Vehicle driving assistance method and driving assistance system
A hybrid driver model in advanced driving support systems switches between rule-based and probabilistic models to ensure safety and comfort, overcoming the limitations of conventional systems by providing predictable and human-like vehicle control.
Patent Information
- Application Number
- JP2022110243
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-07-08
- Publication Date
- 2024-12-09
- Estimated Expiration
- 2042-07-08
AI Technical Summary
Conventional advanced driving support systems face challenges in achieving both passenger comfort and safety due to the uncertainties of probability-based driver models and the difficulty in training models for all driving conditions.
A hybrid driver model combining rule-based and probabilistic models, automatically switching between them to ensure safety and comfort by using a rule-based model for deterministic control when safety is required and a probabilistic model for human-like driving.
The hybrid model achieves both safety and comfort by ensuring safe, predictable vehicle control while mimicking human driving behavior, addressing the limitations of standalone rule-based and probabilistic models.
Smart Images

Figure 0007596334000001 
Figure 0007596334000002 
Figure 0007596334000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a driving assistance method and a driving assistance system for controlling a vehicle by an on-board computer. [Background technology]
[0002] In recent years, vehicles such as automobiles are known that are equipped with systems that use computers to automatically or semi-automatically steer the vehicle. For example, automobiles are known that are equipped with advanced driver-assistance systems (ADAS) that use various sensors to grasp information about the vehicle's surroundings and control the vehicle based on this information to assist in driving. Such advanced driver-assistance systems are equipped with a driver model that defines parameters for generating control operations of the vehicle. The driver model is software that acquires information necessary for driving the vehicle and generates vehicle control operations such as steering and acceleration based on the acquired information.
[0003] In a typical advanced driver assistance system, rules for controlling the vehicle are set in advance, and the control action is determined by comparing the rules with the state of the vehicle (a driver model that generates such control actions is called a "rule-based driver model"). However, human driving is known to be probabilistic, and it is difficult to imitate human driving using pre-set rules. Therefore, the vehicle control action in a rule-based driver model may be far removed from human driving. If control is performed differently from human driving, passengers may lose comfort, or the control may be performed beyond human prediction, causing a sense of danger.
[0004] As a method to solve these problems, technology is being researched that collects data from when a vehicle is actually driven and uses machine learning technology to generate a driver model (such a driver model is called a "probability-based driver model"). By using such a probability-based driver model, the driver model is improved so that it approaches the way a driver actually drives, which may make it possible to achieve behavior that is closer to human driving.
[0005] For example, Patent Document 1 discloses an invention in which a set (combination) of parameters of a formulated mathematical formula is stored as a driver model, and driving characteristics are switched by reflecting one of the stored driver models when autonomous driving is performed. The invention described in Patent Document 1 is said to enable a driver to select a driver model that is close to his or her own driving characteristics by switching between driver models corresponding to each driving characteristic. [Prior art documents] [Patent documents]
[0006] [Patent Document 1] JP 2017-61168 A Summary of the Invention [Problem to be solved by the invention]
[0007] However, the above-mentioned probability-based driver model also has drawbacks. That is, the flip side of being probability-based is that the control content is uncertain, making it difficult to guarantee safe behavior. In addition, driving content is very heterogeneous and has a huge number of variables, making it very difficult to train a model that covers all driving conditions.
[0008] Therefore, an object of the present invention is to provide a driving assistance method and a driving assistance system for a vehicle that can achieve both comfort and safety, which have been difficult to achieve with conventional advanced driving assistance systems. [Means for solving the problem]
[0009] In order to solve the above-mentioned problems, the present invention provides a method in which a computer installed in a vehicle controls the vehicle using a driver model that defines parameters for generating control operations of the vehicle, and includes the steps of selecting one driver model from a plurality of driver models while driving, and controlling the vehicle using the selected driver model, and the driver models include at least a rule-based model for generating control operations based on predetermined conditions without using a probability function, and a probabilistic model for generating control operations using a probability function based on information collected from the driver operating the vehicle. Effect of the Invention
[0010] As described above, the present invention includes, as driver models, at least a rule-based model for generating a control action based on a predetermined condition without using a probability function, and a probability model for generating a control action using a probability function based on information collected from a driver who operates a vehicle. Then, one driver model is selected from the multiple driver models during driving, and the vehicle is controlled using the selected driver model.
