JP7437904B2 - Steering wheel operation prediction method and steering wheel operation prediction device - Google Patents

Description

The present invention relates to a steering wheel operation prediction method and a steering wheel operation prediction device.

BACKGROUND ART Conventionally, an invention has been known in which biological information (potential signal) of a user is acquired using electrodes and a device is controlled based on the obtained potential signal (Patent Document 1). The invention described in Patent Document 1 identifies the electrode where a part of the user's arm is located vertically downward from among a plurality of electrodes, and outputs a potential signal even if all of the plurality of electrodes are not in close contact with the arm. measure.

JP2016-214907A

In the invention described in Patent Document 1, although a potential signal can be measured even if all of the plurality of electrodes are not in close contact with the arm, at least one electrode needs to be installed on the arm of the occupant. However, it may be difficult to install electrodes on a part of the occupant's body, and in this case, it is not possible to obtain an electric potential signal of the occupant.

The present invention has been made in view of the above problems, and its purpose is to obtain the potential signal of the occupant without installing an electrode on a part of the occupant's body, and to use the obtained potential signal to detect the potential of the occupant. It is an object of the present invention to provide a steering wheel operation prediction method and a steering wheel operation prediction device for predicting steering wheel operations.

A steering wheel operation prediction method according to one aspect of the present invention includes detecting a potential signal in the palm of an occupant generated by muscle activity when the occupant operates the steering wheel using a sensor provided in the steering wheel , and detecting the detected potential signal. The system extracts a pre-operation signal that occurs before the occupant operates the steering wheel, and predicts the occupant's steering wheel operation based on the extracted pre-operation signal.

According to the present invention, it is possible to acquire an electric potential signal of an occupant without installing an electrode on a part of the occupant's body, and to predict the operation of the steering wheel by the occupant using the acquired electric potential signal.

FIG. 1 is a block diagram of an operation prediction device according to a first embodiment of the present invention. FIG. 2 is a diagram illustrating a steering wheel according to the first embodiment of the present invention. FIG. 3 is a graph illustrating the relationship between myoelectric potential and lane change timing according to the first embodiment of the present invention. FIG. 4 is a graph illustrating the relationship between lane change timing and TPR according to the first embodiment of the present invention. FIG. 5 is a diagram illustrating a lane change to avoid an obstacle. FIG. 6 is another graph illustrating the relationship between myoelectric potential and lane change timing according to the first embodiment of the present invention. FIG. 7 is another graph illustrating the relationship between lane change timing and TPR according to the first embodiment of the present invention. FIG. 8 is a flowchart illustrating an example of the operation of the operation prediction device according to the first embodiment of the present invention. FIG. 9 is a block diagram of an operation prediction device according to a second embodiment of the present invention. FIG. 10A is a diagram illustrating the sensitivity time according to the second embodiment of the present invention. FIG. 10B is a diagram illustrating the sensitivity time according to the second embodiment of the present invention. FIG. 11 is a flowchart illustrating an example of the operation of the operation prediction device according to the second embodiment of the present invention.

Embodiments of the present invention will be described below with reference to the drawings. In the description of the drawings, the same parts are denoted by the same reference numerals and the description thereof will be omitted.

(First embodiment)
(Example of configuration of operation prediction device)

A configuration example of the operation prediction device according to the first embodiment will be described with reference to FIGS. 1 and 2. The operation prediction device is mounted on the vehicle.

As shown in FIG. 1, the operation prediction device includes a palm potential detection section 10, a controller 20, a storage device 30, a surrounding information acquisition section 40, and an automatic function device 50.

The palm potential detection unit 10 detects the potential (potential signal) of the palm of an occupant (mainly a driver) riding in a vehicle. The palm potential detection unit 10 is, for example, a sensor including an electrode, and a plurality of palm potential detection units 10 are provided in the steering wheel 11 (see FIG. 2). The palm potential detection unit 10 detects a potential signal of the palm of the occupant when the occupant grips the steering wheel 11 . In other words, when the occupant's palm comes into close contact with the steering wheel 11, the potential signal of the occupant's palm is detected.

