control system

Abstract

To solve the problem that when log data related to vehicle control is stored in an on-vehicle memory device in a vehicle control system, a situation where a capacity of the memory device is tight and necessary log data cannot be stored is generated.SOLUTION: A control system includes: one or a plurality of processors for performing vehicle control including a plurality of control functions; and one or a plurality of memory devices. The one or the plurality of processors execute processing of storing log data related to vehicle control in the one or the plurality of memory devices during executing vehicle control, and processing of limiting the actuation of one or a plurality of object control functions selected from a plurality of control functions when a vacant capacity of the one or the plurality of memory devices is less than a threshold value.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to a control system mounted on a vehicle.

Background Art

[0002] In recent years, a technology for performing vehicle control using a machine learning model has been considered. Patent Document 1 discloses a method for collecting training data that can be used for training a machine learning model regarding automatic driving control using the machine learning model. In addition, as documents showing the technical level of this technical field, there are the following Patent Documents 2 or 3.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a control system that performs vehicle control, it may be required to verify the content of vehicle control retrospectively. As a method for retrospectively verifying vehicle control, it has been considered to store log data related to vehicle control in an in-vehicle storage device. However, since the capacity of the in-vehicle storage device is limited, the capacity may become tight depending on the situation of vehicle control. In particular, when a machine learning model is used for vehicle control, it is expected that the amount of log data to be stored will be relatively large. If the capacity of the storage device becomes tight, there is a risk that necessary log data cannot be stored.

[0005] One of the purposes of this disclosure is to address the above-mentioned issues and to provide a technology that can suppress the strain on storage capacity. [Means for solving the problem]

[0006] One aspect of this disclosure relates to a control system installed in a vehicle. The control system comprises one or more processors that perform vehicle control including a plurality of control functions, and one or more storage devices. The one or more processors are further configured to perform the following: saving log data related to vehicle control to one or more storage devices while vehicle control is being performed, and restricting the operation of one or more target control functions selected from the plurality of control functions when the free capacity of one or more storage devices falls below a threshold. [Effects of the Invention]

[0007] According to this disclosure, when the free space of the storage device falls below a threshold, the operation of one or more target control functions selected from multiple control functions is restricted. This reduces the amount of log data stored, and consequently, prevents storage device capacity strain. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows an example configuration related to the automatic driving control of a vehicle according to the embodiment. [Figure 2] This figure shows an example of the hardware configuration of the automated driving system according to the present invention. [Figure 3] This is a flowchart showing the processing performed by the processor according to the embodiment. [Figure 4] This figure shows an example of the amount of data generated. [Figure 5] This figure shows an example of a case in which the operation of one or more target control functions is restricted in stages depending on the available capacity. [Modes for carrying out the invention]

[0009] The embodiments will be described below with reference to the drawings.

[0010] 1. Autonomous driving system This embodiment relates to an automated driving system that performs automated driving control of a vehicle. Figure 1 is a diagram showing an example configuration related to automated driving control of a vehicle 1 by the automated driving system according to this embodiment. Automated driving means that at least one of the following actions of the vehicle 1—steering, acceleration, and deceleration—is performed automatically without driver operation by an operator. Automated driving control is a concept that includes not only fully automated driving control, but also risk avoidance control, lane keeping assist control, etc. The operator may be a driver riding in the vehicle 1, or a remote operator who remotely controls the vehicle 1.

[0011] Vehicle 1 includes a sensor group 10, a recognition unit 20, a planning unit 30, a control quantity calculation unit 40, and a driving device 50.

[0012] The sensor group 10 includes recognition sensors 11 used to recognize the surrounding conditions of the vehicle 1. Examples of recognition sensors 11 include cameras, LIDAR (Laser Imaging Detection and Ranging), radar, etc. The sensor group 10 may further include state sensors 12 for detecting the state of the vehicle 1, position sensors 13 for detecting the position of the vehicle 1, etc. Examples of state sensors 12 include speed sensors, acceleration sensors, yaw rate sensors, steering angle sensors, etc. An example of a position sensor 13 is a GNSS (Global Navigation Satellite System) sensor.

