JP6957440B2 - Controls, control methods and programs - Google Patents

Description

The present invention relates to control devices, control methods and programs.

Technology related to automatic driving of vehicles is known.

For example, one of the first control signal generated based on the automatic driving control information and the second control signal generated based on the relative information between the own vehicle and the surrounding objects is output to the drive device. A technique is known in which a second control signal is output to a drive device instead of the first control signal when an abnormality is detected in the automatic operation control information.

Japanese Unexamined Patent Publication No. 2017-47694

In automatic driving, it is conceivable to adopt a rule-based control system that operates according to predetermined rules. In the case of a rule-based control system, it is possible to construct the control system relatively easily.
However, when a rule-based control system is adopted and processing exceeding the processing capacity is requested, appropriate control may not be in time.

For example, when the input is an external state quantity including spatial information or a large number of conditions, the amount of calculation is large, so that a certain amount of calculation time is required for control. Therefore, the calculation may not be completed within a predetermined time, or even if the calculation is completed, the control may not be in time.
In one aspect, the present invention aims at more reliable control of the controlled object.

A disclosed control device is provided to achieve the above object. This control device stores a second control signal capable of controlling a controlled object by filtering a given state quantity in a storage unit that stores rules prepared in advance for the given state quantity. A generator that performs calculations based on rules and generates a control target at a higher speed than a control target can be transmitted, and a signal that controls the control target based on at least one of the first control signal and the second control signal is transmitted. It has a transmitter and.

In one aspect, the control target can be controlled more reliably.

It is a block diagram which shows the function of the vehicle of embodiment. It is a block diagram which shows the function of the control part of embodiment. It is a flowchart explaining the process of a control part. It is a flowchart explaining the 1st control signal generation processing. It is a flowchart explaining the 2nd control signal generation processing. It is a flowchart explaining the control signal update process. It is a figure which shows an example of the hardware composition of the control part of embodiment.

Hereinafter, the control device of the embodiment will be described in detail with reference to the drawings.
<Embodiment>
FIG. 1 is a block diagram showing a function of the vehicle according to the embodiment.

The vehicle 1 of the first embodiment has sensors 2a and 2b, a drive unit (control target: Actuator) 3a and 3b, and a control unit 4. The vehicle 1 is not particularly limited, and examples thereof include those traveling on a road such as a passenger car and a truck.
The number of sensors and drive units is not limited to those shown in the figure, and may be one or three or more. Further, the control unit may have a multi-stage configuration or a parallel configuration.

Sensors 2a and 2b include, for example, LiDAR (Laser Imaging Detection and Ranging), sonar (SONAR: Sound navigation and ranging) using millimeter waves and ultrasonic waves, GPS, a sensor that detects an external state quantity such as a camera, and a vehicle speed. The one that detects the internal state of the vehicle 1, such as the steering angle and the steering angle, and the like can be mentioned.

The signals detected by these sensors 2a and 2b include, for example, the accelerator opening of the vehicle 1, the opening of the brake, the on / off of the side brake, the steering angle, the current flowing through the steering motor, and the shift position (R / N /). D), presence / absence of door lock, tire angle, axle speed, axle acceleration, vehicle body speed, vehicle body acceleration, vehicle body angular speed, and the like.

Examples of the drive units 3a and 3b include those including a motor, an actuator, and the like and related to the operation of the vehicle 1. For example, brakes, steering motors, shift positions (R / N / D), door locks, drive motors and the like can be mentioned.

The control unit 4 is, for example, an aggregate of a plurality of control circuits. The control unit 4 calculates a given state quantity based on the rules stored in the storage unit and generates a first control signal for controlling the control target. Further, the control unit 4 uses the first control signal to obtain a second control signal that can control the control target at a lower calculation cost than the first control signal generation, such as performing a filter process on a given state quantity. Is also generated at high speed.