The palm potential detection unit 10 detects a potential signal related to the grip of the steering wheel 11 itself, and also detects a potential signal of the occupant's palm caused by muscle activity when the occupant operates the steering wheel 11. In the example shown in FIG. 2, the palm potential detection unit 10 is provided at the gripping position of the steering wheel 11 (as an example, the position from 10:10 to 9:15), but the present invention is not limited thereto. The palm potential detection unit 10 may be provided over the entire outer periphery of the steering wheel 11.

Although the palm potential detection unit 10 has been described as a sensor including electrodes, such a sensor may also include an amplifier, an AD converter, etc. in addition to the electrodes. In this case, the palm potential detection unit 10 amplifies the potential signal detected by the electrode with an amplifier, and then performs signal processing with an AD converter. After signal processing, the palm potential detection unit 10 may transmit the potential signal to the controller 20.

The storage device 30 is a storage medium such as an HDD (Hard Disk Drive). In the first embodiment, the storage device 30 stores a steering operation database 31 and a steering operation timing database 32, but is not limited thereto. The storage device 30 may store various programs, analysis results by the potential analysis section 21, and the like.

The controller 20 is a general-purpose microcomputer that includes a CPU (central processing unit), memory, and an input/output section. A computer program for functioning as an operation prediction device is installed in the microcomputer. By executing the computer program, the microcomputer functions as a plurality of information processing circuits included in the operation prediction device. Note that here we will show an example in which multiple information processing circuits included in the operation prediction device are realized by software, but of course, dedicated hardware for executing each of the information processes shown below is prepared and the information processing circuits are implemented using software. It is also possible to configure Further, the plurality of information processing circuits may be configured by individual hardware. The controller 20 includes a potential analysis section 21, a steering operation prediction section 22, an operation timing prediction section 23, and an automatic function control section 24 as a plurality of information processing circuits.

The potential analysis unit 21 analyzes the potential signal detected by the palm potential detection unit 10. Specifically, the potential analysis unit 21 analyzes the frequency, wavelength, voltage, etc. of the potential signal. Filtering is an example of an analysis method. The potential analysis unit 21 performs analysis such as extracting a specific signal by filtering the potential signal with a bandpass filter of a predetermined frequency band. The potential analysis section 21 outputs the analysis results to the steering operation prediction section 22 and the operation timing prediction section 23.

The steering operation prediction unit 22 determines whether the driver intends to operate the steering wheel 11 based on the potential signal analyzed by the potential analysis unit 21 (specific signal extracted by the potential analysis unit 21). Predict. Specifically, the steering operation prediction unit 22 refers to the steering operation database 31 and determines whether the specific signal extracted by the potential analysis unit 21 matches the data stored in the steering operation database 31. . If the specific signal extracted by the potential analysis unit 21 matches the data stored in the steering operation database 31, the steering operation prediction unit 22 predicts that the driver intends to operate the steering wheel 11. The steering operation prediction section 22 outputs the prediction result to the automatic function control section 24 .

The operation timing prediction unit 23 predicts the time until the steering wheel 11 is actually operated when the steering operation prediction unit 22 predicts that the driver intends to operate the steering wheel 11 . In other words, the operation timing prediction section 23 predicts how many seconds or milliseconds after the specific signal is extracted by the potential analysis section 21, when the steering wheel 11 will actually be operated. This prediction is realized by referring to the steering operation timing database 32. The operation timing prediction unit 23 outputs the prediction result to the automatic function control unit 24.

The automatic function control section 24 outputs a command to the automatic function device 50 based on the prediction results by the steering operation prediction section 22 and the operation timing prediction section 23.

The steering operation database 31 cumulatively stores data regarding potential signals for each driver. Specifically, the specific signal extracted by the potential analysis unit 21 is associated with the actual operation of the steering wheel 11 by the driver, and this correspondence data is stored cumulatively.

