Processing method, processing system, processing procedure, processing device
By detecting the driver's manual operation deviations and using a safety model to set allowable reaction times, the problem of insufficient driving control precision in autonomous driving is solved, and precise control is achieved in different driving modes.
Patent Information
- Application Number
- CN202280013266.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2021-02-05
- Filing Date
- 2022-02-02
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2042-02-02
AI Technical Summary
Existing technologies struggle to ensure the precision of driving control during autonomous driving, particularly when considering the reaction times of the host and target vehicles. This makes it difficult to set appropriate constraints, resulting in insufficient precision in driving control.
By detecting the driver's manual operation deviations and outputting the allowable reaction time, the accuracy of driving control is ensured by setting appropriate constraints using a safety model.
In both manual and automatic driving modes, by setting the allowable reaction time, driving control can be appropriately constrained, thereby improving the accuracy and safety of driving control.
Smart Images

Figure CN116829432B_ABST
Abstract
Description
[0001] Cross-references to related applications
[0002] This application is based on Japanese Patent Application No. 2021-17658, filed on February 5, 2021, and is incorporated herein by reference in its entirety. Technical Field
[0003] This disclosure relates to processing techniques for performing processes related to driving control of a main moving body. Background Technology
[0004] The technology disclosed in Patent Document 1 plans driving control related to the navigation actions of the main vehicle based on detection information related to the internal and external environment of the main vehicle. Therefore, in cases where a potential accident liability is determined based on a safety model according to the driving strategy and the detection information, safety constraints are imposed on the driving control. These safety constraints take into account the reaction times of both the main vehicle and the target vehicle.
[0005] Patent Document 1: Japanese Patent No. 6708793
[0006] In the technology disclosed in Patent Document 1, a reaction time is envisioned for the main vehicle during autonomous driving. Furthermore, in the technology disclosed in Patent Document 1, a shared reaction time is envisioned for both the main vehicle and the target vehicle. Under such an assumption, situations are anticipated where it may be difficult to ensure the accuracy of driving control. Summary of the Invention
[0007] The present disclosure addresses the following: a processing method for ensuring the accuracy of driving control; a processing system for ensuring the accuracy of driving control; a further object of the present disclosure; and a processing apparatus for ensuring the accuracy of driving control.
[0008] The technical means of this disclosure used to solve the problem are described below.
[0009] The first aspect of this disclosure is a processing method executed by a processor for performing processing related to driving control of a primary mobile body, comprising:
[0010] Detecting manual deviations from steady-state operation by the driver in a manually driven main moving body; and
[0011] Output the permissible reaction time, which is the time allowed for the main moving body to react during the generation of manual deviation, based on the safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
[0012] The second aspect disclosed herein is a processing system comprising a processor that performs processing related to the driving control of the primary mobile body.
[0013] The processor is configured to perform the following processing:
[0014] Detecting manual deviations from steady-state operation by the driver in a manually driven main moving body; and
[0015] Output the permissible reaction time, which is the time allowed for the main moving body to react during the generation of manual deviation, based on the safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
[0016] The third aspect of this disclosure is a processing program stored in a storage medium, containing instructions that the processor executes for performing processing related to the driving control of the main mobile body.
[0017] The instruction causes the processor to perform the following processing:
[0018] In a manually driven main moving body, detect the manual deviation of the driver's manual operation from the steady-state operation;
[0019] Output the permissible reaction time, which is the time allowed for the main moving body to react during the generation of manual deviation, based on the safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
[0020] The fourth aspect of this disclosure is a processing device, which includes a processor and is configured to be mounted on a main mobile body to perform processing related to the driving control of the main mobile body.
[0021] The processor is configured to perform the following processing:
[0022] Detecting manual deviations from steady-state operation by the driver in a manually driven main moving body; and
[0023] Output the permissible reaction time, which is the time allowed for the main moving body to react during the generation of manual deviation, based on the safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
[0024] According to these first to fourth methods, in a manually driven main vehicle, during the generation of manual deviations (deviations from steady-state operation) caused by the driver's manual operation, the permissible reaction time of the main vehicle is output based on a safety model. This safety model is a model of the driving strategy, obtained by modeling the safety of the expected functions. Therefore, since a permissible reaction time specifically designed for scenarios where manual deviations occur can be envisioned, appropriate constraints can be set for the manually driven main vehicle to ensure the accuracy of driving control.
[0025] The fifth aspect of this disclosure is a processing method, which is executed by a processor for performing processing related to the driving control of a main mobile body, and includes:
[0026] Detecting a target moving object that travels subsequently relative to the primary moving body in autonomous driving; and
[0027] Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy, which is obtained by modeling the safety of the expected function.
[0028] The sixth aspect of this disclosure is a processing system that includes a processor that performs processing related to the driving control of the primary mobile body.
[0029] The processor is configured to perform the following processing:
[0030] Detecting a target moving object that travels subsequently relative to the primary moving body in autonomous driving; and
[0031] Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy, which is obtained by modeling the safety of the expected function.
[0032] The seventh aspect of this disclosure is a processing program stored in a storage medium, containing instructions that a processor executes for performing processing related to the driving control of the main mobile body.
[0033] The instruction causes the processor to perform the following processing:
[0034] Detecting a target moving object that travels subsequently relative to the primary moving body in autonomous driving; and
[0035] Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy, which is obtained by modeling the safety of the expected function.
[0036] The eighth aspect of this disclosure is a processing apparatus, which includes a processor and is configured to be mounted on a main mobile body to perform processing related to the driving control of the main mobile body.
[0037] The processor is configured to perform the following processing:
[0038] Detecting a target moving object that travels subsequently relative to the primary moving body in autonomous driving; and
[0039] Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy, which is obtained by modeling the safety of the expected function.
[0040] According to these fifth to eighth methods, in the autonomous driving master vehicle, the permissible reaction time of the target vehicle during subsequent driving is output based on a safety model. This safety model is obtained by modeling the safety of the expected functions, based on a driving strategy model. Therefore, since a permissible reaction time specifically for the target vehicle's subsequent driving scenarios can be envisioned, appropriate constraints can be set on the autonomous driving master vehicle to ensure the accuracy of driving control. Attached Figure Description
[0041] Figure 1 This is a descriptive table indicating the explanation of the terms used in this disclosure.
[0042] Figure 2 This is a descriptive table indicating the explanation of the terms used in this disclosure.
[0043] Figure 3 This is a descriptive table indicating the explanation of the terms used in this disclosure.
[0044] Figure 4 This is a table of definitions indicating the terms used in this disclosure.
[0045] Figure 5 This is a table of definitions indicating the terms used in this disclosure.
[0046] Figure 6 This is a block diagram illustrating the processing system of the first embodiment.
[0047] Figure 7This represents the expected driving environment of the main vehicle used in the first embodiment.
[0048] Figure 8 This is a block diagram illustrating the processing system of the first embodiment.
[0049] Figure 9 This is an example of the lane structure in the first embodiment.
[0050] Figure 10 This is a flowchart illustrating the processing method of the first embodiment.
[0051] Figure 11 This is a flowchart illustrating the processing method of the second embodiment.
[0052] Figure 12 This is a block diagram illustrating the processing system of the third embodiment.
[0053] Figure 13 This is a flowchart illustrating the processing method of the third embodiment.
[0054] Figure 14 This is a flowchart illustrating the processing method of the fourth embodiment.
[0055] Figure 15 This is a block diagram illustrating the processing system of the fifth embodiment.
[0056] Figure 16 This is a flowchart illustrating the processing method of the fifth embodiment.
[0057] Figure 17 This is a flowchart illustrating the processing method of the sixth embodiment.
[0058] Figure 18 This is a block diagram illustrating the processing system of the seventh embodiment.
[0059] Figure 19 This is a block diagram illustrating the processing system of the seventh embodiment.
[0060] Figure 20 This is a block diagram illustrating the processing system of the eighth embodiment.
[0061] Figure 21 This is a flowchart illustrating the processing method of the eighth embodiment.
[0062] Figure 22 This is a flowchart illustrating the processing method of the ninth embodiment. Detailed Implementation
[0063] Hereinafter, several embodiments of the present disclosure will be described based on the accompanying drawings. Furthermore, there are cases where repeated descriptions are omitted by using the same reference numerals for corresponding components in each embodiment. Additionally, when only a portion of the structure is described in each embodiment, structures from other previously described embodiments can be applied to the remaining parts of that structure. Moreover, not only combinations of structures explicitly shown in the descriptions of each embodiment are possible, but structures from multiple embodiments can be partially combined with each other even if not explicitly shown, provided that such combinations do not create obstacles.
[0064] Figures 1-5 The description shows the terminology associated with various embodiments of this disclosure. However, the definitions of the terminology are not limited to... Figures 1-5 The explanation is not intended to be based on the description shown, but rather to be interpreted within the scope of the spirit of this disclosure.
[0065] (First Implementation)
[0066] Figure 6 The processing system 1 of the first embodiment shown performs processing related to the driving control of the main mobile body (hereinafter referred to as driving control processing). The main mobile body that is the target of the driving control processing of the processing system 1 is... Figure 7 The main vehicle 2 shown. From the perspective of the main vehicle 2, it can also be said that the main vehicle 2 is the ego-vehicle.