[0011] With this configuration, the rule-based model can be used in situations where safety is required, and the probabilistic model can be used in other situations. By automatically switching between driver models in this way, it is possible to achieve both comfort and safety, which was difficult to achieve with conventional automated driving. [Brief description of the drawings]
[0012] [Figure 1] FIG. 1 is a diagram illustrating an overview of a driving assistance system. [Diagram 2] FIG. 11 is a diagram illustrating the timing for switching a driver model. [Diagram 3]1 is a block diagram showing a schematic configuration of a driving assistance system. [Figure 4] 4 is a flowchart illustrating an example of an automatic driving process. [Diagram 5] 10 is a flowchart illustrating an example of a driver model selection process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0013] An embodiment of the present invention will be described with reference to the drawings. In the following description, an automobile vehicle 10 will be taken as an example of a vehicle. However, the driving assistance method and driving assistance system according to the present invention are not limited to vehicles 10 such as automobiles, and can also be applied to other vehicles (for example, ships, airplanes, etc.).
[0014] The driving assistance system according to the present embodiment grasps information about the surroundings of the vehicle 10 using various sensors and controls the vehicle 10 based on the information to assist driving, and is a so-called Advanced Driver-Assistance System (ADAS). This driving assistance system includes a driver model that defines parameters for generating control operations for the vehicle 10. The driver model is software that acquires information required for driving the vehicle 10 (information about the surroundings of the vehicle 10, etc.) and generates control operations for the vehicle 10, such as steering and acceleration, based on the acquired information. The control operations for the vehicle 10 generated by the driver model are, for example, a list of output signals, and control signals are issued from the computer 12 of the control device 11 to various devices (such as the steering wheel 31, the accelerator 32, and the brake 33) according to this list. As a result, the angle of the steering wheel 31, the pedal pressure of the accelerator 32, the pedal pressure of the brake 33, etc. are controlled by the computer 12.
[0015] The driving assistance system according to this embodiment employs a hybrid driver model as shown in Fig. 1. The hybrid driver model is a computing model that combines a plurality of driver models, and the plurality of driver models are automatically switched and used during driving. Specifically, the driving assistance system selects an optimal driver model according to the vehicle situation, and the control operation of the vehicle 10 is generated according to the selected driver model.
[0016] The hybrid driver model according to this embodiment includes at least two driver models: a rule-based model and a probabilistic model. Note that the number of driver models included in the hybrid driver model does not necessarily have to be two, and the hybrid driver model may include three or more driver models.
[0017] First, the rule-based model is a rule-based driver model that generates a control action based on a predetermined condition without using a probability function (in other words, without using a random variable). This rule-based model can use a driver model used in a general advanced driver assistance system. In the rule-based model, rules used for controlling the vehicle are set in advance by a designer (e.g., an automobile manufacturer), and the computer 12 compares the rule with the state of the vehicle 10 to determine the control action. For example, by setting a speed limit as a rule in advance, it is possible to generate a control action to decelerate the vehicle 10 when the speed of the vehicle 10 reaches the speed limit. Also, by setting a vehicle distance as a rule in advance, it is possible to generate a control to decelerate the vehicle 10 when the vehicle distance becomes equal to or less than a predetermined value. In this way, the rule-based model takes a deterministic approach, and the action of the vehicle 10 is determined based on a predetermined explicit rule.
[0018] The advantage of this rule-based model is that it can determine the control content based on predetermined conditions, so it can deterministically deal with predicted events. Since it is determined in advance how the vehicle 10 will be controlled to operate when the vehicle 10 is in a certain situation, it is possible to generate deterministic operations that ensure safe operation, and thus improve the safety of the vehicle 10.
[0019] On the other hand, this rule-based model has the disadvantage that the control operation is uniform and not human. Human driving is known to be probabilistic, and it is difficult to imitate human driving using pre-set rules. Therefore, the control operation of the vehicle 10 in the rule-based driver model may be far removed from human driving. If control is performed differently from human driving, comfort may be lost, or control beyond human prediction may be performed, which may cause a sense of danger.
[0020] On the other hand, the probabilistic model is a probability-based driver model that depends on information (training data) collected from the driver who operates the vehicle 10. In the probabilistic model, the driver's operation of the vehicle 10 is recorded as training data and reflected in the control content. Specifically, in the probabilistic model, information when the driver actually drives the vehicle (such as the timing when the accelerator 32 and brake 33 are actually operated) is collected, and the driver model is updated based on this information using machine learning technology so as to approach the actual operation content. For example, the information collected from the driver is formulated as a formula or a numerical value, and these formulas and numerical values are applied to the probabilistic model.