Similarly to the steering operation database 31, the steering operation timing database 32 cumulatively stores data regarding potential signals for each driver. However, unlike the steering operation database 31, the steering operation timing database 32 has a correspondence between the specific signal extracted by the potential analysis unit 21 and the timing when the steering wheel 11 is actually operated by the driver. Corresponding data is stored cumulatively. In other words, the steering operation timing database 32 includes the time from when the driver's intention to operate the steering wheel 11 is extracted until the steering wheel 11 is actually operated, and the specific signal extracted by the potential analysis unit 21. are stored in association with each other.

The surrounding information acquisition unit 40 obtains information about the surroundings of the vehicle. The surrounding information acquisition unit 40 includes, for example, a laser range finder that measures distance using light waves, a camera that acquires still images and moving images, and a sonar that measures distance using ultrasonic waves. The surrounding information acquisition unit 40 collects objects surrounding the vehicle (pedestrians, bicycles, motorcycles, other vehicles, etc.) and information about the surroundings of the vehicle (road boundaries, stop lines, traffic lights, signs, crosswalks, intersections, etc.). ) to obtain. The surrounding information acquisition unit 40 outputs the acquired data to the automatic function device 50.

The automatic function device 50 is, for example, a device having an ACC (Adaptive Cruise Control) function, and assists the driving operation of the vehicle. Since the ACC function is a well-known technology, detailed explanation will be omitted. The automatic function device 50 uses commands acquired from the automatic function control section 24 and data acquired from the surrounding information acquisition section 40 to assist in driving the vehicle. Further, the automatic function device 50 may have a lane keep function, an automatic lane change function, etc. in addition to the ACC function. Further, the automatic function device 50 may have an automatic driving function.

Next, an example of potential signal processing by the potential analysis section 21 will be described with reference to FIGS. 3 and 4. As described above, the potential analysis unit 21 uses filtering to process the potential signal.

FIG. 3 shows an analysis result obtained by filtering the potential signal detected by the palm potential detection unit 10 with a bandpass filter (low-pass filter) in a low frequency band of 1 to 6 Hz. The analysis result shown in FIG. 3 is an analysis result of a potential signal of a certain driver (hereinafter referred to as driver A). The vertical axis in FIG. 3 shows myoelectric potential (index), and the horizontal axis shows time. Time 0 in FIG. 3 indicates the timing at which driver A changes lanes (hereinafter referred to as lane change timing). A time before 0 means before a lane change is performed, and a time after 0 means after a lane change is performed.

As shown in FIG. 3, a predetermined potential signal was extracted approximately 500 msec before the lane change timing. This potential signal is a pre-operation signal that is generated before the driver A operates the steering wheel 11. The pre-operation signal is a myoelectric potential signal generated immediately before the driver A makes a muscle movement to operate the steering wheel 11. This myoelectric potential signal is considered to be a precursor operation (muscle movement) in preparation for a lane change. When driver A performs a lane change, it is thought that he performs a preliminary operation in preparation for the lane change, such as by strongly gripping the steering wheel 11 in advance, and a myoelectric potential signal caused by this preliminary operation appears approximately 500 msec before. It is thought that he did. If the myoelectric potential signal resulting from the pioneering operation is extracted approximately 500 msec before the lane change timing, it is predicted that driver A will change lanes approximately 500 msec later.

Here, the generation timing of approximately 500 msec before is the generation timing of driver A's myoelectric potential signal. Naturally, there are individual differences in the timing of occurrence. Characteristics of myoelectric potential signals such as amplitude, voltage, frequency, and generation timing of myoelectric potential signals caused by preemptive maneuvers in preparation for a lane change for each driver are stored in the steering operation database 31 in association with each driver. be done. The steering operation prediction unit 22 identifies the driver in advance, and then compares the analysis result by the potential analysis unit 21 with the steering operation database 31 to accurately generate a specific signal (myoelectric potential signal caused by the preceding operation). It becomes possible to extract. Thereby, the steering operation prediction unit 22 can accurately predict whether the driver intends to operate the steering wheel 11 or not. Note that a well-known technique such as facial recognition technique may be used as a method for identifying the driver.