[0067] In the main vehicle 2, autonomous driving is implemented. Autonomous driving is categorized based on the degree of occupant manual intervention during Dynamic Driving Tasks (DDTs). Autonomous driving can also be achieved by the system performing autonomous driving control of all DDTs while operating at conditions of conditional driving automation, high-level driving automation, or full driving automation. Autonomous driving can also be achieved by the driver, acting as an occupant, performing some or all of the high-level driving assistance controls of DDTs, such as driver assistance or partial driving automation. Autonomous driving can also be achieved through any one, combination, or switching between these autonomous driving controls and high-level driving assistance controls.
[0068] In the main vehicle 2, equipped with Figure 6 , Figure 8 The sensor system 5, communication system 6, map database 7, and information display system 4 are shown. The sensor system 5 acquires sensor data that can be used by the processing system 1 by detecting the external and internal boundaries of the main vehicle 2. For this purpose, the sensor system 5 is composed of an external sensor 50 and an internal boundary sensor 52.
[0069] The external sensor 50 can also detect objects present outside the host vehicle 2. The external sensor 50 of the object detection type is, for example, at least one of a camera, LiDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging), lidar, millimeter-wave radar, and ultrasonic sonar. The external sensor 50 can also detect the state of the atmosphere outside the host vehicle 2. The external sensor 50 of the atmosphere detection type is, for example, at least one of an external temperature sensor and a humidity sensor.
[0070] The interior boundary sensor 52 can also detect specific physical quantities (hereinafter referred to as motion physical quantities) related to vehicle motion within the interior boundary of the main vehicle 2. The interior boundary sensor 52 of the physical quantity detection type is, for example, at least one of a speed sensor, an acceleration sensor, and a gyroscope sensor. The interior boundary sensor 52 can also detect the state of the occupants within the interior boundary of the main vehicle 2. The interior boundary sensor 52 of the occupant detection type is, for example, at least one of an actuator sensor, a driver state monitor, a bio-sensor, a seating sensor, and an in-vehicle equipment sensor. Specifically, as an actuator sensor, at least one of an accelerometer sensor, a brake sensor, and a steering control sensor is used to detect the occupant's operating state related to the motion actuator of the main vehicle 2.
[0071] Communication system 6 acquires communication data that can be used by processing system 1 via wireless communication. Communication system 6 can also receive positioning signals from GNSS (Global Navigation Satellite System) satellites located outside the main vehicle 2. Positioning-type communication system 6 may be, for example, a GNSS receiver. Communication system 6 can also transmit and receive communication signals with V2X systems located outside the main vehicle 2. V2X-type communication system 6 may be, for example, at least one of DSRC (Dedicated Short Range Communications) communicators and cellular V2X (C-V2X) communicators. Communication system 6 can also transmit and receive communication signals with terminals located inside the main vehicle 2. Terminal communication-type communication system 6 may be, for example, at least one of Bluetooth devices, Wi-Fi devices, and infrared communication devices.
[0072] Map DB7 stores map data that can be used by processing system 1. Map DB7 may be constructed using at least one non-transitory tangible storage medium, such as semiconductor memory, magnetic media, or optical media. Map DB7 may also be a DB of a locator used to infer the self-state quantities of the main vehicle 2, including its own position. Map DB may also be a DB of a navigation unit for navigating the driving path of the main vehicle 2. Map DB7 may also be constructed from a combination of various DBs.
[0073] Map DB7 acquires and stores the latest map data, for example, through communication with an external center of communication system 6 via V2X. The map data is digitized into two or three dimensions to represent the driving environment of the main vehicle 2. As three-dimensional map data, high-precision map digital data can also be used. The map data may also include road data, such as the location coordinates, shape, and road surface condition of road structures. The map data may also include label data, such as the location coordinates and shape of road signs, road markings, and zoning lines attached to the road. The label data included in the map data may also represent road signs such as traffic signs, arrow marks, lane markings, stop lines, directional signs, road beacons, rectangular signs, commercial signs, or changes in road line maps. The map data may also include data on at least one structure, such as the location coordinates and shape of buildings and traffic lights facing the road. The label data included in the map data may also represent road signs such as streetlights, road edges, reflectors, utility poles, or the back of road signs.
[0074] Information prompting system 4 provides information to occupants, including the driver of the main vehicle 2. Information prompting system 4 comprises visual prompting units, auditory prompting units, and tactile prompting units. The visual prompting units provide information by stimulating the occupants' vision. Visual prompting units may be at least one of the following: HUD (Head-up Display), MFD (Multi-Function Display), instrument cluster, navigation unit, and luminous unit. The auditory prompting units provide information by stimulating the occupants' hearing. Auditory prompting units may be at least one of the following: speaker, buzzer, and vibration unit. The tactile prompting units provide information by stimulating the occupants' skin sensation. Skin sensations stimulated by the tactile prompting units may include at least one of the following: touch, temperature, and wind. Tactile prompting units may be at least one of the following: steering wheel vibration unit, driver's seat vibration unit, steering wheel reaction force unit, accelerator pedal reaction force unit, brake pedal reaction force unit, and air conditioning unit.
[0075] like Figure 6 As shown, the processing system 1 is connected to the sensor system 5, communication system 6, map DB7, and information prompting system 4 via at least one of the following: LAN (Local Area Network), wiring harness, internal bus, and wireless communication line. The processing system 1 is configured to include at least one dedicated computer. The dedicated computer constituting the processing system 1 may also be an integrated ECU (Electronic Control Unit) that integrates the driving control of the main vehicle 2. The dedicated computer constituting the processing system 1 may also be a determination ECU that determines DDT in the driving control of the main vehicle 2. The dedicated computer constituting the processing system 1 may also be a monitoring ECU that monitors the driving control of the main vehicle 2. The dedicated computer constituting the processing system 1 may also be an evaluation ECU that evaluates the driving control of the main vehicle 2.
[0076] The dedicated computer constituting the processing system 1 can also be a navigation ECU that navigates the driving path of the main vehicle 2. The dedicated computer constituting the processing system 1 can also be a locator ECU that infers its own state variables, including the position of the main vehicle 2. The dedicated computer constituting the processing system 1 can also be an actuator ECU that controls the motion actuators of the main vehicle 2. The dedicated computer constituting the processing system 1 can also be an HCU (Hman Machine Interface) Control Unit that controls information prompts in the main vehicle 2. The dedicated computer constituting the processing system 1 can also be at least one external computer, which may be configured as, for example, an external center or mobile terminal capable of communication via the communication system 6.
[0077] The dedicated computer constituting the processing system 1 has at least one memory 10 and at least one processor 12. The memory 10 is a non-transitory tangible storage medium that stores computer-readable programs and data, such as at least one of semiconductor memory, magnetic media, and optical media. The processor 12 includes, for example, at least one of CPU (Central Processing Unit), GPU (Graphics Processing Unit), and RISC (Reduced Instruction Set Computer) as its core.
[0078] The processor 12 executes multiple instructions contained in a processing program stored as software in the memory 10. Thus, the processing system 1 constructs multiple functional modules for performing driving control processing of the main vehicle 2. In this way, in the processing system 1, multiple functional modules are constructed by the processor 12 executing multiple instructions through a processing program stored in the memory 10 for performing driving control processing of the main vehicle 2. Figure 8 As shown, the multiple functional modules constructed by the processing system 1 include a detection block 100, a planning block 120, a risk monitoring block 140, and a control block 160.
[0079] Detection block 100 acquires sensor data from external sensors 50 and internal sensors 52 of sensor system 5. Detection block 100 acquires communication data from communication system 6. Detection block 100 acquires map data from map DB7. Detection block 100 detects the internal and external environment of the main vehicle 2 by fusing these acquired data as inputs. Through the detection of the internal and external environment, detection block 100 generates detection information for subsequent planning block 120 and risk monitoring block 140. Thus, when generating detection information, detection block 100 acquires data from sensor system 5 and communication system 6, and identifies or understands the meaning of the acquired data. It can be said that it integrates the acquired data to grasp all conditions, including the external conditions of the main vehicle 2, its own position within them, and the internal conditions of the main vehicle 2. Detection block 100 can also provide substantially the same detection information to planning block 120 and risk monitoring block 140. Detection block 100 can also provide different detection information to planning block 120 and risk monitoring block 140.
[0080] The detection information generated by detection block 100 describes the state detected in each scenario within the driving environment of the main vehicle 2. Detection block 100 can also generate object detection information by detecting objects in the external environment of the main vehicle 2, including road users, obstacles, and structures. Object detection information can also represent at least one of, for example, distance to the object, relative velocity of the object, relative acceleration of the object, and inferred state based on object tracking detection. Object detection information can also represent the type identified or determined based on the state of the detected object. Detection block 100 can also generate driving route detection information by detecting the current and future driving routes of the main vehicle 2. Driving route detection information can, for example, represent the state of at least one of, such as road surface, lane, roadside, and free space.
[0081] Detection block 100 can also generate detection information for its own state quantity by inferentially detecting the location of its own state quantity, which includes the position of the main vehicle 2. Simultaneously with the detection information of its own state quantity, detection block 100 can also generate update information for map data related to the driving route of the main vehicle 2 and feed this update information back to map DB7. Detection block 100 can also generate detection information for signs associated with the driving route of the main vehicle 2. The detection information for signs can, for example, represent the state of at least one of signs, lane markings, and traffic lights. The detection information for signs can also represent traffic rules identified or determined based on the state of the signs. Detection block 100 can also generate detection information for weather conditions in each driving scenario of the main vehicle 2. Detection block 100 can also generate detection information for the time of each driving scenario of the main vehicle 2.