[0013] Sensor detection information SEN is information obtained by the sensor group 10. For example, sensor detection information SEN includes image data captured by the camera. Alternatively, sensor detection information SEN may include information (relative position, relative speed, shape, etc.) about specific objects (pedestrians, preceding vehicles, white lines, bicycles, road signs, etc.) that appear in the image. For example, sensor detection information SEN may also include point cloud data obtained by LIDAR. For example, sensor detection information SEN may also include information on the relative position and relative speed of an object detected by radar. Sensor detection information SEN may also include vehicle state information indicating the state of vehicle 1. Sensor detection information SEN may also include position information indicating the location of vehicle 1.

[0014] The recognition unit 20 receives sensor detection information SEN. Based on the information obtained by the recognition sensor 11, the recognition unit 20 recognizes the situation around vehicle 1. For example, the recognition unit 20 recognizes the positions of objects around vehicle 1 on a spatial map. Examples of objects include pedestrians, other vehicles (preceding vehicles, parked vehicles, etc.), white lines, road structures (e.g., guardrails, curbs), fallen objects, traffic lights, intersections, signs, etc. Furthermore, the recognition unit 20 may also predict the behavior of objects around vehicle 1. The recognition unit 20 may also generate a risk map of the area around vehicle 1. The recognition result information RES indicates the recognition result by the recognition unit 20.

[0015] The planning unit 30 receives recognition result information RES from the recognition unit 20. The planning unit 30 may also receive vehicle status information, location information, and pre-generated map information. The map information may be high-precision 3D map information. Based on the received information, the planning unit 30 generates a driving plan for vehicle 1. The driving plan may be for reaching a pre-set destination or for avoiding risks. The driving plan includes driving decisions such as maintaining the current lane, changing lanes, overtaking, turning left or right, steering, accelerating, decelerating, and stopping. Furthermore, the planning unit 30 generates a target trajectory TRJ necessary for vehicle 1 to drive according to the driving plan. The target trajectory TRJ includes a target position and a target speed.

[0016] The control quantity calculation unit 40 receives the target trajectory TRJ from the planning unit 30. The control quantity calculation unit 40 calculates the control quantity CON required for the vehicle 1 to follow the target trajectory TRJ. The control quantity CON can also be described as the control quantity required to reduce the deviation between the vehicle 1 and the target trajectory TRJ. The control quantity CON includes at least one of the steering control quantity, drive control quantity, and braking control quantity. Examples of steering control quantities include target steering angle, target torque, target motor angle, target motor drive current, etc. Examples of drive control quantities include target drive force, target engine torque, etc. Examples of braking control quantities include target braking force, target braking torque, etc.

[0017] The traveling device 50 includes a steering device 51, a driving device 52, and a braking device 53. The steering device 51 steers the wheels. For example, the steering device 51 includes an electric power steering (EPS) device. The driving device 52 is a power source that generates a driving force. Examples of the driving device 52 include an engine, an electric motor, an in-wheel motor, etc. The braking device 53 generates a braking force. The traveling device 50 receives a control amount CON from the control amount calculation unit 40. The traveling device 50 operates the steering device 51, the driving device 52, and the braking device 53 according to the steering control amount, the driving control amount, and the braking control amount, respectively. Thereby, the vehicle 1 travels so as to follow the target trajectory TRJ.

[0018] The recognition unit 20 includes at least one of a rule-based model and a machine learning model. The rule-based model performs recognition processing based on a predetermined rule group. Examples of the machine learning model include a neural network (NN), a support vector machine (SVM), a regression model, a decision tree model, etc. The NN may be a convolutional neural network (CNN), a recurrent neural network (RNN), or a combination thereof. The type of each layer, the number of layers, and the number of nodes in the NN are arbitrary. The machine learning model is generated in advance through machine learning. The recognition unit 20 performs recognition processing by inputting the sensor detection information SEN into the model. The recognition result information RES is output from the model or generated based on the output from the model.