If a myoelectric potential signal caused by a pioneering maneuver is extracted, it means that there is a high possibility that a lane change will actually occur after a predetermined time (approximately 500 msec in the example shown in Figure 3). The inventor has verified. The verification results are shown in Figure 4. The vertical axis in FIG. 4 indicates TPR (true positive ratio), and the horizontal axis indicates time. The TPR value increases as the ratio of predicted lane changes to actual lane changes increases. As is clear from FIG. 4, when a myoelectric potential signal resulting from a precursor operation in preparation for a lane change is extracted, it is found that there is a high possibility that a lane change will actually be performed after a predetermined time.

Next, with reference to FIG. 5, an explanation will be given of an automatic lane change performed by the automatic function device 50 when a myoelectric potential signal caused by a pioneering operation is extracted. Auto lane change is a lane change that occurs automatically.

The scene shown in FIG. 5 shows a scene in which the vehicle 70 changes lanes to avoid the motorcycle 80 (an obstacle). More specifically, the scene shown in the comparative example of FIG. 5 shows a scene in which the driver manually changes lanes at time t=0.

The example of FIG. 5 shows that the myoelectric potential signal caused by the precursor operation was extracted about 500 msec before time t=0, as described in FIG. 3. The driver of the vehicle 70 in the embodiment of FIG. 5 is the driver A described above. The automatic function control unit 24 outputs an operation command for the steering wheel 11 for automatic lane change to the automatic function device 50 when the myoelectric potential signal resulting from the pioneering operation is extracted. At this time, the automatic function control unit 24 determines the operation timing of the steering wheel 11 for automatic lane change, and also outputs this information to the automatic function device 50. Further, the automatic function control unit 24 calculates the amount of operation of the steering wheel 11 necessary for automatic lane change, and outputs this information to the automatic function device 50 as well.

The method for determining the operation timing of the steering wheel 11 for an automatic lane change is not particularly limited, but as an example, the automatic function control unit 24 determines the timing from when a myoelectric potential signal caused by a preceding operation is extracted until when a lane change is actually performed. It can be arbitrarily determined between approximately 500 msec and time t=0. In the embodiment of FIG. 5, as an example, the operation timing of the steering wheel 11 for automatic lane change is determined approximately 400 msec before time t=0.

In the example shown in FIG. 5, since the myoelectric potential signal caused by the advance maneuver is extracted, it is predicted that there is a high possibility that a lane change will actually be performed after a predetermined time (about 500 msec in the example shown in FIG. 5). Therefore, the automatic function control unit 24 determines the operation timing of the steering wheel 11 for automatic lane change, and outputs the determined information to the automatic function device 50. The automatic function device 50 that has acquired the information controls the operation of the steering wheel 11 approximately 400 msec before time t=0 to start an automatic lane change. This makes it possible to provide early support in line with driver A's operating intentions, and gain the trust of driver A. Furthermore, since it is possible to avoid the motorcycle 80 earlier than in the comparative example, the driver A can also feel secure.

Although a low-pass filter has been described as a band-pass filter in FIGS. 3 and 4 above, filtering is not limited to a low-pass filter. For example, as shown in FIGS. 6-7, a high-pass filter may be used as the band-pass filter.

FIG. 6 shows an analysis result obtained by filtering the myoelectric potential signal detected by the palm potential detection unit 10 excluding frequencies below 5 Hz using a band pass filter (high pass filter) in a high frequency band of 20 to 90 Hz. The analysis result shown in FIG. 6 is the analysis result of the myoelectric potential signal of driver A described above. However, the driving scenes are different between FIG. 3 and FIG. 6. The vertical axis in FIG. 6 shows myoelectric potential (index), and the horizontal axis shows time. Time 0 in FIG. 6 indicates the lane change timing by driver A.