[0082] Planning block 120 obtains detection information from detection block 100. Planning block 120 plans the driving control of the main vehicle 2 based on the obtained detection information. In the driving control planning, control commands related to the navigation actions of the main vehicle 2 and the driver's assistance actions are generated. That is, planning block 120 implements the DDT function, which generates control commands as motion control requests for the main vehicle 2. The control commands generated by planning block 120 may also include control parameters for controlling the motion actuators of the main vehicle 2. Motion actuators that become the output objects of the control commands can include at least one of the following: an internal combustion engine, an electric motor, and powertrains composed of combinations thereof, braking devices, and steering mechanisms.
[0083] Planning block 120 can also generate control commands in a manner suitable for the driving strategy by using a safety model described according to the driving strategy and its safety. The driving strategy upon which the safety model is based is, for example, defined by a vehicle-level safety strategy that guarantees the intended function (Safety Of The Intended Functionality: hereinafter referred to as SOTIF). In other words, the safety model is described by modeling SOTIF based on a driving strategy that implements the vehicle-level safety strategy. Planning block 120 can also train the safety model by a machine learning algorithm that backpropagates the driving control results to the safety model. As the training safety model, for example, a learning model based on at least one of deep learning, such as a neural network (DNN), and reinforcement learning can also be used. Here, the so-called safety model can be defined as safety-related models themselves, which represent the safety-related aspects of driving actions based on assumptions of actions reasonably foreseeable by other road users, or it can be defined as a model that constitutes a part of the safety-related model. Such a safety model can be constructed in at least one of the following forms: a mathematical model that formulates vehicle-level safety, and a computer program that performs processing based on the mathematical model.
[0084] Planning block 120 can also plan the future path of the main vehicle 2 through driving control before generating control commands. It can also perform path planning, for example, through simulation or other calculations, to navigate the main vehicle 2 based on detection information. That is, planning block 120 can also implement the DDT (Directed Driving Test) function of planning the path as a tactical action of the main vehicle 2. Planning block 120 can also plan an appropriate track for the main vehicle 2 along the planned path based on the acquired detection information before generating control commands. That is, planning block 120 can also implement the DDT function of planning the track of the main vehicle 2. The track planned by planning block 120 can also specify at least one of the following as motion physical quantities related to the main vehicle 2, such as driving position, speed, acceleration, and yaw rate, in a time series. The time series track planning constructs a future driving scheme based on the navigation of the main vehicle 2. Planning block 120 can generate a track using a planning method that uses a safety model. In this case, a cost function that assigns a cost to the generated track can be calculated, and a safety model can be trained using a machine learning algorithm based on the result of this calculation.
[0085] Planning block 120 can also plan adjustments to the autonomous driving level in the master vehicle 2 based on the acquired detection information. Adjustments to the autonomous driving level can also include handover between autonomous and manual driving. This handover can be achieved by setting the Operational Design Domain (ODD) for autonomous driving, in conjunction with entry or exit schemes for that ODD. In the exit scheme from the ODD, i.e., the handover scheme from autonomous to manual driving, unreasonable situations deemed to pose unreasonable risks based on a safety model are cited as use cases. In this use case, planning block 120 can also plan a DDT rollback, whereby the driver, acting as a rollback candidate, performs a minimum-risk operation on the master vehicle 2 to move it to a minimum-risk state.
[0086] The adjustment of the autonomous driving level may also include the downgrade of the primary vehicle 2. In the downgrade driving scheme, if there is an unreasonable risk due to the handover to manual driving, the unreasonable situation judged based on a safety model is cited as a use case. In this use case, planning block 120 may also plan a DDT rollback to move the primary vehicle 2 to the minimum risk state through autonomous driving and autonomous stopping. The DDT rollback to move the primary vehicle 2 to the minimum risk state is implemented not only in the adjustment of downgrading the autonomous driving level, but also in the adjustment of downgrading driving while maintaining the autonomous driving level, such as in MRM (Minimum Risk Maneuver). In the DDT rollback to move the primary vehicle 2 to the minimum risk state, the visibility of the transition situation may be increased, for example, by using at least one of the following: lighting, horn sound, signal, and gesture.
[0087] Risk monitoring block 140 acquires detection information from detection block 100. Based on the acquired detection information, risk monitoring block 140 monitors the main vehicle 2 and other moving targets 3 in each scenario (see reference). Figure 7 Risk monitoring block 140 performs risk monitoring based on detection information in a time-series manner to ensure the SOTIF of the host vehicle 2 for the target moving body 3. The target moving body 3 envisioned in the risk monitoring is other road users existing in the driving environment of the host vehicle 2. Target moving bodies 3 include, for example, non-vulnerable road users such as cars, trucks, motorcycles, and bicycles, and vulnerable road users such as pedestrians. Target moving bodies 3 may also include animals.
[0088] The risk monitoring block 140 sets a safety envelope in the main vehicle 2 to ensure SOTIF (Safety in Context of Context) based on the detection information acquired for each scenario, such as a vehicle-level safety strategy. The risk monitoring block 140 can use a safety model based on the aforementioned driving strategy to set the safety envelope between the main vehicle 2 and the target moving body 3. The safety model used in setting the safety envelope can also be designed to comply with accident liability rules to avoid potential accident liability caused by unreasonable risks or misuse by road users. In other words, the safety model can also be designed to ensure that the main vehicle 2 complies with accident liability rules based on the driving strategy. Examples of such a safety model include, for instance, the responsibility-sensitive safety model disclosed in Patent Document 1.
[0089] Here, the so-called safety envelope can be defined as a set of constraints and conditions designed to allow the system to operate within an acceptable risk level, with the system acting as a constraint or controllable object. Such a safety envelope, as a physics-based margin around each road user including the main vehicle 2 and the target moving body 3, can be set using margins related to at least one of the following physical quantities of motion: distance, speed, and acceleration. For example, in setting the safety envelope, a safe distance can be envisioned based on a safety model of the main vehicle 2 and the target moving body 3, assumed to be based on a driving strategy, according to a curve related to at least one physical quantity of motion. The safe distance defines the boundaries of the physics-based margin around the main vehicle 2 for the predicted motion of the target moving body 3. The safe distance can also be envisioned taking into account the reaction time of the road user until they perform an appropriate response. The safe distance can also be envisioned to comply with accident liability rules. For example, in scenarios with lane structures such as lanes, a safe distance can be calculated to avoid the risk of rear-end collisions and frontal collisions longitudinally to the main vehicle 2, and a safe distance can be calculated to avoid the risk of side collisions laterally to the main vehicle 2. On the other hand, in scenarios where there is no lane structure, the safe distance to avoid the risk of track collision in any direction of the main vehicle 2 can be calculated.
[0090] The risk monitoring block 140 can also determine the conditions of each relative motion scenario between the main vehicle 2 and the target moving body 3 before setting the aforementioned safety envelope. For example, in scenarios with lane structures such as lanes, it can also determine the conditions under which the risk of longitudinal rear-end collision and frontal collision is envisioned, as well as the conditions under which the risk of lateral collision is envisioned. In these longitudinal and lateral condition determinations, the state variables related to the main vehicle 2 and the target moving body 3 can also be transformed to a coordinate system based on straight lanes. On the other hand, in scenarios without lane structures, it can also determine the conditions under which the risk of track collision is envisioned in any direction of the main vehicle 2. Furthermore, for the above condition determination function, at least a portion of it can be performed by the detection block 100 to provide the condition determination results as detection information to the risk monitoring block 140.
[0091] The risk monitoring block 140 performs a safety determination between the main vehicle 2 and the target moving body 3 based on the set safety envelope and the detection information acquired for each scenario. That is, the risk monitoring block 140 determines safety by testing whether there is a safety envelope violation in the driving scenario interpreted based on the detection information between the main vehicle 2 and the target moving body 3. If a safety distance is assumed in the setting of the safety envelope, a determination of no safety envelope violation can be made if the actual distance between the main vehicle 2 and the target moving body 3 exceeds the safety distance. Conversely, a determination of a safety envelope violation can be made if the actual distance between the main vehicle 2 and the target moving body 3 is below the safety distance.
[0092] In the event of a determination that a safety envelope violation has occurred, the risk monitoring block 140 can also use simulation to calculate a reasonable plan for providing the appropriate action to the master vehicle 2 as a suitable response. In simulating the reasonable plan, the actions to be taken for each transition state can be set as constraints on the master vehicle 2 by inferring the state transitions between the master vehicle 2 and the target moving body 3 (described in detail later). In setting the actions, the assumed limit value for the motion physical quantity can also be calculated by restricting at least one motion physical quantity given to the master vehicle 2 as a constraint on the master vehicle 2.
[0093] The risk monitoring block 140 can also directly calculate the limit value for compliance with accident liability rules based on a safety model of the main vehicle 2 and the target moving body 3, which are assumed to be based on a driving strategy, and according to a curve related to at least one physical quantity of motion. The calculation of the direct limit value is itself a setting of the safety envelope, or a setting of constraints on driving control. Therefore, if a real value closer to the safe side than the limit value is detected, a determination can be made that there is no violation of the safety envelope. On the other hand, if a real value deviating from the limit value is detected, a determination can be made that there is a violation of the safety envelope.
[0094] The risk monitoring block 140 can store at least one type of evidentiary information, such as detection information for setting a security envelope, determination information indicating the determination result of the security envelope, detection information affecting the determination result, and simulated schemes, in the memory 10. The memory 10 storing the evidentiary information may be installed inside the main vehicle 2, or, for example, in an external center outside the main vehicle 2, depending on the type of dedicated computer constituting the processing system 1. The evidentiary information can be stored unencrypted, encrypted, or hashed. Storage of the evidentiary information is performed at least when a security envelope violation is determined. Of course, storage of the evidentiary information can also be performed when no security envelope violation is determined. Evidence information where no security envelope violation is determined can be effectively used as a lagging indicator at the time of storage, and can also be effectively used as a forward-looking indicator for the future.
