EP4775095A1

EP4775095A1 - Seat device

Info

Publication number: EP4775095A1
Application number: EP25856778.3A
Authority: EP
Inventors: Yoritaka Doi; Naoya Matsumoto
Original assignee: TS Tech Co Ltd
Current assignee: TS Tech Co Ltd
Priority date: 2024-08-22
Filing date: 2025-08-08
Publication date: 2026-07-15
Also published as: CN121925203A; WO2026042641A1

Abstract

[TASK] To provide a seat device capable of accurately estimating a position of a sole.[SOLUTION] A seat device 1 includes: a seat main body 2 provided behind an operation pedal 3; a sensor 10 installed on a seat surface 7 of the seat main body 2 and configured to detect a body pressure distribution of a seated person; and a controller configured to estimate a position of a sole of a right foot of the seated person based on the body pressure distribution. The controller is configured to estimate the position of the sole of the right foot based on the body pressure distribution in each of a plurality of monitoring areas set in different positions on the seat surface 7.

Description

TECHNICAL FIELD

The present invention relates to a seat device.

BACKGROUND ART

Patent Document 1 discloses a vehicle control system that detects a preparatory action or movement unconsciously performed by a driver and controls a vehicle based on the action or movement. The vehicle control system detects a posture change that appears immediately before the driver operates a brake pedal or an accelerator pedal, based on a pressure distribution and a center-of-gravity movement vector acquired from pressure sensors provided in a seat cushion, and performs vehicle control, such as acceleration or deceleration, when the posture change corresponds to the preparatory action or movement. As a result, a delay in the vehicle response to driving operations is reduced, thereby improving safety and comfort.

PRIOR ART DOCUMENT(S)

PATENT DOCUMENT(S)

Patent Document 1: Japanese Patent No. 4740399

SUMMARY OF THE INVENTION

TASK TO BE ACCOMPLISHED BY THE INVENTION

In the invention described in Patent Document 1, a posture is estimated based on the pressure distribution over an entire area on the seat cushion, i.e., the pressure distribution in an area including ischial positions of the seated person. Accordingly, pressure applied by the ischial bones of the buttocks becomes dominant, and changes in leg pressure are relatively small and tend to be obscured. Accordingly, it is difficult to accurately detect and estimate the position and movement of a foot, and there is a limitation in the prediction accuracy of driving operations by the legs, such as brake or accelerator operations.
In view of the above background, an object of the present invention is to provide a seat device capable of accurately estimating a position of a sole.

MEANS TO ACCOMPLISH THE TASK

To achieve such an object, one aspect of the present invention provides a seat device, comprising: a seat main body provided behind an operation pedal; a sensor installed on a seat surface of the seat main body and configured to detect a body pressure distribution of a seated person; and a controller configured to estimate a position of a sole of a right foot of the seated person based on the body pressure distribution, and the controller is configured to estimate the position of the sole of the right foot based on the body pressure distribution in each of a plurality of monitoring areas set in different positions on the seat surface.

EFFECT OF THE INVENTION

According to this aspect, it is possible to provide a seat device capable of accurately estimating the position of the sole.

BRIEF DESCRIPTION OF THE DRAWING(S)

[Figure 1] a perspective view of a seat device according to an embodiment
[Figure 2] an explanatory diagram showing a configuration of the seat device
[Figure 3] an explanatory diagram showing a position of a sole of a right foot of a seated person
[Figure 4] an explanatory diagram showing a pressure distribution acquired by a first sensor provided in a seat cushion
[Figure 5] an explanatory diagram of feature amounts acquired based on the pressure distribution
[Figure 6] a flowchart showing a procedure for estimating the position of the sole of the right foot of the seated person
[Figure 7] an explanatory diagram showing a body pressure distribution acquired by the first sensor and a second sensor
[Figure 8] an explanatory diagram showing an example of a monitoring area according to a modified embodiment
[Figure 9] an explanatory diagram showing an example of the monitoring area according to the modified embodiment
[Figure 10] an explanatory diagram showing an example of the monitoring area according to the modified embodiment