The control unit 4 controls the drive units 3a and 3b by properly using or combining the generated first control signal and second control signal. For example, it is possible to perform control such as braking and decelerating the vehicle 1 using a second control signal generated faster than the first control signal, and then determining an appropriate braking amount using the first control signal. Become. Further, it is also possible to integrate the first control signal and the second control signal according to the situation to more appropriately control the drive units 3a and 3b.
Hereinafter, the disclosed control unit 4 will be described in more detail.
FIG. 2 is a block diagram showing a function of the control unit according to the embodiment.

The control unit 4 includes a sensor control unit 41, a rule control unit (first control signal generation unit) 42, an inference engine 43, a feature amount extraction unit (second control signal generation unit) 44, and the like. It has a drive control unit 45. The functions of each of these parts may be formed by one integrated circuit, or some functions may be collectively formed by one integrated circuit.

The sensor control unit 41 receives a signal (Knowledge Data) detected by the sensors 2a and 2b according to a user's instruction or under a predetermined situation. The sensor control unit 41 sends the received detection signal to the rule control unit 42 and the feature amount extraction unit 44.

The rule control unit 42 calculates the detection signal given from the sensor control unit 41 based on the rules stored in advance, and generates a first control signal for controlling the vehicle 1.

The reference engine 43 refers to the information stored in the knowledge database (not shown) that stores the signals detected by the sensors 2a and 2b, the control signals generated by the rule control unit 42 and the feature quantity extraction unit 44, and the like. Then, in consideration of the functional characteristics of the entire system, the internal state quantity and the external state quantity of the system at a certain point in time are comprehensively specified and the state is recognized. The inference engine 43 infers and selects an action necessary for achieving the goal based on the difference or relationship between the recognized state and the goal set by the user's instruction.

Further, the inference engine 43 receives the setting of the destination desired by the user, and based on the inference engine 43 and the planning unit (not shown) that generates the route information (route) to be traveled, the current rules. Generate an update rule to update. The inference engine 43 sends an update rule to the rule control unit 42.

In addition, the inference engine 43 generates an update filter that updates the current filter. The inference engine 43 sends an update filter to the feature amount extraction unit 44.

The feature amount extraction unit 44 may be realized by, for example, an FPGA (field-programmable gate array). The feature quantity extraction unit 44 performs filtering processing on the given state quantity, and generates a second control signal capable of controlling the vehicle 1 at a speed higher than that of the rule control unit 42. When generating the second control signal, the feature amount extraction unit 44 can take various methods such as waveform approximation of the input signal, noise, and a periodic approach.

Here, when the sensor control unit 41 sends a signal to the rule control unit 42 and the feature amount extraction unit 44, the rule control unit 42 and the feature amount extraction unit 44 use their respective methods to control the first control signal and the second control. Start signal generation. As described above, as a result, the second control signal is generated faster than the first control signal. As a result, for example, it is possible to control the vehicle 1 by first braking and decelerating, and then determining an appropriate braking amount.

The drive control unit 45 is an example of a transmission unit, and generates and transmits a signal for controlling the drive units 3a and 3b based on at least one of a first control signal and a second control signal. As an example, the drive control unit 45 regards each signal of the first control signal and the second control signal as one vector, and uses the vector for each predetermined element such as presence / absence of control, control method, control amount, and the like. Disassemble into.

Then, the drive control unit 45 evaluates, selects, weights, adds and subtracts each decomposed element, integrates the first control signal and the second control signal, and generates a control signal for controlling the drive units 3a and 3b. ..

Examples of the (circuit) configuration of the drive control unit 45 include a vector decomposer (2), an input side multiplier (matching the number of vectors after decomposition = a predetermined number of elements × 2), and an adder (addition). A combination of a combination of an output side multiplier (matching the number of vectors after addition = a predetermined number of elements) and a vector synthesizer (1) can be mentioned. ..
The drive control unit 45 sends a control signal to each drive unit 3a and 3b.