[0095] Control block 160 obtains control commands from planning block 120. Control block 160 obtains decision information related to the safety envelope from risk monitoring block 140. That is, control block 160 implements the DDT function to control the movement of the main vehicle 2. If control block 160 obtains decision information indicating that there is no violation of the safety envelope, it executes the planned driving control of the main vehicle 2 according to the control commands.
[0096] In contrast, when control block 160 receives a determination that a safety envelope violation has occurred, it imposes constraints on the planned driving control of the main vehicle 2 based on the determination information and the driving strategy. These constraints can be functional restrictions, degraded constraints, or constraints different from these. Driving control is constrained by limiting control commands. If a reasonable scenario has been simulated by risk monitoring block 140, control block 160 can also limit control commands based on that scenario. In this case, if a limit value is set for the motion physical quantities of the main vehicle 2, the control parameters of the motion actuators included in the control commands can be modified based on that limit value.
[0097] The details of the first embodiment will be described below.
[0098] like Figure 9 As shown, the first embodiment envisions a lane structure Ls that divides the lanes. The lane structure Ls uses the direction in which the lanes extend as the longitudinal direction to restrict the movement of the main vehicle 2 and the target moving body 3. The lane structure Ls also uses the width direction or the direction in which the lanes are arranged as the lateral direction to restrict the movement of the main vehicle 2 and the target moving body 3.
[0099] The driving strategy between the main vehicle 2 and the target moving body 3 in the lane structure Ls, for example, when the target moving body 3 is the target vehicle 3a, is specified as (A) to (E) below. Furthermore, the so-called forward direction based on the main vehicle 2 means, for example, the direction of travel on the rotation circle at the current steering angle of the main vehicle 2, the direction of travel through the vehicle's center of gravity orthogonal to the axle in the main vehicle 2, or the direction of travel on the axis from the front camera module to the FOE (Focus of Expansion) of the sensor system 5 in the main vehicle 2, etc.
[0100] (A) A vehicle does not rear-end a vehicle traveling in front.
[0101] (B) The vehicle shall not forcibly insert itself between other vehicles.
[0102] (C) Even when a vehicle has priority, it will yield to other vehicles depending on the situation.
[0103] (D) Drive carefully in areas with poor visibility.
[0104] (E) Regardless of whether the accident is the fault of the driver or the other party, the driver shall take reasonable action if the accident could have been prevented.
[0105] This is a safety model based on driving strategies and modeled using SOTIF. It envisions the actions of road users in situations that do not reach the level of unreasonableness as appropriate and reasonable actions to be taken. The unreasonable situations between the main vehicle 2 and the target moving body 3 in the lane structure Ls are frontal collisions, rear-end collisions, and side collisions. For example, if the target moving body 3 for the main vehicle 2 is target vehicle 3a, reasonable actions in a frontal collision include braking by vehicles traveling in the opposite direction. For example, if the target moving body 3 for the main vehicle 2 is target vehicle 3a, reasonable actions in a rear-end collision include the vehicle in front not applying sufficient emergency braking, and thus avoiding a rear-end collision by vehicles traveling behind. For example, if the target moving body 3 for the main vehicle 2 is target vehicle 3a, reasonable actions in a side collision include swerving parallel vehicles in their respective directions of separation. When envisioning reasonable actions, the state variables related to the main vehicle 2 and the target moving body 3, whether it is a lane structure Ls for lane turning or a lane structure Ls for lane elevation, are converted into a hypothetical straight and planar lane structure Ls to define orthogonal coordinate systems in the longitudinal and lateral directions.
[0106] The safety model can be designed to assign liability for accidents to moving bodies that fail to take reasonable action, according to accident liability rules. Under the accident liability rules in the lane structure Ls, the safety model, used to monitor the risk between the main vehicle 2 and the target moving body 3, sets a safety envelope for the main vehicle 2 in a way that avoids potential accident liability through reasonable action. Therefore, the risk monitoring block 140 of the processing system 1, under normal conditions, determines whether there is a violation of the safety envelope by comparing the actual distance between the main vehicle 2 and the target moving body 3 with the safe distance based on the safety model for each driving scenario. In the case of a violation of the safety envelope, the risk monitoring block 140 simulates a scheme for giving reasonable action to the main vehicle 2. Through simulation, the risk monitoring block 140 sets, for example, a limit value related to at least one of speed and acceleration, as a constraint on driving control in the control block 160.
[0107] In the first embodiment, multiple functional modules jointly execute according to Figure 10 The flowchart shown illustrates a method for executing driving control processing. The processing method of the first embodiment is repeatedly executed during manual driving planned by the planning block 120. Furthermore, in the following description, each "S" in the processing method refers to a plurality of steps executed by a plurality of instructions included in the processing program.
[0108] In processing method S100, risk monitoring block 140 determines whether detection block 100 has detected a manual deviation in the driver's manual operation of the manually driven main vehicle 2, indicating a departure from steady-state operation. The detection of manual deviation by detection block 100 is based on operation data representing the driver's operating state, which is sensor data acquired from inner boundary sensors 52, such as actuator sensors, in sensor system 5.
[0109] When determining whether a manual deviation is detected, the risk monitoring block 140 acquires steady-state information related to steady-state operation. Steady-state operation refers to a reasonable or minimal-risk operation of the motion actuators controlled in each scenario within the autonomous driving master vehicle 2. Therefore, the steady-state information can be acquired by including the reasonable or minimal-risk operation amount, and may also include the dispersion value (i.e., the tolerance error) of that operation amount. The risk monitoring block 140 can also acquire steady-state information from the planning block 120, which plans trajectories based on steady-state operation in autonomous driving. The risk monitoring block 140 acquires steady-state information through calculations such as simulations based on the motion physical quantity curves envisioned according to the safety model in autonomous driving.
[0110] When determining whether a manual deviation is detected, the risk monitoring block 140 acquires detection information regarding the deviation-generating operation, which is a manual operation by the driver that induces the manual deviation, generated by the detection block 100 based on operational data. The deviation-generating operation can be an additional manual operation manually given for risk avoidance, performed to mitigate risk in each scenario where the driver psychologically or perceptually deviates from the steady-state operation in high-level driver assistance control. The deviation-generating operation can also be a manual operation in each scenario of manual driving without autonomous driving intervention, where the driver deviates from the steady-state operation by manually giving the operation. The deviation-generating operation can also be, for example, a fine-tuning operation involving turning and returning the steering wheel (i.e., correcting the steering) for a steady-state operation of steering maneuvering consistent with the curvature of a curve. The deviation-generating operation can also be, for example, a brake-on operation for a steady-state operation of brake disengagement at a straight road or curve. The deviation-generating operation can also be, for example, an accelerator-on operation for a steady-state operation of accelerator depressing at the exit of a straight road or curve. To represent such a deviation-generating operation, the detection information is acquired based on the amount of driver input to the motion actuator during manual driving.
[0111] In S100, the risk monitoring block 140 determines whether a manual deviation has been detected by checking whether the difference between the operating quantities represented by the steady-state information and the detection information is outside a set range. The set range for determining whether a deviation has been detected can be set to be less than the lower limit of the difference that should be considered as a manual deviation, or the upper limit of the difference that should be considered as a steady-state operation range. When the steady-state information includes a dispersion value of the operating quantity, the set range of the difference can be set to be less than the upper limit obtained by adding the dispersion value to the operating quantity on the safety side.
[0112] In S100, if the difference between the operational quantities is within a set range, the risk monitoring block 140 determines that no manual deviation has been detected, and the current process of the processing method ends. On the other hand, if the difference between the operational quantities is outside the set range in S100, the risk monitoring block 140 determines that a manual deviation has been detected, and the processing method moves to S101.
[0113] In processing step S101, the risk monitoring block 140 acquires the permissible reaction time ρp, which is the time allowed as the reaction time ρ of the main vehicle 2 for the target moving body 3, based on a safety model. This safety model is a model based on the driving strategy and is modeled using SOTIF. When a manual deviation occurs in the main vehicle 2, the reaction time ρ refers to the time required for the main vehicle 2, including the driver's reaction, to react to the driver's deviation.
[0114] The reaction time ρ of the main vehicle 2 is related to the safe distance dmin, which determines the driving control of the main vehicle 2 in the safety model. That is, the reaction time ρ of the main vehicle 2 is used as a variable in the safety function L, which represents the curve of the kinematic physical quantity used to calculate the safe distance dmin according to Equation 1. In Equation 1, Q is at least one kinematic physical quantity used for the kinematic curve. As the kinematic physical quantity Q, it is selected according to each scheme or scenario of the envisioned safety model, and is related to at least one of the main vehicle 2 and the target moving body 3, such as speed, acceleration / deceleration, azimuth angle, azimuth angular velocity, and position offset.
[0115] [Formula 1]
[0116] dmin=L(ρ,Q)
[0117] The inverse function R of the safety function L is defined by a function or algorithm based on the safety model between the main vehicle 2 and the target moving body 3, satisfying Equation 2. In Equation 2, dr is the actual distance compared to the safety distance dmin when a safety envelope violation is determined, i.e., the separation distance between the main vehicle 2 and the target moving body 3 at the time of executing S101. Based on these, the risk monitoring block 140 in S101 calculates the permissible reaction time ρp of the main vehicle 2 according to Equation 3 using the inverse function R. After completing the execution of S101, the processing method moves to S102.