As shown in FIG. 6, since the predetermined myoelectric potential signal was extracted approximately 100 msec before the lane change timing, it is predicted that driver A will change lanes approximately 100 msec later. Other explanations are the same as those in FIG. 3, so details will be omitted, but even when filtering is performed with a high-pass filter, the same results as when filtering with a low-pass filter are obtained.

The inventor also verified the significance of filtering with a high-pass filter. The verification results are shown in FIG. The vertical axis in FIG. 7 shows TPR, and the horizontal axis shows time. As is clear from FIG. 7, when a myoelectric potential signal resulting from a precursor operation in preparation for a lane change is extracted, it is found that there is a high possibility that a lane change will actually be performed after a predetermined time.

Next, an example of the operation of the operation prediction device according to the first embodiment will be described with reference to the flowchart in FIG. 8 .

In step S101, the palm potential detection unit 10 detects a potential signal of the driver's palm when the driver grips the steering wheel 11. The process proceeds to step S103, where the potential analysis unit 21 analyzes the potential signal detected by the palm potential detection unit 10.

The process proceeds to step S105, where the steering operation prediction unit 22 refers to the steering operation database 31 and determines whether the specific signal extracted by the potential analysis unit 21 matches the data stored in the steering operation database 31. judge. If the specific signal extracted by the potential analysis unit 21 matches the data stored in the steering operation database 31, the steering operation prediction unit 22 predicts that the driver intends to operate the steering wheel 11.

The process proceeds to step S107, and when the steering operation prediction unit 22 predicts that the driver intends to operate the steering wheel 11, the operation timing prediction unit 23 calculates the time until the steering wheel 11 is actually operated. Predict.

The process proceeds to step S109, where the automatic function control unit 24 determines the operation timing of the steering wheel 11 for automatic lane change from when the myoelectric potential signal caused by the pioneering operation is extracted until when the lane change is actually performed. decide. The process proceeds to step S111, where the automatic function control unit 24 calculates the amount of operation of the steering wheel 11 necessary for automatic lane change.

The process proceeds to step S113, where the automatic function device 50, which has acquired the operation timing of the steering wheel 11 for automatic lane change and the operation amount of the steering wheel 11 necessary for automatic lane change, operates the steering wheel 11 at the specified timing. Controls operations and initiates automatic lane change.

(effect)
As explained above, according to the operation prediction device according to the first embodiment, the following effects can be obtained.

The palm potential detection unit 10 is provided in the steering wheel 11 and detects a potential signal of the palm of the occupant when the occupant grips the steering wheel 11. The potential analysis unit 21 extracts a pre-operation signal generated before the operation of the steering wheel 11 by the occupant from the potential signal detected by the palm potential detection unit 10. The steering operation prediction unit 22 predicts the operation of the steering wheel 11 by the occupant based on the pre-operation signal extracted by the potential analysis unit 21. This makes it possible to acquire the potential signal of the occupant without installing an electrode on a part of the occupant's body. Furthermore, it is possible to predict the operation of the steering wheel 11 by the occupant using the acquired potential signal.

Further, the potential analysis unit 21 refers to the steering operation database 31 that stores information indicating the relationship between the actual operation of the steering wheel 11 by the occupant and the potential signal, and calculates the potential signal detected by the palm potential detection unit 10. A pre-operation signal generated before the operation of the steering wheel 11 by the occupant is extracted from the pre-operation signal. Thereby, the potential analysis unit 21 can accurately extract the pre-operation signal that occurs before the operation of the steering wheel 11.

The pre-operation signal is a myoelectric potential signal generated immediately before the occupant makes a muscle movement to operate the steering wheel 11. In other words, the pre-operation signal is a signal indicating that the driver intends to operate the steering wheel 11.

Further, the potential analysis unit 21 filters the potential signal detected by the palm potential detection unit 10 excluding signals below 5 Hz using a bandpass filter in the high frequency band of 20 to 90 Hz to extract a pre-operation signal. Thereby, the potential analysis section 21 can accurately extract the pre-operation signal.