[0019] Similarly, the planning unit 30 includes at least one of a rule-based model and a machine learning model. The planning unit 30 performs planning processing by inputting the recognition result information RES into the model. The target trajectory TRJ is output from the model or generated based on the output from the model.

[0020] Similarly, the control amount calculation unit 40 includes at least one of a rule-based model and a machine learning model. The control amount calculation unit 40 performs control amount calculation processing by inputting the target trajectory TRJ into the model. The control amount CON is output from the model or generated based on the output from the model.

[0021] Two or more of the recognition unit 20, the planning unit 30, and the control amount calculation unit 40 may be integrally configured. All of the recognition unit 20, the planning unit 30, and the control amount calculation unit 40 may be integrally configured (End-to-End configuration). For example, the recognition unit 20 and the planning unit 30 may be integrally configured by a neural network (NN) that outputs the target trajectory TRJ from the sensor detection information SEN. Even in the case of an integrated configuration, intermediate products such as the recognition result information RES and the target trajectory TRJ may be output. For example, when the recognition unit 20 and the planning unit 30 are integrally configured by an NN, the recognition result information RES may be the output of the intermediate layer of the NN.

[0022] The recognition unit 20, the planning unit 30, and the control amount calculation unit 40 constitute an "automatic driving control unit" that controls the automatic driving of the vehicle 1. The automatic driving control unit performs automatic driving control of the vehicle 1 so as to achieve a plurality of control functions. That is, the automatic driving control by the automatic driving control unit includes a plurality of control functions. For example, as the plurality of control functions, there are cruise control, lane keep, lane change, pre-crash safety, risk avoidance control, overtaking of low-speed vehicles, recognition of road signs and control according to the recognition result, recognition of signals and control according to the recognition result, and the like.

[0023] The automatic driving control by the automatic driving control unit can consider various patterns depending on the control functions to be implemented. For example, it is possible to consider an automatic driving control unit in which all of the above control functions are implemented, or an automatic driving control unit in which only cruise control and lane keep are implemented.

[0024] In the automated driving control unit, the recognition unit 20, the planning unit 30, and the control quantity calculation unit 40 each have configurations corresponding to the implemented control functions. For example, the rule-based model is given conditional statements and function expressions corresponding to the implemented control functions. Also, for example, the machine learning model performs machine learning corresponding to the implemented control functions.

[0025] Figure 2 shows an example of the hardware configuration of the automated driving system 100 according to this embodiment. The automated driving system 100 has at least the functions of the automated driving control unit described above. The automated driving system 100 may further include a sensor group 10 and a driving device 50.

[0026] The autonomous driving system 100 includes one or more processors 110 (hereinafter simply referred to as processor 110) and one or more storage devices 120 (hereinafter simply referred to as storage devices 120).

[0027] The processor 110 performs various processes. The processor 110 can be composed of, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), etc. The recognition unit 20, the planning unit 30, and the control variable calculation unit 40 may be implemented by a single processor 110 or by separate processors 110. The storage device 120 stores various information necessary for the execution of processes by the processor 110. The storage device 120 can be composed of, for example, a recording medium such as ROM (Read Only Memory), RAM (Random Access Memory), HDD (Hard Disk Drive), or SSD (Solid State Drive).

[0028] The storage device 120 stores the computer program 121, model data 122, and log data LOG.

[0029] The computer program 121 is executed by the processor 110. Various processes by the autonomous driving system 100 may be realized through the cooperation of the processor 110 executing the computer program 121 and the storage device 120. The computer program 121 may be recorded on a computer-readable recording medium.

[0030] Model data 122 is model data included in the recognition unit 20, the planning unit 30, and the control variable calculation unit 40. When the processor 110 executes processing related to automatic driving control, it selects and uses a model from the model data 122 to configure the recognition unit 20, the planning unit 30, and the control variable calculation unit 40.