[0118] [Equation 2]
[0119] dr≥L(R(dr,Q),Q)
[0120] [Formula 3]
[0121] ρp = R(dr, Q)
[0122] In processing step S102, the risk monitoring block 140 obtains the operational margin time ρo granted to the driver's manual operation in the main vehicle 2 based on the permissible reaction time ρp obtained in S101. The operational margin time ρo can also be described as the sufficient time allowed for operation due to deviations of the driver in the main vehicle 2, based on the safety model between the main vehicle 2 and the target moving body 3. The operational margin time ρo is calculated according to Equation 4 using the permissible reaction time ρp. In Equation 4, ρv is the minimum time required for the main vehicle 2 to avoid unreasonable risks under unreasonable conditions, defined as the action-required time. The action-required time ρv is set as the anticipated time required to avoid risks according to each scenario after automatic driving intervention with manual driving. After completing S102, the processing method proceeds to S103.
[0123] [Formula 4]
[0124] ρo=ρp-ρv
[0125] In processing step S103, the risk monitoring block 140 outputs evidence information, including at least one of the allowable reaction time ρp obtained in S101 and the operational margin time ρo obtained in S102, to the memory 10. The evidence information is stored in the memory 10 in association with at least one of the output times ρp and ρo, and a timestamp indicating the time of occurrence of the operation object scene. The evidence information may include, for example, a calculated variable including the allowable reaction time ρp of the motion physical quantity Q, a calculated variable including the operational margin time ρo of the behavior-required time ρv, detection information identifying the target moving body 3, and detection information including the behavior of the target moving body 3. The output of the evidence information may be performed at intervals corresponding to each cycle of the processing method corresponding to the control cycle. The output of the evidence information in S103 may be performed at intervals longer than one cycle of the processing method, or at intervals of multiple cycles of the processing method, for example, to exclude noise information. In the case of output for each set cycle or multiple cycles, S103 itself is skipped at non-output times.
[0126] In S103, evidence information can be stored either by outputting to a memory 10 installed within the main vehicle 2 or by remotely distributing to a memory 10 that serves as an external center, which is outside the main vehicle 2. When the memory 10 outputting evidence information is installed within the main vehicle 2, it can be mechanically protected even if the main vehicle 2 is involved in a collision. When the memory 10 outputting evidence information is installed within the main vehicle 2, it can also be protected in a fire-resistant mounting location. When the memory 10 outputting evidence information is installed within the main vehicle 2, it can also be protected in a water-resistant mounting location. In this protective memory 10 installed within the main vehicle 2, encrypted or hashed evidence information can also be stored. In the case of encrypted evidence information, the decryption key can be stored in at least one of the protective memory 10 within the main vehicle 2, the unprotected memory 10 within the main vehicle 2, and the memory 10 at the external center. In the case of evidence information that has been hashed, the transaction containing the hash value can be stored in at least one of the following: the protection memory 10 inside the main vehicle 2, the non-protection memory 10 inside the main vehicle 2, and the memory 10 in an external center. After completing S103, the processing method moves to S104.
[0127] In processing step S104, the risk monitoring block 140 determines whether the operational margin time ρo obtained in S102 is outside the allowable range. The allowable range that serves as the criterion for determining the operational margin time ρo can be set, for example, above the upper limit of the time ρo at which risk avoidance such as DDT rollback or downgrade is deemed necessary, or above the lower limit of the time ρo at which risk avoidance is deemed unnecessary. The allowable range of the operational margin time ρo can be set to exceed the upper limit value (which is assumed to be greater than 0). In this case, "outside the range" refers to the area between the positive and negative sides of the 0 value below the upper limit value. The allowable range of the operational margin time ρo can also be set to exceed the range of 0 values (which is assumed to be the upper limit value). In this case, "outside the range" refers to the negative side below the 0 value.
[0128] If the risk monitoring block 140 determines in S104 that the operating margin time ρo is outside the allowable range, the processing method moves to S105. On the other hand, if the risk monitoring block 140 determines in S104 that the operating margin time ρo is within the allowable range, the processing method moves to S108.
[0129] In step S105 of the processing method, the risk monitoring block 140 sets constraints on the motion control of the main vehicle 2 to enable automatic driving intervention in the manual driving of the main vehicle 2. The constraint for intervention can be an intervention command to the control block 160. In this case, by providing the control commands from the automatic driving system along with the control commands from the manual driving system from the planning block 120 to the control block 160, the control block 160 can select the control command from the automatic driving system based on the intervention command. After completing step S105, the processing method proceeds to step S106.
[0130] In processing step S106, the risk monitoring block 140 sets constraints on the motion control of the master vehicle 2 to prevent the autonomous driving master vehicle 2 from incurring unreasonable risks. The constraint for risk avoidance can be a degradation instruction to the control block 160, which is a best-effort action for the vehicle 2, executing degraded driving such as an emergency retreat or MRM (Missing Ride) to ensure continued driving control in autonomous driving. The constraint for risk avoidance can also be a restriction instruction to the control block 160 based on a safety model to move the autonomous driving master vehicle 2 to a minimum-risk state, according to a determination of a safety envelope violation. When a restriction instruction is given as a constraint, the determination of whether the operating margin time ρo is outside the allowable range is used as a determination of whether a safety envelope violation has occurred.
[0131] In S106, the allowable range of the operation margin time ρo can be set such that when the range exceeds the upper limit value greater than 0, a degradation instruction is used as a constraint for the positive side of the operation margin time ρo (greater than 0), and a restriction instruction is used as a constraint for the operation margin time ρo (below 0). This switching allows for a stricter safety restriction when the positive side of the operation margin time ρo disappears compared to the degradation instruction when the positive side retains the operation margin time ρo. After completing S106, the processing method moves to S107.
[0132] In processing step S107, the risk monitoring block 140 stores (i.e., accumulates) the evidence information output from S103, including at least one of the allowable reaction time ρp and the operational margin time ρo, in the memory 10. The memory 10 that serves as the destination for storing the evidence information can be either the same as or different from the memory 10 that stores the evidence information via S103. In different cases, the evidence information can be stored after the storage destination is changed to the memory 10 installed inside the main vehicle 2, or after the storage destination is changed to the memory 10 that becomes an external center outside the main vehicle 2. Furthermore, in different cases, by setting the interval from storage in S103 (i.e., temporary primary storage) to storage based on the change of storage destination in S107 (i.e., secondary storage) to be shorter than the storage interval described above with respect to S103, the evidence information can be reliably stored even if the power supply to the main vehicle 2 is cut off.
[0133] When the memory 10 storing evidence information in S107 is installed inside the main vehicle 2, it can be mechanically protected even if the main vehicle 2 is involved in a collision. When the memory 10 storing evidence information is installed inside the main vehicle 2, it can also be protected in a fire-resistant mounting location. When the memory 10 storing evidence information is installed inside the main vehicle 2, it can also be protected in a water-resistant mounting location. In this way, the protected memory 10, which is protected inside the main vehicle 2, can store encrypted or hashed evidence information. In the case of encrypted evidence information, the decryption key can be stored in at least one of the following: the protected memory 10 inside the main vehicle 2, the unprotected memory 10 inside the main vehicle 2, and the memory 10 at an external center, before or after a change in the storage destination. In the case of hashed evidence information, the transaction containing the hash value can be stored in at least one of the following: the protected memory 10 inside the main vehicle 2, the unprotected memory 10 inside the main vehicle 2, and the memory 10 at an external center.
[0134] By executing S107, the driver's operational history under unreasonable conditions or unreasonable risk states can be preserved as evidence. After completing S107, this phase of the processing method ends. Furthermore, unlike the preservation of evidence, during the execution of S107, the time change of the operating margin time ρo can be observed based on the evidence stored in memory 10 when a manual deviation occurs. In this case, by judging the driver's state, such as fatigue, based on the time change, the judgment result can be effectively used for, for example, planning or executing driving control, or determining a violation of the safety envelope.
[0135] like Figure 10 As shown, in S108 of the processing method where the operating margin time ρo is within the allowable range, the risk monitoring block 140 determines whether the deviation generation operation of the driver in the main vehicle 2 has ended, that is, whether the detection block 100 detects the end of the deviation generation operation. Therefore, the determination of the end of the deviation generation operation is based on the detection information in the detection block 100 of S100. If the risk monitoring block 140 determines in S108 that the deviation generation operation is still continuing, this phase of the processing method ends. On the other hand, if the risk monitoring block 140 determines in S108 that the deviation generation operation has ended, the processing method proceeds to S109.
[0136] That is, if it is determined that the operation has ended due to deviation caused by the operation margin time ρo being kept within the allowable range, processing method S109 is executed. In S109, risk monitoring block 140 updates the safe distance dmin envisioned in the safety model based on the operation margin time ρo output by S103. Here, the safe distance dmin is the distance ensured between the master vehicle 2 and the target moving body 3 according to the safety model in autonomous driving, and is calculated by the safety function L of Equation 1. Therefore, the update of the safe distance dmin can also be performed by risk monitoring block 140 adjusting or learning the parameter coefficients of the safety function L. After completing the execution of S109, the processing method moves to S110.