Next, the processing of the control unit 4 will be described with reference to a flowchart. The procedure shown in the flowchart below is an example, and a part of the procedure may be omitted, another procedure may be inserted, a part of the procedure may be replaced, or a plurality of processes may be executed in parallel.
FIG. 3 is a flowchart illustrating the processing of the control unit.

[Step S1] The sensor control unit 41 accepts an input of the automatic driving permission signal "ON" according to a user's instruction or under a predetermined situation. In the state of the automatic driving permission signal "ON", the vehicle 1 maintains, for example, the speed of the vehicle 1 and controls the distance between vehicles.

[Step S2] The sensor control unit 41 determines whether the automatic driving permission signal is “on” or “off”. If the automatic driving permission signal is "ON" (Yes in step S2), the process proceeds to step S3. If the automatic driving permission signal is "off" (No in step S2), the process shown in FIG. 3 is terminated.

[Step S3] The sensor control unit 41 receives the signals detected by the sensors 2a and 2b and sends them to a knowledge database (not shown). After that, the transition to step S4 and step S5 occurs. The processes of steps S4 and S5 can be executed in parallel.

[Step S4] The rule control unit 42 calculates the detection signal given from the sensor control unit 41 based on the rules stored in advance, and generates a first control signal for controlling the vehicle 1. After that, the process proceeds to step S6.

[Step S5] The feature quantity extraction unit 44 performs a filter process on the given state quantity, and generates a second control signal capable of controlling the vehicle 1 at a speed higher than that of the rule control unit 42. Further, the drive control unit 45 updates the control signal according to the situation. After that, the process proceeds to step S6.

[Step S6] The drive control unit 45 attempts to generate a drive signal by arbitrating the first control signal and the second control signal by using the method described above. Examples of the arbitration referred to here include selecting one of a plurality of signals, and using the function of the drive control unit 45 to integrate the plurality of signals to generate a single signal. ..

[Step S7] The drive control unit 45 determines whether or not the arbitration was successful in step S6. If the arbitration is successful (Yes in step S7), the process proceeds to step S8. If the arbitration fails (No in step S7), the process proceeds to step S2.

[Step S8] The drive control unit 45 generates knowledge database update information for updating the knowledge database. The drive control unit 45 sends the generated knowledge database update information to the knowledge database. After that, the process proceeds to step S9.

[Step S9] The drive control unit 45 executes a control signal update process for updating the control signals for controlling the drive units 3a and 3b using the control signal generated in step S6. The control signal update process will be described in detail later.
This is the end of the description of the process of FIG.
Next, the process of generating the first control signal (first control signal generation process) in step S4 will be described.
FIG. 4 is a flowchart illustrating the first control signal generation process.

[Step S4a] The rule control unit 42 performs rule matching. Specifically, the rule control unit 42 selects an option (the only solution) in which the state quantity given by the sensor control unit 41 only matches, based on the rule given by the inference engine 43.

[Step S4b] The rule control unit 42 determines whether or not the only solution has been found. When the only solution is found (Yes in step S4b), the only solution is set as the first control signal, and the first control signal generation process is terminated. If the only solution cannot be found (No in step S4b), the process proceeds to step S4c.
[Step S4c] The rule control unit 42 sends an escalation and rule update request to the inference engine 43.
As an example of updating the rule, it is possible to generate a stop signal, perform a steering operation, and the like.
[Step S4d] The rule control unit 42 sets the answer prepared in advance as the first control signal. After that, the first control signal generation process is terminated.
[Step S4e] The rule control unit 42 generates a control signal for updating the control state of the drive units 3a and 3b based on the only solution.
Next, the process of generating the second control signal in step S5 of FIG. 3 and the process of updating the control signal according to the situation (second control signal generation update process) will be described.
FIG. 5 is a flowchart illustrating the second control signal generation update process.
[Step S5a] The feature amount extraction unit 44 filters the input detection signal.
[Step S5b] The feature amount extraction unit 44 determines whether or not a significant feature amount (sign) can be extracted from the signal filtered in step S5a.