Further, the potential analysis unit 21 filters the potential signal detected by the palm potential detection unit 10 with a bandpass filter in a low frequency band of 1 to 6 Hz to extract a pre-operation signal. Thereby, the potential analysis section 21 can accurately extract the pre-operation signal.

(Second embodiment)
Next, a second embodiment of the present invention will be described with reference to FIG. The second embodiment differs from the first embodiment in that a sensitivity time database 33 is stored in the storage device 30. For configurations that overlap with those of the first embodiment, reference numerals will be cited and descriptions thereof will be omitted. The differences will be mainly explained below.

The sensitivity time database 33 stores the sensitivity time Δt when the driver operates the steering wheel 11 for each driver. This sensitivity time Δt is used to classify drivers based on their driving skill level. In the first embodiment, drivers are classified into three categories: advanced drivers with a high driving skill level, beginners with a low driving skill level, and intermediate drivers with a driving skill level in between. With this classification, for example, when changing lanes by operating the steering wheel 11 to avoid the motorcycle 80 as shown in FIG. It is possible to distinguish between drivers who change lanes after the lane has become shorter. In the second embodiment, as an example, the sensitivity time Δt for a beginner is 800 msec, the sensitivity time Δt for an intermediate user is 500 msec, and the sensitivity time Δt for an advanced user is 200 msec.

Note that the driving skill level may be set based on the driver's past operation history. Alternatively, the driving skill level may be set such that a driver with less than one year of driving experience is a beginner, a driver with one year or more but less than 10 years of driving experience is an intermediate level, and a driver with 10 years or more of driving experience is an advanced level. Note that the driving skill level is not limited to three, but may be classified into four or more.

In the first embodiment, the automatic function control unit 24 arbitrarily controls the operation timing of the steering wheel 11 for an automatic lane change between when a myoelectric potential signal caused by a pioneering operation is extracted and when a lane change is actually performed. He explained that he could decide to do so. On the other hand, in the second embodiment, the automatic function control unit 24 determines the operation timing of the steering wheel 11 for automatic lane change in consideration of the sensitivity time Δt. Details will be explained with reference to FIGS. 10A and 10B.

Time T0 shown in FIGS. 10A and 10B is the time when a myoelectric potential signal resulting from a precursor operation in preparation for a lane change is extracted. Time T1 shown in FIGS. 10A and 10B is the time when the lane change is actually performed by the driver. Alternatively, the time T1 may be the time when the steering wheel 11 is predicted to be actually operated by the driver. Time T2 shown in FIGS. 10A and 10B is an arbitrary time indicating the operation timing of the steering wheel 11 for automatic lane change determined by the automatic function control unit 24.

The automatic function control unit 24 refers to the sensitivity time database 33 and obtains the driving skill level of the driver. Note that, as a method for identifying the driver, a well-known technique such as face recognition technique as described above may be used. Then, the automatic function control unit 24 obtains the sensitivity time Δt set for each driving skill level.

The automatic function control unit 24 calculates the difference "T1-T2" between time T1 and time T2 (hereinafter this difference will be referred to as time difference α), and compares this time difference α with the driver's sensitivity time Δt.

As shown in FIG. 10A, when the time difference α is larger than the sensitivity time Δt, the automatic function control unit 24 sets the operation timing Tr of the steering wheel 11 to “T1−Δt”. That is, as shown in FIG. 10A, a timing T3a that is back by a sensitivity time Δt from time T1 is set as the operation timing Tr for automatically operating the steering wheel 11.

On the other hand, as shown in FIG. 10B, when the time difference α is less than or equal to the sensitivity time Δt, the automatic function control unit 24 sets the operation timing Tr of the steering wheel 11 to time T2. That is, as shown in FIG. 10B, the timing T3b, which is the sensitivity time Δt back from the time T1, is earlier than the time T2, so the automatic function control unit 24 sets the time T2 as the operation timing Tr.

By determining the operation timing Tr of the steering wheel 11 for automatic lane change in consideration of the sensitivity time Δt in this manner, the automatic operation by the automatic function device 50 is performed in accordance with the timing when the driver operates the steering wheel 11. will be executed. As a result, the steering wheel 11 operates automatically as if the driver were operating the steering wheel 11 himself. That is, it becomes possible to automatically operate the steering wheel 11 without making the driver aware that an automatic function is activated.