[0137] In step S110 of the processing method, the risk monitoring block 140 stores and maintains scenario information representing the end scenario of the deviation-causing operation in the memory 10. Scenario information can also be described as event information representing the end event of the deviation-causing operation. The scenario information is stored and maintained in association with a timestamp representing the end time of the deviation-causing operation. The storage and maintenance of scenario information can also be performed based on the aforementioned evidence information. By executing S110, the driver's operational history in scenarios where risks were avoided without autonomous driving intervention can be retained as evidence information different from at least one of the allowable reaction time ρp and the operational margin time ρo. After completing the execution of S110, this phase of the processing method ends.
[0138] Upon completion of S108 and S110, the risk monitoring block 140 may delete the evidence information stored in S103, which includes at least one of the allowable reaction time ρp and the operational margin time ρo. Upon completion of S108 and S110, the risk monitoring block 140 retains the evidence information stored in S103 in memory 10 according to S107. The risk monitoring block 140 may also overwrite the evidence information stored in S103 of the next process after the completion of S108 or S110 of this process with new evidence information.
[0139] As described above, the technology disclosed in Patent Document 1 envisions a reaction time for the host vehicle during autonomous driving. However, during manual driving, the reaction time differs from that during autonomous driving, for example, due to the individualization of driver operation. Therefore, even if the technology disclosed in Patent Document 1 is applied to manual driving, there are concerns that it may be difficult to ensure the accuracy of driving control through appropriate safety constraints in the host vehicle. Furthermore, as described above, the technology disclosed in Patent Document 1 envisions a shared reaction time for both the host vehicle and the target vehicle. However, the reaction time of the target vehicle traveling behind the host vehicle differs from that of the host vehicle, for example, depending on whether it is autonomous or manual driving, or the type of vehicle. Therefore, there are concerns that it may be difficult to ensure the accuracy of driving control in the host vehicle following the target vehicle through appropriate safety constraints as in autonomous driving.
[0140] In contrast, according to the first embodiment described above, when a manual deviation occurs in the manually driven master vehicle 2, causing the driver's manual operation to deviate from steady-state operation, an allowable reaction time ρp for the master vehicle 2 is output based on a safety model. This safety model is a model of the driving strategy and is modeled using SOTIF. Therefore, since a specific allowable reaction time ρp for the scenario in which the manual deviation occurs can be envisioned, appropriate constraints can be set for the manually driven master vehicle 2 to ensure the accuracy of driving control.
[0141] (Second Implementation)
[0142] The second embodiment is a variation of the first embodiment.
[0143] like Figure 11 As shown, in S200 of the processing method of the second embodiment, the risk monitoring block 140 determines whether the detection block 100 has detected a target vehicle 3a (hereinafter referred to as the follower vehicle 3a) that is a target moving body 3 traveling subsequently relative to the master vehicle 2 of the autonomous driving system. The detection of the follower vehicle 3a by the detection block 100 is based on data obtained from at least one of the external sensors 50 of the sensor system 5 and the V2X type communication system 6. When determining whether the follower vehicle 3a has been detected, the risk monitoring block 140 obtains detection information containing information about the follower vehicle 3a.
[0144] If the risk monitoring block 140 determines in S200 that no subsequent vehicle 3a has been detected, the current process of the processing method ends. On the other hand, if the risk monitoring block 140 determines in S200 that a subsequent vehicle 3a has been detected, the processing method moves to S201.
[0145] In processing step S201, the risk monitoring block 140 obtains the permissible reaction time ρp. This permissible reaction time ρp is the time allowed as the reaction time ρ for the subsequent vehicle 3a of the main vehicle 2, based on a safety model. This safety model is a model obtained by modeling SOTIF based on the driving strategy. In the subsequent vehicle 3a, whether it is performing autonomous driving or manual driving, the reaction time ρ means the time required for the subsequent vehicle 3a to make a complete reaction, including the driver's reaction.
[0146] The reaction time ρ of the subsequent vehicle 3a is used as a variable in Equation 1 for the safety function L, according to S101. The inverse function R of the safety function L is defined by a function or algorithm satisfying Equation 2 according to S101. Here, dr in Equation 2 is the actual distance compared with the safety distance dmin when a safety envelope violation is determined; that is, the separation distance between the primary vehicle 2 and the subsequent vehicle 3a at the time of executing S201. Based on these, the risk monitoring block 140 in S201, according to S101 and Equation 3, infers the permissible reaction time ρp of the subsequent vehicle 3a.
[0147] Figure 8The risk monitoring block 140 shown manages the state and transitions of a reasonable plan between the primary vehicle 2 and the subsequent vehicle 3a during autonomous driving. In this plan management, the risk monitoring block 140 maintains the state of the reasonable plan between the primary vehicle 2 and the subsequent vehicle 3a. Furthermore, whenever the state of the maintained plan transitions, the risk monitoring block 140 determines the appropriate response, i.e., reasonable action, for both the primary vehicle 2 and the subsequent vehicle 3a.
[0148] Therefore, in Figure 11 In S201, the risk monitoring block 140 focuses on the reasonable actions under consideration during the calculation of the allowable reaction time ρp, managing the start and end scenarios of the state transition synchronization of the plan. The start scenario for the reasonable actions under consideration can be, for example, an event requiring avoidance due to a high-importance accident risk, such as the collision risk of the following vehicle 3a while the main vehicle 2 is in a signal-stopped state. The end scenario for the reasonable actions under consideration can be either a safe end event of the reasonable plan or an event of a violation of the safety envelope. Based on these, the risk monitoring block 140 can also perform the calculation of the allowable reaction time ρp according to Equation 3 during the period of consideration for reasonable actions. After completing the execution of S201, the processing method moves to S202.
[0149] In processing step S202, the risk monitoring block 140 obtains the operational margin time ρo of the automatic operation given to the master vehicle 2 during autonomous driving, based on the allowable reaction time ρp of the subsequent vehicle 3a obtained in S201. The operational margin time ρo can also be described as the allowable surplus time for risk avoidance operations based on the safety model between the master vehicle 2 and the subsequent vehicle 3a. The risk monitoring block 140 calculates the operational margin time ρo of the master vehicle 2 relative to the subsequent vehicle 3a according to Equation 4, based on S102. Here, the action-required time ρv is set as the estimated time required to avoid the risk from the occurrence of an unreasonable situation or unreasonable risk state, according to each scheme or scenario. After completing the execution of S202, the processing method proceeds to S203.
[0150] In processing step S203, the risk monitoring block 140 outputs evidence information, including at least one of the allowable reaction time ρp obtained in S201 and the operational margin time ρo obtained in S202, to the memory 10 according to S103. In S203, the evidence information may include scenario information indicating at least one of the start and end scenarios of a reasonable action of concern. The operational margin time ρo stored in the memory 10 at the output destination via S203 can also be used to update the safety distance dmin envisioned in the safety model according to S109. After completing S203, the processing method proceeds to S204.
[0151] In processing step S204, the risk monitoring block 140 determines, based on S104, whether the operating margin time ρo obtained in S202 is outside the acceptable range. If the risk monitoring block 140 determines in S204 that the operating margin time ρo is outside the acceptable range, the processing method proceeds to S205. Conversely, if the risk monitoring block 140 determines in S204 that the operating margin time ρo is within the acceptable range, the processing method proceeds to S208.
[0152] In step S205 of the processing method, the risk monitoring block 140 treats the automatic operation that restricts the motion control of the main vehicle 2 as a risk avoidance operation and sets a risk avoidance flag in the memory 10 to indicate the risk avoidance operation. After completing the execution of S205, the processing method moves to S206.
[0153] In processing step S206, the risk monitoring block 140 sets motion control settings for the master vehicle 2 to constrain the autonomous driving master vehicle 2 from unreasonable risks to the following vehicle 3a. The constraint for risk avoidance can also be an effort-based approach for the master vehicle 2, such as avoiding collisions with the following vehicle 3a as much as possible by early deceleration or reducing deceleration. The constraint for risk avoidance can also serve as a constraint to move the autonomous driving master vehicle 2 to a minimum-risk state based on a safety model, or it can be a restriction instruction on the control block 160 based on a determination of a safety envelope violation. When a restriction instruction is given as a constraint, the determination of whether the operating margin time ρo is outside the allowable range can also be used as a determination of whether a safety envelope violation has occurred.
[0154] In S206, the allowable range of the operation margin time ρo can be set to a range exceeding the upper limit value greater than 0. For operation margin time ρo on the positive side (greater than 0), an avoidance instruction is used as a constraint; for operation margin time ρo below 0, a restriction instruction is used as a constraint. This switching allows setting a stricter safety restriction when the positive side of the operation margin time ρo disappears compared to the avoidance instruction when the positive side retains the operation margin time ρo. After completing S206, the processing method moves to S207.
[0155] In S207 of the processing method, the risk monitoring block 140 retains (i.e., accumulates) the evidence information output by S107, which includes at least one of the allowable reaction time ρp and the operational margin time ρo, in the memory 10. The evidence information retained in S207 may also include scenario information such as the start and end scenarios of the reasonable action of interest, and the start scenario indicating a risk avoidance operation. The evidence information retained in S207 may include detection information indicating the behavior of the subsequent vehicle 3a in relation to the risk avoidance operation in the main vehicle 2, such as brake light illumination information. By executing S207, the operational history of vehicles 2 and 3a in scenarios or situations of unreasonable conditions or unreasonable risk states can be retained as evidence information. After completing the execution of S207, the processing method proceeds to S208.
