicle.

3. The method according to claim 1, characterized in that, The method further includes: The first vehicle data is uploaded to a cloud server to learn from vehicle data uploaded by other vehicles.

4. A throttle control device for an autonomous vehicle, characterized in that, include: The first determining module is used to acquire environmental information around the vehicle and determine the driving environment of the vehicle from a pre-created driving environment database based on the environmental information. The second determining module is used to determine the first control strategy corresponding to the driving environment in the first functional logic set of the dataset; The acquisition module is used to control the operation of the vehicle based on the first control strategy and acquire the first vehicle data after the first control strategy is executed. The training module is used to input the first vehicle data into a set machine learning framework for training to obtain the second functional logic set. The third determining module is used to determine the second control strategy corresponding to the driving environment in the second functional logic set; The fourth determining module is used to determine a target control strategy for vehicle throttle control based on the first control strategy and the second control strategy, including: The first control strategy is compared with the second control strategy to determine a first evaluation score; the first evaluation score represents the similarity between the first control strategy and the second control strategy. If the first evaluation score is greater than the first preset threshold, the first control strategy and the second control strategy are combined to obtain the target control strategy. When the first evaluation score is less than or equal to a first preset threshold, the following applies: The second vehicle data is input into the set machine learning framework for training to obtain the third functional logic set; the second vehicle data represents data related to the second functional logic set, including data obtained in the process of generating the second control strategy based on the second functional logic set, and data obtained in the process of comparing the second control strategy and the first control strategy; The third control strategy corresponding to the driving environment is determined in the third functional logic set; Determining the target control strategy based on the first control strategy, the second control strategy, and the third control strategy includes: The first control strategy, the second control strategy, and the third control strategy are compared to determine the second evaluation score of the three items; the second evaluation score of the three items includes the similarity between the first control strategy and the third control strategy, the similarity between the second control strategy and the third control strategy, and the similarity between the first control strategy, the second control strategy, and the third control strategy. If the second evaluation score of at least two items is greater than the first preset threshold, the first control strategy, the second control strategy and the third control strategy are combined to obtain the target control strategy; If at least two of the second evaluation scores are less than or equal to the first preset threshold, the third vehicle data is input into the set machine learning framework for training to obtain a fourth functional logic set; the third vehicle data represents data related to the third functional logic set; a fourth control strategy corresponding to the driving environment is determined in the fourth functional logic set, and the fourth control strategy is determined as the target control strategy; the fourth control strategy is used to control the vehicle to operate in a safe state.

5. An autonomous vehicle, comprising a memory and a processor, the memory having a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images