[0021] These formulas and numerical values include probability functions and random variables. For example, a probability function may be created (updated) based on information collected from the driver, and the control operation of the vehicle 10 may be generated using this probability function. Also, random variables for expressing the individuality of the driver may be prepared in advance in the probability model, and the random variables may be selected based on information collected from the driver to define the steering profile of the driver. Note that such designs of the probability functions and random variables are merely examples. The probability model may be one that probabilistically determines the control operation based on information collected from the driver, and the probability functions and random variables may be freely designed. Thus, the probabilistic model includes a probability function based on the training data, which is used to generate a control action.
[0022] The advantage of this probabilistic model is that it can reproduce behavior close to the driver's actual operation because the control content is determined based on information collected from the driver. This improves comfort. Another advantage is that the driver can easily predict the behavior of the automatic control because the control that is close to the driver's own operation is selected.
[0023] However, this probabilistic model also has disadvantages. First, driving behavior is highly heterogeneous and has a huge number of variables, so it is very difficult to train a model that covers all driving conditions. Second, the control content is uncertain, so it is difficult to guarantee safe behavior.
[0024] As described above, the rule-based model and the probabilistic model each have their own advantages and disadvantages, and complement each other. For this reason, the driving assistance system according to the present embodiment employs a hybrid driver model that combines the rule-based model and the probabilistic model, and is configured to utilize the advantages of both models by automatically switching between these driver models while driving.
[0025] Specifically, the switching control as shown in FIG. 2 is executed. First, as shown in A of FIG. 2, when comfort is prioritized, a probabilistic model is used to realize driving close to human operation. If a situation occurs that deviates from a safety boundary while driving using the probabilistic model in this way (for example, if an event such as the speed becoming too fast, the distance between vehicles becoming too close, or an obstacle being found in the driving direction is detected by various sensors, etc.), the driving assistance system switches the driver model to the rule-based model as shown in B and drives. As a result, when the safety boundary is deviated, safe driving is realized by the rule-based model. After that, when it is determined that the safety of the vehicle 10 is ensured, the driving assistance system switches the driver model to the probabilistic model as shown in C and drives. By switching the driver model in this way, it is possible to maintain the vehicle 10 in a safe state while providing the most comfortable driving possible.
[0026] Fig. 3 is a block diagram illustrating the general configuration of the driving assistance system. As shown in Fig. 3, the driving assistance system according to this embodiment is configured mainly with a control device 11 mounted on a vehicle 10. Although not shown in particular, this control device 11 is configured mainly with a computer 12 and includes a storage device 20. The computer 12 is configured to control various steering devices 30 and an environment acquisition device 40 by reading the programs and data stored in the storage device 20.
[0027] The storage device 20 may be of any type accessible to the computer 12, such as a hard disk drive, a solid state drive, a memory card, a ROM, a RAM, a computer readable medium, or a combination of these.
[0028] 3, the storage device 20 stores rule-based model data 21, probability model data 22, a state determination program 23, a flight control program 24, etc. It goes without saying that the storage device 20 also stores various other data and programs in addition to those described above.
[0029] The rule-based model data 21 is data for realizing a rule-based model. This rule-based model data 21 includes rule data 21a. The rule data 21a is a formula or a numerical value for generating a control action according to the state of the vehicle 10. An input parameter from an environment acquisition device 40, which will be described later, is converted into a control action based on this rule data 21a. This rule data 21a also includes data on a safety boundary, and is used when a state determination program 23, which will be described later, determines a safe state. This rule data 21a is defined in advance by a designer of the vehicle 10 (such as an automobile manufacturer), and is not changed autonomously.
[0030] The probability model data 22 is data for realizing a probability model. The probability model data 22 according to the present embodiment includes profile data 22a and a probability function 22b. The profile data 22a is a formula of information collected from the driver who drives the vehicle 10 as a numerical value. The probability function 22b is a formula of information collected from the driver who drives the vehicle 10 as a mathematical expression. The profile data 22a and the probability function 22b are generated based on, for example, the operation timing and wheel angle of the steering wheel 31, the operation timing and pedal pressure of the accelerator 32 and the brake 33, and the like, which are recorded. Therefore, as the driver drives the vehicle 10, the profile data 22a and the probability function 22b are updated from time to time, so that control closer to the driver's operation can be realized. The profile data 22a may be defined in a form of selecting a random variable prepared in advance in the probability model.