MODE(S) FOR CARRYING OUT THE INVENTION

In the following, an embodiment in which a seat device of the present invention is applied to a driver's seat of an automobile will be described with reference to the drawings.
As shown in Figure 1, a seat device 1 includes a seat main body 2 provided on a floor of a vehicle. The seat main body 2 constitutes the driver's seat of the vehicle. The seat main body 2 is provided behind operation pedals 3 of the vehicle. The operation pedals 3 include an accelerator pedal 3A and a brake pedal 3B.
The seat main body 2 includes a seat cushion 5 that supports buttocks of a seated person from below, and a seat back 6 that extends upward from a rear portion of the seat cushion 5 and supports a back of the seated person from behind. The seat cushion 5 and the seat back 6 each include a frame, a pad supported by the frame, and a skin material covering the pad. The seat cushion 5 is provided on the floor of the vehicle. A sliding device or a lifting device may be provided between the floor and the seat cushion 5.
The seat main body 2 includes a seat surface 7 that supports the seated person. The seat surface 7 includes an upper surface of the seat cushion 5 and a front surface of the seat back 6. The seat surface 7 is provided with a sensor 10 capable of detecting a body pressure distribution of the seated person.
The sensor 10 is preferably a sheet-shaped pressure distribution sensor capable of measuring the body pressure distribution of the seated person. The pressure distribution sensor may be configured to have a plurality of pressure detection elements arranged in a grid pattern or matrix pattern to acquire a pressure distribution on a two-dimensional plane in an area-wise manner. The pressure sensing elements are incorporated into a layered sensor sheet including a pressure-sensitive material or structure.
The pressure detection element may be, for example, a resistive sensor in which the resistance of a conductive polymer or conductive rubber changes depending on the pressure. The pressure detection element may also be a capacitance sensor that detects a change in capacitance between electrodes, a piezoelectric sensor that generates an electric charge in response to pressure, or a force sensor array.
A base material that constitutes the sensor sheet may be a flexible PET film, a silicone-based material, or the like. The sensor sheet flexibly conforms to a curved shape of the seat surface 7 and bending deformation during use.
The sensor 10 includes a first sensor 10A provided on the upper surface of the seat cushion 5 that constitutes the seat surface 7, and a second sensor 10B provided on the front surface of the seat back 6. The first sensor 10A is preferably provided between the upper surface of the pad constituting the seat cushion 5 and the skin material. The second sensor 10B is preferably provided between the upper surface of the pad constituting the seat back 6 and the skin material.
As shown in Figure 4, the first sensor 10A and the second sensor 10B each include a plurality of detecting portions (cells) 11 arranged in a grid pattern along the surface. The plurality of detecting portions 11 is arranged in an X direction and a Y direction. In the present embodiment, the first sensor 10A and the second sensor 10B each have 16×16 detecting portions 11. The position of each detecting portion 11 is represented by a position in the X direction and a position in the Y direction. The first sensor 10A is arranged such that the X direction corresponds to the left-and-right direction and the Y direction corresponds to the front-and-rear direction. The X-direction value increases from right to left (X01 to X16), and the Y-direction value increases from front to rear (Y01 to Y16). The second sensor 10B is arranged such that the X direction corresponds to the left-and-right direction and the Y direction corresponds to the up-and-down direction. In the second sensor 10B, the X-direction value increases from right to left (X01 to X16), and the Y-direction value increases from bottom to top (Y01 to Y16).
As shown in Figure 2, the first sensor 10A and the second sensor 10B are connected to a seat controller 20. The seat controller 20 is an electronic control device including a processor 21 and a memory 22 communicatively connected to the processor 21. The processor 21 may include, as a core, at least one of a central processing unit (CPU), a graphics processing unit (GPU), and a reduced instruction set computer (RISC). The memory 22 may include at least one of a volatile memory and a non-volatile memory. The volatile memory may be, for example, a dynamic random access memory (DRAM) or a static random access memory (SRAM). The non-volatile memory may be a solid state drive (SSD), a flash memory, a magnetic disk storage, or an optical disk storage. At least a portion of the seat controller 20 may be realized by hardware, such as an LSI, an ASIC, or an FPGA, or may be realized by a combination of software and hardware. The processor 21 executes programs stored in the memory 22 to realize various applications. The seat controller 20 may be composed of a single piece of hardware, or may be composed of a plurality of pieces of hardware that can communicate with each other.