For example, if the driver is an intermediate driver, the sensitivity time Δt is set to 500 msec, and the automatic function is activated 500 msec earlier than time T1. Furthermore, if the driver is an advanced driver, the sensitivity time Δt is set to 200 msec, and the automatic function is activated 200 msec earlier than time T1. If the sensitivity time Δt is set to 500 msec for an advanced driver, the steering wheel 11 will operate slightly earlier than the timing at which the advanced driver would normally operate the steering wheel 11, giving him the feeling that he is driving himself. may be inhibited, leading to a feeling of discomfort.

In the second embodiment, the driving skill level is classified into advanced, intermediate, and beginner, and the operation timing Tr of the steering wheel 11 is determined using the sensitivity time Δt corresponding to each driving skill level. This allows the driver to control the operation of the steering wheel 11 as if he were operating it himself. This makes it possible to provide support that is tailored to the driver's personal driving skill level and that does not make the driver feel uncomfortable.

Next, an example of the operation of the operation prediction device according to the second embodiment will be described with reference to the flowchart of FIG. 11. However, the processing in steps S201 to 209 and 219 to 221 is the same as the processing shown in steps S101 to 109 and 111 to 121 shown in FIG. 8, so a description thereof will be omitted.

In step S211, the automatic function control unit 24 refers to the sensitivity time database 33 and obtains the sensitivity time Δt set for each driving skill level.

If the difference "T1-T2" between time T1 and time T2 is larger than the sensitivity time Δt (Yes in step S213), the process proceeds to step S217, and the automatic function control unit 24 sets the operation timing Tr of the steering wheel 11 to "T1". −Δt” (see FIG. 10A). On the other hand, if the difference "T1-T2" between time T1 and time T2 is less than or equal to the sensitivity time Δt (No in step S213), the process proceeds to step S215, and the automatic function control unit 24 adjusts the operation timing Tr of the steering wheel 11. Set at time T2 (see FIG. 10B).

Each functionality described in the embodiments described above may be implemented by one or more processing circuits. Processing circuitry includes programmed processing devices, such as processing devices that include electrical circuitry. Processing circuitry also includes devices such as application specific integrated circuits (ASICs) or circuit components arranged to perform the described functions.

As described above, embodiments of the present invention have been described, but the statements and drawings that form part of this disclosure should not be understood as limiting the present invention. Various alternative embodiments, implementations, and operational techniques will be apparent to those skilled in the art from this disclosure.

For example, in the first embodiment, a low-pass filter or a high-pass filter is used, but both filters may be used. That is, the potential analysis unit 21 filters the potential signal detected by the palm potential detection unit 10 by excluding signals below 5 Hz using a band pass filter (high pass filter) in a high frequency band of 20 to 90 Hz, and The signal related to the operation of the steering wheel 11 by the occupant may be extracted by filtering the potential signal detected by the steering wheel 10 with a band pass filter (low pass filter) in a low frequency band of 1 to 6 Hz. Even when both filters are used, the same effect as when each filter is used can be obtained.

Further, the pre-operation signal (myoelectric potential signal generated immediately before the occupant's muscle movement) may be extracted by the following method. The potential signal detected by the palm potential detection unit 10 is filtered with a bandpass filter in a high frequency band of 20 to 90 Hz, and the filtered signal is subjected to absolute value processing to convert all signal intensities into positive values. By filtering the signal with a 5 Hz low-pass filter, a myoelectric potential signal that occurs immediately before the occupant's muscle movement may be extracted. Even when this extraction method is used, effects similar to those described above can be obtained.