[0031] The state determination program 23 is a program for determining whether or not the vehicle 10 is in a safe state. This state determination program 23 uses data on safety boundaries included in the rule data 21a to determine the state of the vehicle 10. This determination of the safe state is executed at a cycle short enough to ensure the safety of the vehicle 10, and is executed every 50 milliseconds in this embodiment. By executing the determination of the safe state at a constant cycle in this manner, it is possible to constantly grasp the state of the vehicle 10.
[0032] In addition, in the data regarding the safety boundary, a number of conditions are set in advance for the state in which the vehicle 10 is not safe. For example, the conditions are "the speed change rate is equal to or greater than a predetermined value (possibility of sudden vehicle breakdown)", "the time until the collision with the preceding vehicle is less than a predetermined value", "the distance to the preceding vehicle is less than a predetermined value", "the vehicle is outside the driving lane", etc. The state determination program 23 checks whether the current state of the vehicle 10 corresponds to any of these conditions, and determines whether the vehicle 10 is in a safe state or not.
[0033] However, even if the state determination program 23 determines that the vehicle 10 is "unsafe," this does not necessarily mean that the vehicle 10 is in a dangerous state. That is, the determination by the state determination program 23 is made solely for the purpose of ensuring safety, and there may be cases where a strict safety boundary is set in order to prevent accidents from occurring. For example, when a rare event occurs in which the behavior is unstable due to insufficient training data for the probabilistic model, the vehicle may be determined to be "unsafe."
[0034] The result of the determination by the state determination program 23 is used to select a driver model. In this embodiment, if the state determination program 23 determines that the vehicle 10 is "unsafe" while driving using the probabilistic model, a process of switching the driver model to the rule-based model is executed. Also, if the state determination program 23 determines that the vehicle 10 is in a safe state for a predetermined period (e.g., 3 seconds) while driving using the rule-based model, a process of switching the driver model to the probabilistic model is executed.
[0035] The driving control program 24 is a program that uses the driver model described above to control the vehicle 10. By continuously using sensor data, the driving control program 24 can identify traffic signs, people, other vehicles, obstacles, driving lanes, etc., and autonomously control the vehicle 10. In addition, the driving control program 24 can use map data and position data such as GPS to assist driving to a destination.
[0036] This steering control program 24 acquires data on the state of the vehicle 10 from various devices described below, generates a control operation for the steering device 30 based on the acquired data, and outputs a control signal for the steering device 30 based on the control operation. In addition, this steering control program 24 selects a driver model to be used in accordance with the determination result by the state determination program 23. When the steering control program 24 generates a control operation, either the rule-based model data 21 or the probabilistic model data 22 is used in accordance with the selected driver model. For this reason, the generated control operation may differ depending on which driver model is selected.
[0037] The control device 11 described above receives input signals from, for example, the steering device 30 and the environment acquisition device 40 as shown in Fig. 3, and thereby grasps the current state of the vehicle 10. Then, using the current driver model, the control device 11 generates a control operation suitable for the current state of the vehicle 10. In accordance with this control operation, the control device 11 outputs a control signal to the steering device 30 to assist in driving the vehicle 10.
[0038] The steering device 30 is a group of devices for steering the vehicle 10. As shown in FIG. 3, the steering device 30 includes a steering wheel 31 for controlling the direction of travel of the vehicle, an accelerator 32 for adjusting the speed of the vehicle, a brake 33 for decelerating or stopping the vehicle, and a transmission 34 for changing the rotation speed and torque. The output of these steering devices 30 can be controlled by a control signal from the control device 11, and can also be controlled by operation by the driver. For example, the angle of the steering wheel 31 can be controlled by the computer 12, or by the driver operating the steering wheel. The accelerator 32 and the brake 33 can also be controlled by the computer 12, or by the driver operating pedals, etc. The steering device 30 is not limited to these, and other steering devices 30 may be included.
[0039] The environment acquisition device 40 is a group of devices for acquiring the state of the vehicle 10, and includes various sensors, cameras, and the like. The environment acquisition device 40 enables the control device 11 to grasp the traveling state of the vehicle 10 and the external environment, and the like. Possible environment acquisition devices 40 include, for example, a speed detection device 41, a position detection device 42, an object detection device 43, a lane detection device 44, a passenger detection device 45, and an image detection device 46. Note that the environment acquisition device 40 is not limited to these, and other environment acquisition devices 40 may be provided.