The processor 21 of the seat controller 20 executes the programs stored in the memory 22 to function as a sole position estimating unit 25, a hazard prediction level estimating unit 26, a cause estimating unit 27, and a notification unit 28.
The sole position estimating unit 25 estimates the position of the sole of the right foot of the seated person based on the body pressure distribution of the seated person. The position of the sole of the right foot of the seated person includes at least a position facing the accelerator pedal 3A (see Figure 3A), a position facing the brake pedal 3B (see Figure 3B), and a position on the floor. The sole position estimating unit 25 acquires the body pressure distribution of the seated person based on signals from the first sensor 10A and the second sensor 10B. Here, the body pressure distribution includes the pressure at the plurality of detecting portions 11 of the first sensor 10A and the second sensor 10B. The sole position estimating unit 25 estimates the position of the sole of the right foot of the seated person based on the body pressure distribution in each of a plurality of monitoring areas set in different positions of the seat surface 7.
The plurality of monitoring areas includes a first monitoring area. The first monitoring area includes at least a central portion in the front-and-rear direction on the right side of the seat surface 7 of the seat cushion 5. The first monitoring area is an area facing the right thigh of the seated person, and is intended to detect the pressure distribution received from the right thigh of the seated person. The first monitoring area may be a central area obtained by dividing the right half of the seat surface 7 of the seat cushion 5 into thirds in the front-and-rear direction.
As shown in Figure 4, it is preferable that the plurality of monitoring areas be set in the right half of the seat surface 7 of the seat cushion 5. Some of the plurality of monitoring areas may include left and right ischial support portions 31 arranged on the seat surface 7 of the seat cushion 5. The left and right ischial support portions 31 are portions corresponding to the left and right ischia of the seated person, and may be preset based on a typical ischial position of the seated person. Alternatively, the ischial support portion 31 may be set at the positions of maximum pressure in the right and left halves, obtained by dividing the pressure distribution acquired by the first sensor 10A into the right and left halves. The ischial support portion 31 may correspond to an area associated with a single detecting portion 11 having the maximum pressure, or to an area including the single detecting portion 11 having the maximum pressure and the surrounding detecting portions 11.
The plurality of monitoring areas is composed of four monitoring areas A1 to A4. The monitoring area A1 is set in the central portion in the front-and-rear direction of the right half of the seat surface 7 of the seat cushion 5. The rear portion of the monitoring area A1 may overlap with the right ischial support portion 31. The monitoring areas A2 and A3 may be areas obtained by dividing the monitoring area A1 into right and left halves. The monitoring area A2 may correspond to the left half of the monitoring area A1, and the monitoring area A3 may correspond to the right half of the monitoring area A1. The rear portion of the monitoring area A2 may overlap with the right ischial support portion 31. The rear portion of the monitoring area A3 may be set so as not to overlap with the right ischial support portion 31. The monitoring area A4 is set at the front right portion of the seat surface 7 of the seat cushion 5. The monitoring area A4 may be set so as not to overlap with the right ischial support portion 31. In this example, the monitoring area A1 and the monitoring area A2 correspond to the first monitoring area.
In the present embodiment, the first sensor 10A includes 16 detecting portions 11 in the X direction (the lateral direction), designated as X01 to X16, and 16 detecting portions 11 in the Y direction (the front-and-rear direction), designated as Y01 to Y16, thereby forming a 16×16 array of detecting portions 11. The monitoring area A1 is an 8×8 square area defined by (X1, Y7) and (X8, Y14). The monitoring area A2 is a 4×8 square area defined by (X5, Y7) and (X8, Y14). The monitoring area A3 is a 4×8 square area defined by (X1, Y7) and (X4, Y14). The monitoring area A4 is an 8×8 square area defined by (X1, Y1) and (X8, Y8).