[0156] like Figure 11 As shown, in step S208 of the processing method where the operating margin time ρo is within the acceptable range, the risk monitoring block 140 determines whether a risk avoidance flag is set in the memory 10. If the risk monitoring block 140 determines in step S208 that the risk avoidance flag is not set, this step of the processing method ends. On the other hand, if the risk monitoring block 140 determines in step S208 that the risk avoidance flag is set, the processing method proceeds to step S209.
[0157] That is, if the operational margin time ρo, which is determined to be temporarily outside the permissible range, returns to the permissible range through risk avoidance operation, then S209 of the processing method is executed. In S209, the risk monitoring block 140 stores and maintains the scenario information indicating the end scenario of the risk avoidance operation in the memory 10 as evidence information different from at least one of the permissible reaction time ρp and the operational margin time ρo. By executing S209, the operational behavior history of vehicles 2 and 3a in the case of recovery from a scheme or scenario that has become unreasonable or an unreasonable risk state can be retained as evidence information different from at least one of the permissible reaction time ρp and the operational margin time ρo. After completing the execution of S209, the processing method moves to S210.
[0158] In step S210 of the processing method, the risk monitoring block 140 removes the risk avoidance flag from the memory 10. After completing the execution of S210, this phase of the processing method ends.
[0159] When executing S208 without setting a risk avoidance flag, and upon completion of S210, the risk monitoring block 140 may delete the evidence information stored in S103, which includes at least one of the allowable reaction time ρp and the operational margin time ρo. When executing S208 without setting a risk avoidance flag, and upon completion of S210, the risk monitoring block 140 may, according to S107 and S207, retain the evidence information stored in S103 in the memory 10. The risk monitoring block 140 may also, in S208 without setting a risk avoidance flag in this process, or in S103 of the next process after completing S210 of this process, overwrite the evidence information stored in S103 of this process with new evidence information.
[0160] According to the second embodiment described above, in the autonomous driving master vehicle 2, during the following of a subsequent vehicle 3a, which is the target moving body 3, the permissible reaction time ρp allowed by the subsequent vehicle 3a is output based on a safety model, wherein the safety model is a model based on the driving strategy and is obtained by modeling SOTIF. Therefore, since the permissible reaction time ρp specific to the following scenario of the subsequent vehicle 3a can be envisioned, appropriate constraints can be set on the autonomous driving master vehicle 2 to ensure the accuracy of driving control.
[0161] (Third Implementation)
[0162] The third embodiment is a variation of the first embodiment.
[0163] like Figure 12 As shown, in the control block 3160 of the system structure in the third embodiment, the acquisition process of determination information related to the safety envelope is omitted from the risk monitoring block 140. Therefore, the planning block 3120 of the third embodiment acquires the determination information related to the safety envelope from the risk monitoring block 140. When the planning block 3120 acquires determination information indicating no safety envelope violation, it plans the driving control of the main vehicle 2 according to the planning block 120. On the other hand, when the planning block 3120 acquires determination information indicating a safety envelope violation, it imposes constraints on the driving control based on the determination information during the stage of planning the driving control according to the planning block 120. That is, the planning block 3120 restricts the driving control to be planned. In any case, the driving control of the main vehicle 2 planned by the planning block 3120 is executed by the control block 3160.
[0164] like Figure 13As shown, in S305 of the processing method of the third embodiment, the risk monitoring block 140, in addition to executing constraints for automatic driving intervention to manual driving by issuing intervention commands to the planning block 3120, proceeds according to S105. In S306 of the processing method of the third embodiment, the risk monitoring block 140, in addition to executing constraint settings for risk avoidance by issuing downgrade commands or restriction commands to the planning block 3120, proceeds according to S106. In this third embodiment, by applying the principles of the first embodiment, appropriate constraints can be set for the manually driven master vehicle 2 to ensure the accuracy of driving control.
[0165] (Fourth Implementation)
[0166] The fourth embodiment is a variation of applying the system structure of the third embodiment to the processing method of the second embodiment.
[0167] like Figure 14 As shown, in S406 of the processing method of the fourth embodiment, the risk monitoring block 140, in addition to executing the constraint setting for risk avoidance by issuing avoidance or restriction instructions to the planning block 3120, according to S206. In this fourth embodiment, by applying the principles of the second embodiment, appropriate constraints can be set for the autonomous driving master vehicle 2 to ensure the accuracy of driving control.
[0168] (Fifth Implementation)
[0169] The fifth embodiment is a variation of the first embodiment.
[0170] like Figure 15 As shown, in the control block 5160 of the system structure in the fifth embodiment, the processing of obtaining determination information related to the safety envelope from the risk monitoring block 5140 is omitted. Therefore, the risk monitoring block 5140 of the fifth embodiment obtains information representing the result of the driving control performed on the main vehicle 2 by the control block 5160. The risk monitoring block 5140 evaluates the driving control by performing a safety determination based on the safety envelope on the result of the driving control.
[0171] like Figure 16 As shown, in S505 of the processing method of the fifth embodiment, the risk monitoring block 5140 evaluates situations where constraints need to be set for automatic driving intervention in manual driving, according to S105. In S506 of the processing method of the fifth embodiment, the risk monitoring block 5140 evaluates situations where constraints need to be set to avoid risks, according to S106. In this fifth embodiment, driving control can be evaluated based on constraints appropriately set for the manually driven master vehicle 2 according to the principles of the first embodiment, ensuring the accuracy of driving control.
[0172] (Sixth Implementation Method)
[0173] The sixth embodiment is a variation of applying the system structure of the fifth embodiment to the processing method of the second embodiment.
[0174] like Figure 17 As shown, in S606 of the processing method of the sixth embodiment, the risk monitoring block 5140 evaluates situations where constraints need to be set to avoid risks, according to S206. In this sixth embodiment, driving control can be evaluated based on constraints appropriately set for the autonomous driving master vehicle 2 according to the principles of the second embodiment, ensuring the accuracy of driving control.
[0175] (Seventh Implementation)
[0176] The seventh embodiment is a variation of the fifth or sixth embodiment.
[0177] like Figure 18 , 19 As shown, in the system structure of the seventh embodiment, a test block 7180 is added. This test block 7180 is used to test the driving control of the control block 160 for purposes such as safety approval. The test block 7180 is given the functions of the detection block 100 and the risk monitoring block 5140. The test block 7180 can also be... Figure 18 The processing system 1 shown executes test programs that are appended to the processing programs used to build blocks 100, 120, 5140, and 160. Test block 7180 can also be built by, as shown in... Figure 19 The test processing system 7001, which is different from the processing system 1, executes a test processing program that is different from the processing program used to build each of blocks 100, 120, 5140, and 160. Here, the test processing system 7001 can also be composed of at least one dedicated computer connected to the processing system 1 for testing driving control (illustration of the connection via communication system 6 is omitted), and has a memory 10 and a processor 12.
[0178] In the processing method of the seventh embodiment, the test block 7180 replaces the risk monitoring block 5140 or performs the steps of the processing method of the fifth or sixth embodiment based on the risk monitoring block 5140. Figure 18 , Figure 19 The diagram illustrating the path through which test block 7180 acquires detection information is omitted. In this seventh embodiment, driving control can be evaluated based on appropriately set constraints on the manually driven master vehicle 2 according to the principles of the first embodiment, ensuring the accuracy of driving control; or driving control can be evaluated based on appropriately set constraints on the automatically driven master vehicle 2 according to the principles of the second embodiment, ensuring the accuracy of driving control.
[0179] (Eighth Implementation Method)
[0180] The eighth embodiment is a variation of the third embodiment.
[0181] like Figure 20 As shown, in the planning block 8120 of the eighth embodiment, the functions of the risk monitoring block 140 are incorporated as a risk monitoring sub-block 8140. Therefore, when the planning block 8120 of the eighth embodiment obtains determination information such as no safety envelope violation through the risk monitoring sub-block 8140, it plans the driving control of the main vehicle 2 according to the planning block 120. On the other hand, when the risk monitoring sub-block 8140 obtains determination information such as a safety envelope violation, the planning block 8120 imposes constraints on the driving control based on the determination information during the stage of planning the driving control according to the planning block 120. That is, the planning block 8120 restricts the driving control to be planned. In any case, the driving control of the main vehicle 2 planned by the planning block 8120 is executed by the control block 3160.
[0182] like Figure 21 As shown, in S805 of the processing method of the eighth embodiment, the risk monitoring sub-block 8140, in addition to executing the intervention plan in the planning block 8120 to impose constraints for automatic driving intervention in manual driving, follows S105 as described in the first embodiment. In S806 of the processing method of the eighth embodiment, the risk monitoring sub-block 8140, in addition to executing the setting of constraints for risk avoidance through the degradation plan or restriction plan in the planning block 8120, follows S106 as described in the first embodiment. In this eighth embodiment, by following the principles of the first embodiment, appropriate restrictions can be set on the manually driven master vehicle 2 to ensure the accuracy of driving control.
[0183] (Ninth Implementation)
[0184] The ninth embodiment is a variation of applying the system structure of the eighth embodiment to the processing method of the second embodiment.
[0185] like Figure 22 As shown, in S906 of the processing method of the ninth embodiment, the risk monitoring sub-block 8140, in addition to executing the constraint setting for risk avoidance through the avoidance plan or restriction plan in the planning block 3120, follows S206 as described in the second embodiment. In this ninth embodiment, based on the principle of the second embodiment, appropriate constraints can be set for the autonomous driving master vehicle 2 to ensure the accuracy of driving control.
[0186] (Other implementation methods)
[0187] The above describes several embodiments, but this disclosure should not be construed as limiting it to these embodiments. It can be applied to various embodiments and combinations without departing from the spirit of this disclosure.