[0040] The speed detection device 41 is a device that detects the speed of the vehicle 10 (for example, a speedometer, an accelerometer, a gyroscope, etc.). The position detection device 42 is a device that detects the position of the vehicle 10 (for example, a GPS receiver, etc.). The object detection device 43 is a device that detects objects around the vehicle 10 (for example, a laser, a radar, a camera, a sonar, etc.). The lane detection device 44 is a device that detects the driving lane (for example, a laser that detects the amount of reflected light from a lane line, etc.). The passenger detection device 45 is a device that detects the driver and passengers in the passenger car (for example, a seat weight measuring device, etc.). The image detection device 46 is a device that detects images around the vehicle 10 (for example, a camera used to identify signs, etc.). The data detected by these devices may be used as raw data as is, or a calculated value from each detection result may be used (for example, the speed change rate, the difference in speed with the preceding vehicle, the time until a collision with the preceding vehicle, etc. may be calculated and used). The automatic driving process using the above-mentioned driving assistance system will be described with reference to the flowchart of FIG.
[0041] First, in step S100 shown in Fig. 4, it is checked whether the current mode is an autonomous driving mode (a mode in which a driving assistance system is used). If it is the autonomous driving mode, the process proceeds to step S115. On the other hand, if it is not the autonomous driving mode, the process ends and the vehicle 10 is driven by the driver.
[0042] In step S115, environmental information about the vehicle 10 is acquired using the steering device 30 and the environment acquisition device 40. Note that the information acquired here is wide-ranging and is not particularly limited, but includes, for example, information related to the driver's operation, such as how much the driver is depressing the pedal, information related to the output of the vehicle 10, such as the current driving speed, and information related to the environment around the vehicle 10, such as the presence or absence of obstacles and the position of the driving lane. Then, the process proceeds to step S120.
[0043] In step S120, one driver model is selected from the multiple driver models based on the information acquired in step S115. Note that this driver model selection process will be described in detail later. Then, the process proceeds to step S125.
[0044] In step S125, the vehicle 10 is controlled using the selected driver model. Specifically, a control action is generated using the environmental information acquired in step S115 and the driver model selected in step S120. For example, when environmental information is input to a driver model defined as a mathematical expression, a list of control actions is obtained as an output. A control signal is output to the steering device 30 according to the list of control actions thus obtained, and the vehicle 10 is controlled. This completes one cycle of the automatic driving process.
[0045] The above-described cycle of the automatic driving process is continuously executed while the vehicle 10 is traveling. For convenience of explanation, step S115 (obtaining environmental information), step S120 (selecting a driver model), and step S125 (automatic driving control) have been explained as a series of operations, but in reality, these steps can be executed asynchronously. Next, the driver model selection process will be described with reference to the flowchart of FIG.
[0046] First, in step S200 shown in Fig. 5, a determination is made by the state determination program 23. Specifically, the information acquired in step S115 is compared with data on the safety boundary included in the rule data 21a. If it is determined that the vehicle 10 is in a safe state as a result of this determination, the process proceeds to step S210. On the other hand, if it is determined that the vehicle 10 is not in a safe state, the process proceeds to step S230.
[0047] When the process proceeds to step S230, since the vehicle 10 is outside the safety boundary, it is necessary to immediately ensure the safety of the vehicle 10 and stabilize the state of the vehicle 10. For this reason, a recovery plan for stabilizing the state of the vehicle 10 is created. Thereafter, the vehicle 10 is controlled according to this recovery plan for a certain period of time. Note that if another dangerous state is detected while the recovery plan is being executed, it is also possible to re-create (update) the recovery plan. Then, the process proceeds to step S235.
[0048] When the process proceeds to step S210, it is checked whether the current driver model is a rule-based model. If the current driver model is a rule-based model, the process proceeds to step S215. Otherwise, the process proceeds to step S220.
[0049] When the process proceeds to step S215, it is checked whether the safe state has been maintained continuously for a certain period of time (3 seconds or more in this embodiment). If the safe state has been maintained continuously for the certain period of time, it is determined that the state of the vehicle 10 is stable, and the process proceeds to step S220. On the other hand, if the safe state has not been maintained continuously for the certain period of time, even if the vehicle 10 is currently in a safe state, the process proceeds to step S235 since the state of the vehicle 10 may not yet be stable.