The sole position estimating unit 25 calculates the feature amounts related to the body pressure distribution in each of the plurality of monitoring areas A1 to A4 based on the body pressure distribution in each of the plurality of monitoring areas A1 to A4, and estimates the position of the sole of the right foot based on the feature amount. The feature amounts are calculated for each monitoring area. The feature amounts may include, for example, at least one of the first to seventh feature amounts.
The first feature amount is a position in the lateral direction (X direction) of a centroid position G of the body pressure distribution in each monitoring area A1 to A4 (see Figure 5). Here, the centroid position G of the body pressure distribution in each of the monitoring areas A1 to A4 is a centroid position G calculated based on the body pressure distribution limited to the corresponding monitoring area. The centroid position G may be calculated, for example, by multiplying the pressure value at each detecting portion 11 of the body pressure distribution by its position in the X direction, and dividing the sum by the total pressure value.
The second feature amount is the position in the front-and-rear direction (Y direction) of the centroid position G of the body pressure distribution in each monitoring area A1 to A4.
The third feature amount is the ratio of the area in each monitoring area A1 to A4 where the pressure is equal to or higher than the determination value. For example, in an 8×8 monitoring area, when the number of detecting portions (cells) 11 where the pressure is equal to or higher than the determination value is 32, the ratio of the area where the pressure is equal to or higher than the determination value is 0.5 (=32/64). The third feature amount may be the number of detecting portions (cells) 11 where the pressure is equal to or higher than the determination value, instead of the ratio of the area where the pressure is equal to or higher than the determination value.
The fourth and fifth feature amounts may be the angles on the seat surface 7 of the centroid position G of the body pressure distribution in each monitoring area A1 to A4, centered on the reference point set on the seat surface 7. In the fourth feature amount, the reference point is a center point C of the seat surface 7 of the seat cushion 5 (see Figure 5). In the fifth feature amount, the reference point is the ischial support portion 31 on the right side (see Figure 5). The center point C of the seat surface 7 of the seat cushion 5 and the right ischial support portion 31 may be at preset positions. Further, the right ischial support portion 31 may be set at the position of the detecting portion 11 having a maximum pressure in the right half of the pressure distribution. The angle on the seat surface 7 of the centroid position G of the body pressure distribution in each monitoring area A1 to A4, centered on the reference point set on the seat surface 7, may be the angle with respect to a reference line that passes through the reference point and extends in the lateral direction. The angle may be calculated based on the centroid position G of the body pressure distribution in each of the monitoring areas A1 to A4 and the position of the reference point.
The sixth and seventh feature amounts include the distance between the centroid position G of the body pressure distribution in each monitoring area A1 to A4 and the reference point set on the seat surface 7. In the sixth feature amount, the reference point is the center point C of the seat surface 7 of the seat cushion 5 (the center of the seat surface). In the seventh feature amount, the reference point is the right ischial support portion 31. The distance between the centroid position G of the body pressure distribution in each monitoring area A1 to A4 and the reference point set on the seat surface 7 may be calculated based on the centroid position G of the body pressure distribution in each monitoring area A1 to A4 and the position of the reference point.
In the present embodiment, since the first to seventh feature amounts are acquired for each of the four monitoring areas A1 to A4, a total of 28 feature amounts are acquired.
The sole position estimating unit 25 estimates the position of the sole of the right foot based on the feature amounts using the machine learning model that receives the feature amounts as input and outputs the position of the sole of the right foot. The machine learning model may be configured by a known neural network model, a deep learning model, or the like. The machine learning model may be a trained model trained using training data. Each record of the training data includes a plurality of feature amounts and the position of the sole of the right foot stored in association with the feature amounts. The position of the sole preferably includes at least three positions: "on the brake pedal," "on the accelerator pedal," and "on the floor." In the present embodiment, 28 feature amounts are input to the machine learning model, and the machine learning model outputs, as the position of the sole of the right foot, one of the positions: "on the brake pedal," "on the accelerator pedal," and "on the floor."