[0031] While executing automated driving control, the processor 110 collects log data LOG related to the automated driving control. The processor 110 stores the collected log data LOG in the storage device 120. The stored log data LOG is expected to be used for verifying the automated driving control. The log data LOG may include sensor detection information SEN input to the automated driving control unit. The log data LOG may include control quantity CON output from the automated driving control unit. The log data LOG may include recognition result information RES output from the recognition unit 20. The log data LOG may include target trajectory TRJ output from the planning unit 30. The log data LOG may include the reason for the decision made in the recognition process by the recognition unit 20. The log data LOG may include the reason for the decision made in the planning process by the planning unit 30. The log data LOG may include whether or not there was operator intervention in the automated driving control.

[0032] The automated driving system 100 according to this embodiment is configured to allow modification of the contents of multiple control functions implemented in the automated driving control unit. In particular, the automated driving system 100 is configured to allow restriction of the operation of each control function. Restriction here includes not only turning off the operation of a control function, but also reducing the frequency and range of operation of a control function, or reducing the content of a control function.

[0033] For example, when an automated driving control unit is configured to implement cruise control, lane keeping, and lane changing, the automated driving system 100 is configured to allow the operation of lane changing to be turned off, so that the automated driving control unit does not perform lane changes.

[0034] The processor 110 may change the content of multiple control functions implemented in the autonomous driving control unit by, for example, switching the model selected from the model data 122. Alternatively, the processor 110 may change the content of multiple control functions implemented in the autonomous driving control unit by changing the parameters included in the model.

[0035] 2. Restrictions on the operation of control functions As described above, in the autonomous driving system 100, the processor 110 stores log data LOG related to autonomous driving control in the storage device 120 for post-event verification of autonomous driving control. However, in order to properly verify autonomous driving control in the event of any incident, it is desirable that all of the stored log data LOG be retained until a single process is completed. However, the storage device 120 provided in the vehicle 1 has a limited capacity. Therefore, depending on the status of autonomous driving control, the capacity may become strained. If the capacity becomes strained, there is a risk that it may not be possible to save the necessary log data LOG.

[0036] Therefore, in the automated driving system 100 according to this embodiment, when the free capacity of the storage device 120 falls below a predetermined threshold, the processor 110 restricts the operation of one or more control functions (hereinafter referred to as "one or more target control functions") among the multiple control functions implemented in the automated driving control unit.

[0037] The amount of data collected in the log data LOG during the execution of automated driving control correlates with the content of the control functions being operated. This is because data associated with the execution of processing related to each control function is collected as log data LOG. In particular, the amount of data collected in the log data LOG is thought to increase as the number of operating control functions increases and as the content of the operating control functions becomes more complex. Therefore, by restricting the operation of some of the control functions among the multiple control functions implemented in the automated driving control unit, the amount of data collected and stored in the log data LOG can be reduced. Based on the above viewpoint, the automated driving system 100 according to this embodiment makes it possible to suppress the strain on the capacity of the storage device 120.

[0038] 2-1. Processing Figure 3 is a flowchart showing an example of the processing performed by the processor 110 regarding the restriction of the operation of one or more target control functions. The processing shown in the flowchart in Figure 3 is repeatedly executed at a predetermined processing cycle, for example, while automatic driving control is being performed.

[0039] In step S110, the processor 110 obtains free space in the storage device 120.

[0040] Next, in step S120, the processor 110 determines whether the free space of the storage device 120 falls below a predetermined threshold. The threshold is set to, for example, a value at which it is expected that log data LOG will no longer be able to be saved during the journey to the destination. The processor 110 may also be configured to change the threshold depending on the expected driving time, driving environment, etc.

[0041] If the available space is above the threshold (step S120; No), the processor 110 terminates the current process without restricting the operation of the control function. If the available space is below the threshold (step S120; Yes), the process proceeds to step S130.

[0042] In step S130, the processor 110 selects one or more target control functions from among multiple control functions. From various perspectives, several embodiments can be considered regarding how to select one or more target control functions. Embodiments relating to the selection of one or more target control functions will be described later.