As significant features, for example, an object appeared at high speed in front of the vehicle 1, the information detected by the sensors 2a and 2b showed an abnormal value, and the state of the vehicle 1 deviated from the safety control range, ideally. A large deviation from the sensor state was discovered (for example, sudden spin due to an external factor).

If a significant feature amount can be extracted (Yes in step S5b), the process proceeds to step S5c. When a significant feature amount cannot be extracted (No in step S5b), the second control signal generation process is terminated.

[Step S5c] The feature amount extraction unit 44 generates a control signal for updating the control state of the drive units 3a and 3b. Unlike rule-based systems and pattern matching, even if the only correct answer cannot be found, it operates in parallel for each feature and generates a control signal. For example, a control signal is generated that causes a gradual deceleration without waiting for the result of object recognition. This increases the avoidability.
In addition, steps S5a to S5c may be executed as an integrated step by a logic circuit or the like.

[Step S5d] The feature amount extraction unit 44 generates knowledge database update information for updating the knowledge database. The feature amount extraction unit 44 sends the generated knowledge database update information to the knowledge database. Further, the feature amount extraction unit 44 sends the generated control signal to the drive control unit 45. After that, the process proceeds to step S5e. The process of step S6 can be executed in parallel with the process of step S5e.
[Step S5e] The drive control unit 45 executes a control signal update process for updating the control signal. The control signal update process will be described in detail later.
This is the end of the description of the process of FIG.
Next, the control signal update process (control signal update process) executed by the drive control unit 45 in steps S5e and S9 will be described with reference to the flowchart.
FIG. 6 is a flowchart illustrating the control signal update process.

[Step S9a] The drive control unit 45 receives control signals from the rule control unit 42 or the feature amount extraction unit 44 to the drive units 3a and 3b. The control signal from the rule control unit 42 is generated by the process of step S8 described above. The control signal from the feature amount extraction unit 44 is generated by the process of step S5d described above.
[Step S9b] The drive control unit 45 sends a control signal to each drive unit 3a and 3b. As a result, each of the drive units 3a and 3b drives according to the control signal.

As described above, according to the control unit 4, the drive units 3a and 3b perform calculations based on the knowledge database that stores the rules prepared in advance and the rules stored in the knowledge database for the given state quantity. The rule control unit 42 that generates the first control signal that can control the above, and the second control signal that can control the drive units 3a and 3b by filtering the given state quantity are faster than the first control signal. It has a feature amount extraction unit 44 to be generated.

More specifically, in step S5d, the feature amount extraction unit 44 generates knowledge database update information and control signals. The feature amount extraction unit 44 sends the generated knowledge database update information to the knowledge database. Further, the feature amount extraction unit 44 sends the generated control signal to the drive control unit 45. At this stage, the control signal can be updated at high speed, so it is possible to handle processing in which speed is prioritized over accuracy.

Then, in step S8, the knowledge database update information and the control signal are generated again. As a result, the processing can be accurately handled. As a result, the initial operation of the control can be made faster than in the case where the vehicle 1 is controlled using only the first control signal. Therefore, the probability of success of the action can be increased.

The processing performed by the control unit 4 may be distributed processing by a plurality of devices. For example, one device may generate the first control signal and the second control signal, and the other device may integrate the first control signal and the second control signal.

The control device, control method, and program of the present invention have been described above based on the illustrated embodiment, but the present invention is not limited thereto, and the configuration of each part is arbitrary having the same function. It can be replaced with the one of the configuration. Further, any other constituents and processes may be added to the present invention.
For example, in the present embodiment, the control unit 4 is realized by using a control circuit. However, the present invention is not limited to this, and the control unit 4 may be realized by a computer having the same function.
FIG. 7 is a diagram showing an example of the hardware configuration of the control unit of the embodiment.