The position of the sole may include "in transition." The position of the sole being "in transition" refers to a state in which the position of the sole does not correspond to any of "on the brake pedal," "on the accelerator pedal," or "on the floor," and the position of the sole is changing. The machine learning model outputs, as the position of the sole of the right foot, one of "on the brake pedal," "on the accelerator pedal," "on the floor," and "in transition" in response to the input of the feature amount.
In this case, the position of the sole in the training data for creating the machine learning model may include "on the brake pedal," "on the accelerator pedal," "on the floor," and "in transition."
As shown in Figure 2, the seat controller 20 is connected to a vehicle control system 40. The vehicle control system 40 is an electronic control device that controls the traveling of the vehicle and includes a processor 41 and a memory 42 communicatively connected to the processor 41. The configurations of the processor 41 and the memory 42 may be the same as those of the processor 21 and the memory 22.
The processor 41 of the vehicle control system 40 functions as a travel control unit 44, an obstacle recognition unit 45, and a hazard predicting unit 46 by executing the programs stored in the memory 42.
The vehicle control system 40 is connected to the accelerator pedal 3A and the brake pedal 3B. The accelerator pedal 3A and the brake pedal 3B each include a sensor that detects the position thereof, and outputs a signal corresponding to the position to the vehicle control system 40. The travel control unit 44 of the vehicle control system 40 controls a propulsion device, such as an internal combustion engine or an electric motor, according to the position of the accelerator pedal 3A, and controls the brake according to the position of the brake pedal 3B.
The vehicle control system 40 is connected to an external environment recognizer 51 that detects the obstacles present around the vehicle. The external environment recognizer 51 is a sensor that captures electromagnetic waves and light from the surroundings of the vehicle to detect objects outside the vehicle. The external environment recognizer 51 includes at least one of a radar, a lidar, and a camera.
The obstacle recognition unit 45 of the vehicle control system 40 recognizes the obstacle present around the vehicle based on the detection results of the external environment recognizer 51. The obstacle recognition unit 45 recognizes the surrounding environment (external environment) including the obstacle around the vehicle, a road shape, presence or absence of a sidewalk, road markings, and the like, based on the detection results of the external environment recognizer 51. The obstacle includes, for example, a guardrail, a utility pole, a surrounding vehicle, and people, such as pedestrians. The obstacle recognition unit 45 can acquire the state of the obstacle, such as the position, speed, and acceleration of the surrounding vehicle, pedestrian, and the like, from the detection results of the external environment recognizer 51.
The hazard predicting unit 46 of the vehicle control system 40 determines whether the vehicle is in a hazardous state. Here, a hazardous state refers to a state in which the vehicle may collide with an obstacle, such as another vehicle, a structure, or a pedestrian. The hazard predicting unit 46 determines, for example, the possibility of collision between the obstacle and the vehicle, and determines a hazardous state based on the possibility of collision. For example, when the time to collision (TTC) between the obstacle and the vehicle is equal to or lower than a collision threshold, the hazard predicting unit 46 determines that there is a possibility of collision between the obstacle and the vehicle. The hazard predicting unit 46 may determine the possibility of collision with the obstacle whose distance from the vehicle is within the prescribed value among the obstacles detected by the obstacle recognition unit 45. The time to collision may be calculated by dividing the distance between the obstacle and the vehicle by the relative speed between the obstacle and the vehicle. Further, the hazard predicting unit 46 may determine a hazardous state based on blind spot information obtained by communicating with another vehicle. The blind spot information may include information about pedestrians and the like that are present around the other vehicle and are present at positions not visible from the vehicle. The hazard predicting unit 46 may determine a hazardous state based on the possibility of collision with the obstacle included in the blind spot information.