[0043] Next, in step S140, the processor 110 restricts one or more selected target control functions. For example, if lane change is selected as one or more target control functions, the processor 110 turns off the operation of lane change. Alternatively, the processor 110 reduces the frequency of lane change operation. Alternatively, the processor 110 reduces the functions related to lane change determination so that control is initiated only when the operator makes a decision to perform a lane change (e.g., the operator operates the turn ramp). The processor 110 may also be configured to notify the operator that one or more target control functions have been restricted.

[0044] By restricting the operation of one or more target control functions in this way, the amount of log data LOG associated with the execution of processing related to one or more target control functions can be reduced. For example, by restricting the operation of one or more target control functions, the amount of sensor detection information SEN and recognition result information RES used by the automatic driving control unit can be reduced. Also, for example, information related to the reasons for decisions in the recognition processing by the recognition unit 20 and the planning processing by the planning unit 30 can be reduced.

[0045] 2-2. Selection of Target Control Function The following describes embodiments relating to the selection of one or more target control functions.

[0046] The first embodiment involves selecting one or more target control functions based on the amount of log data LOG generated per unit time by each of the multiple control functions implemented in the automatic driving control unit (hereinafter referred to as "amount of generated data"). Typically, in the first embodiment, the processor 110 selects one or more control functions with a large amount of generated data from among the multiple control functions as target control functions. The number of target control functions may be determined, for example, based on the desired reduction in the amount of generated data and the expected reduction in the amount of generated data by restricting the operation.

[0047] Figure 4 is a diagram illustrating an example of the amount of generated data. Figure 4 shows a list of multiple control functions implemented in the automatic driving control unit, and a bar graph showing the magnitude of the amount of generated data for each control function. According to the first embodiment, for example, the processor 110 selects risk avoidance control, slow vehicle overtaking, and lane keeping as one or more target control functions, which have a large amount of generated data. Here, in order to ensure the functionality of the automatic driving control at a certain level, there may be control functions that are not selected as one or more target control functions regardless of the amount of generated data. For example, cruise control, lane keeping, and pre-collision safety are set as control functions that are not selected as one or more target control functions. In this case, for example, the processor 110 selects risk avoidance control, slow vehicle overtaking, and lane change as one or more target control functions.

[0048] According to the first embodiment, the amount of data stored in the log data LOG can be effectively reduced.

[0049] The second embodiment involves selecting one or more target control functions based on the necessity for the driving environment of vehicle 1. Examples of the driving environment of vehicle 1 include the type of road on which vehicle 1 travels (whether it is a public road or an expressway, etc.), the number of lanes on the road on which vehicle 1 travels, and the number of objects around vehicle 1.

[0050] In the second embodiment, the processor 110 selects one or more control functions from among a plurality of control functions that are of low necessity for the driving environment of the vehicle 1 as target control functions. For example, when the vehicle 1 is driving on a highway, risk avoidance control and signal recognition and control according to the recognition result are considered to be of low necessity. Therefore, in this case, for example, the processor 110 selects risk avoidance control and recognition and control according to the recognition result as one or more target control functions. Also, for example, when the vehicle 1 is driving on a single-lane road, lane changes and overtaking of slow-moving vehicles are considered to be of low necessity. Therefore, in this case, for example, the processor 110 selects lane changes and overtaking of slow-moving vehicles as one or more target control functions. The processor 110 can be configured to select one or more target control functions by using, for example, an array that defines control functions according to the necessity for the driving environment of the vehicle 1.

[0051] According to the second embodiment, the amount of log data LOG can be reduced while maintaining the functionality of the automatic driving control unit as much as possible.

[0052] The third embodiment involves selecting one or more target control functions based on the need for the operator's driving characteristics of the vehicle 1. Examples of the operator's driving characteristics include tendencies in how they direct their awareness of the vehicle 1's surroundings (e.g., high awareness of left and right, low awareness of rear), and preferences for driving operations (e.g., dislike of overtaking or lane changes). The processor 110 may be configured to acquire the operator's driving characteristics from information obtained from a monitor camera (e.g., the operator's gaze and face direction). Alternatively, data managing the operator's driving characteristics may be stored in the storage device 120.