Although the myoelectric potential signal generated immediately before the occupant's muscle movement has been described as an example of the pre-operation signal, the pre-operation signal is not limited to this. The pre-operation signal may be a signal caused by a change in impedance between the palm potential detection unit 10 and the palm of the occupant. The impedance change between the palm potential detection unit 10 and the occupant's palm is due to a change in the contact surface pressure between the palm potential detection unit 10 and the occupant's palm, or the change in impedance between the palm potential detection unit 10 and the occupant's palm. This is due to changes in the contact area between the two.

An example of a method for extracting a signal caused by a change in impedance between the palm potential detection unit 10 and the palm of an occupant will be described. The potential analysis unit 21 filters the potential signal of the occupant's palm with a bandpass filter in a low frequency band of 1 to 6 Hz, thereby extracting a signal caused by an impedance change between the palm potential detection unit 10 and the occupant's palm. You may.

10 Palm potential detection section 11 Steering wheel 20 Controller 21 Potential analysis section 22 Steering operation prediction section 23 Operation timing prediction section 24 Automatic function control section 30 Storage device 31 Steering operation database 32 Steering operation timing database 33 Sensitivity time database 40 Surrounding information acquisition section 50 Automatic function device

Claims (6)

A sensor provided on the steering wheel detects a potential signal in the occupant's palm caused by muscle activity when the occupant operates the steering wheel;
extracting a pre-operation signal generated before the operation of the steering wheel by the occupant from the potential signal detected by the sensor;
A steering wheel operation prediction method for predicting an operation of the steering wheel by the occupant based on the extracted pre-operation signal, the method comprising:
The pre-operation signal is a myoelectric potential signal generated immediately before the occupant's muscle movement to operate the steering wheel.
filtering the potential signal detected by the sensor with a bandpass filter in a high frequency band of 20 to 90 Hz;
By performing absolute value processing on the signal after filtering, all signal strengths are converted into positive values,
By filtering the signal after absolute value processing with a 5 Hz low-pass filter, a myoelectric potential signal generated immediately before the muscle movement of the occupant is extracted.
How to predict steering wheel operation. The myoelectric potential signal is extracted from the potential signal detected by the sensor by referring to a database storing information indicating a relationship between the operation of the steering wheel by the occupant and the potential signal. The steering wheel operation prediction method according to claim 1. A sensor provided on the steering wheel detects a potential signal in the occupant's palm caused by muscle activity when the occupant operates the steering wheel;
extracting a pre-operation signal generated before the operation of the steering wheel by the occupant from the potential signal detected by the sensor;
A steering wheel operation prediction method for predicting an operation of the steering wheel by the occupant based on the extracted pre-operation signal, the method comprising:
The steering wheel operation prediction method, wherein the pre-operation signal is a signal caused by an impedance change between the sensor and the palm of the occupant. The impedance change between the sensor and the occupant's palm is caused by a change in the contact surface pressure between the sensor and the occupant's palm, or by a change in the contact area between the sensor and the occupant's palm. 4. The steering wheel operation prediction method according to claim 3 , wherein the steering wheel operation prediction method is based on a change. The electric potential signal detected by the sensor is filtered with a bandpass filter in a low frequency band of 1 to 6 Hz to extract a signal caused by an impedance change between the sensor and the palm of the occupant. The steering wheel operation prediction method according to claim 3 or 4 . a sensor that is provided on a steering wheel and detects an electric potential signal of the occupant's palm caused by muscle activity when the occupant operates the steering wheel;
Extracting a pre-operation signal generated before the operation of the steering wheel by the occupant from the potential signal detected by the sensor, and predicting the operation of the steering wheel by the occupant based on the extracted pre-operation signal. a controller to
A steering wheel operation prediction device,
The pre-operation signal is a myoelectric potential signal generated immediately before the occupant's muscle movement to operate the steering wheel.
The controller includes:
filtering the potential signal detected by the sensor with a bandpass filter in a high frequency band of 20 to 90 Hz;
By performing absolute value processing on the signal after filtering, all signal strengths are converted into positive values,
By filtering the signal after absolute value processing with a 5 Hz low-pass filter, a myoelectric potential signal generated immediately before the muscle movement of the occupant is extracted.
Steering wheel operation prediction device.