[0188] In a variation, the dedicated computer constituting processing system 1 may also include at least one of digital circuitry and analog circuitry as a processor. The digital circuitry referred to here includes, for example, at least one of ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), SOC (System on a Chip), PGA (Programmable Gate Array), and CPLD (Complex Programmable Logic Device). Furthermore, such digital circuitry may also have a memory storing programs.
[0189] The processing method of the modified example can also be limited to S100-S103 and S107. In the processing method of the modified example, the execution of at least S106, S306, S506, and S806 can be omitted respectively. The processing method of the modified example can also be limited to S200-S203 and S207. In the processing method of the modified example, the execution of S206, S406, S606, and S906 can be omitted respectively. In the processing method of the modified example, the execution of S108-S110 can be omitted. In the processing method of the modified example, the execution of S205 and S208-S210 can be omitted.
[0190] In addition to the above description, the above-described embodiments and variations can also be implemented as devices that can be mounted on a main mobile body and have at least one processor 12 and one memory 10, and can be implemented as processing circuits (e.g., processing ECUs) or semiconductor devices (e.g., semiconductor wafers).
Claims
1. A processing method, executed by a processor for performing processing related to driving control of a primary mobile body, comprising: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operating margin time is outside the allowable range, at least one of the aforementioned allowable response time and the aforementioned operating margin time that has been output is stored.
2. The processing method according to claim 1, wherein, It also includes: when it is determined that the above-mentioned operation margin time is within the allowable range and the above-mentioned manual operation that gave the above-mentioned manual deviation has ended, updating the safe distance between the main moving body and the target moving body in autonomous driving based on the above-mentioned operation margin time.
3. A processing method, executed by a processor for performing processing related to driving control of a primary mobile body, comprising: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Output operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body. If it is determined that the aforementioned operational margin time is outside the permissible range, a constraint is set on the aforementioned driving control to allow the automatic driving system to intervene in the aforementioned manual driving of the main moving body; and If it is determined that the aforementioned operational margin time is outside the permissible range, the aforementioned driving control is set with constraints to prevent the aforementioned main moving body from unreasonable risks during autonomous driving.
4. The processing method according to claim 3, wherein, It also includes: when it is determined that the above-mentioned operational margin time has disappeared, setting constraints on the above-mentioned driving control to enable the above-mentioned main moving body of the autonomous driving system to move to the minimum risk state based on the above-mentioned safety model.
5. The processing method according to claim 3 or 4, wherein, It also includes: when it is determined that the above-mentioned operation margin time is within the allowable range and the above-mentioned manual operation that gave the above-mentioned manual deviation has ended, updating the safe distance between the main moving body and the target moving body in autonomous driving based on the above-mentioned operation margin time.
6. A processing method, executed by a processor for performing processing related to driving control of a primary mobile body, comprising: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operational margin time remains within the allowable range and the aforementioned manual operation that gave the aforementioned manual deviation has ended, the safe distance between the main moving body and the target moving body in autonomous driving, based on the aforementioned operational margin time, is updated.
7. A processing system comprising a processor for performing processing related to driving control of a primary moving body. The processor described above is configured to perform the following processes: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operating margin time is outside the allowable range, at least one of the aforementioned allowable response time and the aforementioned operating margin time that has been output is stored.
8. A processing system comprising a processor for performing processing related to driving control of a primary moving body. The processor described above is configured to perform the following processes: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Output operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body. If it is determined that the aforementioned operational margin time is outside the permissible range, a constraint is set on the aforementioned driving control to allow the automatic driving system to intervene in the aforementioned manual driving of the main moving body; and If it is determined that the aforementioned operational margin time is outside the permissible range, the aforementioned driving control is set with constraints to prevent the aforementioned main moving body from unreasonable risks during autonomous driving.
9. A processing system comprising a processor for performing processing related to driving control of a primary moving body. The processor described above is configured to perform the following processes: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operational margin time remains within the allowable range and the aforementioned manual operation that gave the aforementioned manual deviation has ended, the safe distance between the main moving body and the target moving body in autonomous driving, based on the aforementioned operational margin time, is updated.
10. A storage medium for a processing program, the processing program being stored in the storage medium and comprising instructions for execution by a processor to perform processing related to driving control of a primary mobile body. The above instructions cause the processor to perform the following processing: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operating margin time is outside the allowable range, at least one of the aforementioned allowable response time and the aforementioned operating margin time that has been output is stored.
11. A storage medium for a processing program, the processing program being stored in the storage medium and comprising instructions for execution by a processor to perform processing related to driving control of a primary mobile body. The above instructions cause the processor to perform the following processing: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Output operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body. If it is determined that the aforementioned operational margin time is outside the permissible range, a constraint is set on the aforementioned driving control to allow the automatic driving system to intervene in the aforementioned manual driving of the main moving body; and If it is determined that the aforementioned operational margin time is outside the permissible range, the aforementioned driving control is set with constraints to prevent the aforementioned main moving body from unreasonable risks during autonomous driving.
12. A storage medium for a processing program, the processing program being stored in the storage medium and comprising instructions for execution by a processor to perform processing related to driving control of a primary mobile body. The above instructions cause the processor to perform the following processing: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operational margin time remains within the allowable range and the aforementioned manual operation that gave the aforementioned manual deviation has ended, the safe distance between the main moving body and the target moving body in autonomous driving, based on the aforementioned operational margin time, is updated.
13. A processing method, executed by a processor for performing processing related to driving control of a primary mobile body, comprising: Detecting the target moving body that subsequently travels relative to the aforementioned primary moving body in autonomous driving; and Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
14. The processing method according to claim 13, wherein, It also includes an output operating margin time, which is obtained based on the allowable reaction time and is given to the automatic operation in the above-mentioned automatic driving in the main moving body.
15. The processing method according to claim 14, wherein, It also includes: when it is determined that the above-mentioned operating margin time is outside the allowable range, maintaining the storage of at least one of the above-mentioned allowable response time and the above-mentioned operating margin time that has been output.
16. The processing method according to claim 14 or 15, wherein, It also includes: when it is determined that the above-mentioned operating margin time is outside the allowable range, setting constraints on the above-mentioned driving control to prevent the above-mentioned main moving body in autonomous driving from unreasonable risks.
17. The processing method according to claim 16, wherein, It also includes: when it is determined that the above-mentioned operational margin time has disappeared, setting constraints on the above-mentioned driving control to enable the above-mentioned main moving body of the autonomous driving system to move to the minimum risk state based on the above-mentioned safety model.
18. A processing system comprising a processor for performing processing related to driving control of a primary moving body. The processor described above is configured to perform the following processes: Detecting the target moving body that subsequently travels relative to the aforementioned primary moving body in autonomous driving; and Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
19. A storage medium for a processing program, the processing program being stored in the storage medium and comprising instructions for execution by a processor to perform processing related to driving control of a primary mobile body. The above instructions include: Detecting the target moving body that subsequently travels relative to the aforementioned primary moving body in autonomous driving; and Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
20. A storage medium, wherein, In the processing method according to claim 1 or 15, if it is determined that the above-mentioned operating margin time is outside the allowable range, the storage medium retains storage of at least one of the above-mentioned allowable response time and the above-mentioned operating margin time that has been output.
21. A processing apparatus comprising a processor configured to be mounted on a main mobile body and to perform processing related to driving control of the main mobile body. The processor described above is configured to perform the following processes: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operating margin time is outside the allowable range, at least one of the aforementioned allowable response time and the aforementioned operating margin time that has been output is stored.
22. A processing apparatus comprising a processor configured to be mounted on a main mobile body and to perform processing related to driving control of the main mobile body. The processor described above is configured to perform the following processes: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Output operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body. If it is determined that the aforementioned operational margin time is outside the permissible range, a constraint is set on the aforementioned driving control to allow the automatic driving system to intervene in the aforementioned manual driving of the main moving body; and If it is determined that the aforementioned operational margin time is outside the permissible range, the aforementioned driving control is set with constraints to prevent the aforementioned main moving body from unreasonable risks during autonomous driving.
23. A processing apparatus comprising a processor configured to be mounted on a main mobile body and to perform processing related to driving control of the main mobile body. The processor described above is configured to perform the following processes: In the aforementioned manually driven main moving body, the manual deviation of the driver's manual operation from the steady-state operation is detected; Output the allowable reaction time, which is the time allowed for the main moving body to react during the generation of the manual deviation, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function. Outputting an operating margin time, which is obtained based on the allowable reaction time and is given to the driver's manual operation in the main moving body; and If it is determined that the aforementioned operational margin time remains within the allowable range and the aforementioned manual operation that gave the aforementioned manual deviation has ended, the safe distance between the main moving body and the target moving body in autonomous driving, based on the aforementioned operational margin time, is updated.
24. A processing apparatus comprising a processor and configured to be mounted on a main mobile body to perform processing related to driving control of the main mobile body. The processor described above is configured to perform the following processes: Detecting the target moving body that subsequently travels relative to the aforementioned primary moving body in autonomous driving; and Output the permissible reaction time, which is the time allowed for the target moving body to react, obtained based on a safety model. The safety model is a model based on the driving strategy and is obtained by modeling the safety of the expected function.
Citation Information
Patent Citations
Work clothing with cooling function
JP2021017658A
Vehicle and road collaborative information-based signalized intersection vehicle speed guidance system and method
CN105160917A
Lane deviation preventing device
JP2001310719A