[0053] In a third embodiment, the processor 110 selects one or more control functions from among a plurality of control functions that are of low necessity for the operator's driving characteristics as target control functions. For example, if the operator does not like overtaking, then overtaking slow-moving vehicles is considered to be of low necessity. Therefore, in this case, for example, the processor 110 selects overtaking slow-moving vehicles as one or more target control functions. The processor 110 can be configured to select one or more target control functions by using, for example, an array that defines control functions according to the operator's needs for their driving characteristics.

[0054] According to the third embodiment, the amount of log data (LOG) can be reduced while minimizing the increase in the burden on the operator.

[0055] The above embodiments can also be combined. For example, the processor 110 may select one or more target control functions based on both the amount of data generated and the need for the driving environment of vehicle 1.

[0056] 2-3. Limiting the operation of target control functions according to available capacity The processor 110 may be configured to restrict the operation of one or more selected target control functions in stages according to the free space in the storage device 120. In particular, the processor 110 may be configured to restrict the operation of one or more target control functions in order from those with a large amount of generated data as the free space decreases.

[0057] Figure 5 shows an example of restricting the operation of control functions in order from those with the largest generated data volume. In Figure 5, lane changes and overtaking of slow vehicles are selected as one or more target control functions. The amount of generated data related to overtaking of slow vehicles is larger than the amount of generated data related to lane changes. In this case, as shown in Figure 5, the processor 110 may be configured to restrict the operation of overtaking of slow vehicles at time T1 when the available capacity falls below a threshold, and to further restrict the operation of lane changes at time T2 when the available capacity has decreased further.

[0058] By gradually restricting the operation of one or more target control functions according to available capacity, the control functions can be maintained to the extent that the shortage of available capacity can be suppressed.

[0059] The processor 110 may also perform incremental restrictions on the operation of a single target control function. For example, if lane change is selected as one or more target control functions, the processor 110 may be configured to first reduce the frequency of lane change operation when the available capacity falls below a threshold, and then turn off lane change operation when the available capacity decreases further.

[0060] 3. Effects As described above, according to this embodiment, when the free capacity of the storage device 120 falls below a predetermined threshold, the operation of one or more target control functions selected from the multiple control functions is restricted. This reduces the amount of log data LOG collected and stored. Consequently, it is possible to prevent the storage device 120 from becoming overloaded.

[0061] 4. Other control systems This embodiment can also be applied to other control systems mounted on the vehicle 1. For example, a driver assistance system can be considered another control system that performs vehicle control including multiple control functions. A driver assistance system is a system that performs vehicle control including control functions such as collision damage mitigation braking, inter-vehicle distance control, lane keeping, and road sign recognition. Similarly, when applied to a driver assistance system, when the free capacity of the storage device 120 falls below a predetermined threshold, the operation of one or more target control functions selected from the multiple control functions can be restricted. This makes it possible to achieve the same effect as described above. [Explanation of symbols]

[0062] 1 vehicle, 100 autonomous driving systems, 110 processors, 120 memory devices, LOG Log data

Claims

1. One or more processors that perform vehicle control including multiple control functions, One or more storage devices, Equipped with, The one or more processors further include: During the execution of the vehicle control, a process is performed to save log data related to the vehicle control to the one or more storage devices. When the free space of the one or more storage devices falls below a threshold, a process is performed to restrict the operation of one or more target control functions selected from the multiple control functions. Configured to perform In the process of restricting the operation of the one or more target control functions, the one or more processors With respect to each of the control functions of the plurality of control functions, the amount of generated data, which is the amount of log data generated per unit time, is obtained. Select one or more target control functions based on the size of the generated data. Control system.

2. The control system according to Claim 1, In the process of restricting the operation of the one or more target control functions, the one or more processors As the available capacity decreases, the operation of the control functions among the one or more target control functions will be restricted in order from the control function with the largest amount of generated data. Characterized by Control system.

Images