[0050] If the process proceeds to step S220, the vehicle 10 is in a safe and stable state, so the probabilistic model is selected as the driver model. Then, the driver model selection process ends.
[0051] If the process proceeds to step S235, the rule-based model is selected as the driver model because the vehicle 10 is not in a safe state or the state of the vehicle 10 is not stable, and the driver model selection process then ends.
[0052] In this way, the driving assistance system and the driving assistance method according to the present embodiment include, as driver models, at least a rule-based model for generating a control action based on a predetermined condition without using a probability function, and a probabilistic model for generating a control action using a probability function based on information collected from a driver who operates a vehicle. Then, one driver model is selected from the multiple driver models during driving, and the vehicle 10 is controlled using the selected driver model.
[0053] Specifically, the safety boundary of the vehicle 10 is defined by the rule data 21a, which is a part of the rule-based model, and the rule-based model is used when the safety boundary is exceeded (when it is determined to be unsafe). Even while driving using the probabilistic model, a check as to whether or not the safety boundary is exceeded is always performed at a constant interval, and when the safety boundary is exceeded (when a known dangerous situation defined by the rule data 21a occurs), the system is quickly switched to the rule-based model, and the safe state defined by the rule data 21a is restored.
[0054] With this configuration, the rule-based model can be used in situations where safety is required, and the probabilistic model can be used in other situations. By automatically switching between driver models in this way, it is possible to achieve both comfort and safety, which was difficult to achieve with conventional automated driving.
[0055] In other words, if it is determined that the vehicle 10 is not in a safe state, the rule-based model is selected, so that the system can automatically switch to the rule-based model in situations where safety is required, thereby improving the safety of the vehicle 10.
[0056] Also, when the rule-based model is used to control the vehicle 10, if it is determined that the vehicle 10 is in a safe state for a predetermined period of time, the probabilistic model is used to control the vehicle 10. Therefore, when the state of the vehicle 10 is stable, the probabilistic model can be used to increase passenger comfort and predictability. [Explanation of symbols]
[0057] 10 Vehicles 11 Control device 12. Computers 20 Storage device 21 Rule-based model data 21a Rules Data 22 Probabilistic Model Data 22a Profile Data 22b Probability Functions 23 Status Judgment Program 24 Flight control program 30 Control equipment 31 Steering 32 Axel 33 Brake 34 Transmission 40 Environment acquisition device 41 Speed detection device 42 Position detection device 43 Object detection device 44 Lane Detection Device 45 Passenger detection device 46 Video detection device
Claims
1. A driving assistance method in which a computer mounted on a vehicle controls the vehicle using a driver model that defines parameters for generating a control operation of the vehicle, comprising: selecting a driver model from the plurality of driver models while driving; controlling the vehicle using a selected driver model (i.e., without using a non-selected driver model); Equipped with The driver model includes at least a rule-based model for generating a control action based on a predetermined condition without using a probability function, and a probability model for generating a control action using a probability function based on information collected from a driver who operates the vehicle; The computer includes: When a situation occurs in which the vehicle deviates from a preset safety boundary while using the probabilistic model, the vehicle is switched to the rule-based model, and When it is determined that the vehicle is in a safe state for a predetermined period of time while traveling using the rule-based model, a process of switching to the probabilistic model is executed. A method for assisting driving of a vehicle.
2. A driving assistance system in which a computer mounted on a vehicle controls the vehicle using a driver model that defines parameters for generating a control operation of the vehicle, comprising: The computer includes: selecting one driver model from the plurality of driver models while driving; configured to control the vehicle using a selected driver model (i.e., without using a non-selected driver model); The driver model includes at least a rule-based model for generating a control action based on a predetermined condition without using a probability function, and a probability model for generating a control action using a probability function based on information collected from a driver who operates the vehicle; The computer includes: When a situation occurs in which the vehicle deviates from a preset safety boundary while using the probabilistic model, the vehicle is switched to the rule-based model, and When it is determined that the vehicle is in a safe state for a predetermined period of time while traveling using the rule-based model, a process of switching to the probabilistic model is executed. Vehicle driving assistance system.
Citation Information
Patent Citations
Driving support device
JP2009205646A
Automatic operation device of vehicle
JP2017061168A
Automatic operation device
JP2018024286A
Vehicle control device, vehicle control method, and program
JP2021160530A