The entire device of the control unit 4 is controlled by the CPU (Central Processing Unit) 101. A RAM (Random Access Memory) 102 and a plurality of peripheral devices are connected to the CPU 101 via a bus 105.

The RAM 102 is used as the main storage device of the control unit 4. The RAM 102 temporarily stores at least a part of an OS (Operating System) program or an application program to be executed by the CPU 101. Further, the RAM 102 stores various data used for processing by the CPU 101.
A built-in memory 103 and a communication interface 104 are connected to the bus 105.

The built-in memory 103 writes and reads data. The built-in memory 103 is used as a secondary storage device of the mobile terminal device 2. The OS program, application program, and various data are stored in the built-in memory 103. Examples of the built-in memory include a semiconductor storage device such as a flash memory.

The communication interface 104 is connected to the network 50. The communication interface 104 transmits / receives data to / from another computer or communication device via the network 50. The communication interface 104 may be omitted.
With the hardware configuration as described above, the processing function of the present embodiment can be realized.
Further, the present invention may be a combination of any two or more configurations (features) of the above-described embodiments.

The above processing function can be realized by a computer. In that case, a program that describes the processing content of the function of the control unit 4 is provided. By executing the program on a computer, the above processing function is realized on the computer. The program describing the processing content can be recorded on a computer-readable recording medium. Examples of the recording medium that can be read by a computer include a magnetic storage device, an optical disk, an optical magnetic recording medium, a semiconductor memory, and the like. Examples of the magnetic storage device include a hard disk drive, a flexible disk (FD), and a magnetic tape. Examples of the optical disk include DVD, DVD-RAM, and CD-ROM / RW. Examples of the magneto-optical recording medium include MO (Magneto-Optical disk).

When a program is distributed, for example, a portable recording medium such as a DVD or a CD-ROM on which the program is recorded is sold. It is also possible to store the program in the storage device of the server computer and transfer the program from the server computer to another computer via the network.

The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes the processing according to the program. The computer can also read the program directly from the portable recording medium and execute the processing according to the program. In addition, the computer can sequentially execute processing according to the received program each time the program is transferred from the server computer connected via the network.

Further, at least a part of the above processing functions can be realized by an electronic circuit such as a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or a PLD (Programmable Logic Device).

1 Vehicle 2a, 2b Sensor 3a, 3b Drive unit 4 Control unit 41 Sensor control unit 42 Rule control unit 43 Difference engine 44 Feature extraction unit 45 Drive control unit

Claims (6)

The operation based on rules stored in the storage unit for storing a prepared rule faster than the first control signal capable of controlling the line have control object with respect to given state quantity, the state quantity to A generator that generates a second control signal that can control the control target by performing a filter process that extracts significant features.
A transmission unit that controls the control target based on the second control signal and then transmits a signal for controlling the control target based on the first control signal.
A control device characterized by having. A claim that the transmitter decomposes the first control signal and the second control signal into predetermined control elements, compares each element, and integrates the first control signal and the second control signal. The control device according to 1. A first control signal capable of controlling the control target is generated by performing an operation based on a rule stored in the storage unit for a given state quantity.
A second control signal capable of controlling the control target is generated at a higher speed than the first control signal by performing a filter process for extracting a significant feature amount with respect to the state quantity.
A control method characterized by that. On the computer
The operation based on rules stored in the storage unit for storing a prepared rule faster than the first control signal capable of controlling the line have control object with respect to given state quantity, the state quantity to A second control signal that can control the control target is generated by performing a filtering process to extract a significant feature amount.
A program characterized by executing processing. The operation based on rules stored in the storage unit for storing a prepared rule faster than the first control signal capable of controlling the line have control object with respect to given state quantity, the state quantity to A control device having a generation unit that generates a second control signal capable of controlling a control target by performing a filter process for extracting a significant feature amount. The control device according to claim 1, wherein the generation unit determines whether or not a significant feature amount can be extracted by performing a filtering process, and generates the second control signal when the significant feature amount can be extracted.

Images