The vehicle control system 40 outputs information related to the respective positions of the accelerator pedal 3A and the brake pedal 3B to the seat controller 20. The information related to the respective positions of the accelerator pedal 3A and the brake pedal 3B may be information indicating whether each of the accelerator pedal 3A and the brake pedal 3B is being depressed, or may be information indicating the respective positions of the accelerator pedal 3A and the brake pedal 3B. The vehicle control system 40 also outputs information related to the hazardous state determined by the hazard predicting unit 46 to the seat controller 20. The information related to the hazardous state may be the presence or absence of the hazardous state, or may be a value obtained by quantifying the hazard state.
The sole position estimating unit 25 may estimate the position of the sole of the right foot of the seated person based on information related to the respective positions of the accelerator pedal 3A and the brake pedal 3B received from the vehicle control system 40. For example, when the brake pedal 3B is being depressed, the sole position estimating unit 25 may determine that the sole of the right foot of the seated person is operating the brake pedal 3B.
Figure 6 shows an example of the procedure for estimating the position of the sole performed by the sole position estimating unit 25 of the seat controller 20. First, the seat controller 20 acquires the body pressure distribution of the seat cushion 5 based on a signal from the first sensor 10A (ST1). Next, the seat controller 20 corrects the acquired body pressure distribution (ST2). The correction may be performed to reduce noise in the body pressure distribution. For example, the pressure value of the detecting portion (cell) 11 whose pressure is equal to or lower than the prescribed lower limit value may be set to 0. Further, the pressure value of the detecting portion (cell) 11 whose pressure is equal to or higher than the prescribed upper limit value may be set to the upper limit value. The correction process in step ST2 may be omitted.
Next, the seat controller 20 acquires the body pressure distribution in each of the monitoring areas A1 to A4 (ST3). Next, the seat controller 20 calculates the first to seventh feature amounts for each of the monitoring areas A1 to A4 (ST4).
Next, the seat controller 20 inputs the first to seventh feature amounts acquired for each of the monitoring areas A1 to A4 into the machine learning model, and obtains an output of the position of the sole of the right foot of the seated person (ST5). The position of the sole is one of "on the brake pedal," "on the accelerator pedal," "on the floor," and "in transition."
Next, the seat controller 20 determines whether the position of the sole is "in transition" (ST6). When the position of the sole is "in transition" (ST6: Yes), the seat controller 20 determines that the position of the sole is "in transition" (ST7).
When the position of the sole is not "in transition" (ST6: No), the seat controller 20 determines whether the accelerator pedal 3A or the brake pedal 3B is being depressed based on information related to the respective positions of the accelerator pedal 3A and the brake pedal 3B received from the vehicle control system 40 (ST8). When the accelerator pedal 3A or the brake pedal 3B is being depressed (ST8: Yes), the seat controller 20 determines the position of the sole as "pedal-depressing" (ST9).
When neither the accelerator pedal 3A nor the brake pedal 3B is being depressed (ST8: No), the seat controller 20 determines the position of the sole as one of "on the brake pedal," "on the accelerator pedal," and "on the floor" based on the estimated position of the sole derived from the output from the machine learning model in step ST5 (ST10).
The hazard prediction level estimating unit 26 determines the hazard prediction level of the seated person, who is the driver, based on the sole position determined by the sole position estimating unit 25 and information related to the hazardous state received from the hazard predicting unit 46. The hazard prediction level is expressed as high or low. When the hazard prediction level is "high," the seated person is highly aware of monitoring the surroundings of the vehicle and can respond promptly to unexpected situations.
The hazard prediction level estimating unit 26 determines the hazard prediction level of the seated person based on the position of the sole of the right foot of the seated person when it is determined, based on information related to the hazardous state received from the hazard predicting unit 46, that the vehicle is in a hazardous state. The hazardous state may be, for example, a state in which the possibility of collision between the vehicle and the obstacle is equal to or higher than the prescribed value. More specifically, the hazardous state may be a state in which the time to collision (TTC) is equal to or lower than the prescribed value.
The hazard prediction level estimating unit 26 estimates the hazard prediction level to be "high" when the vehicle is in a hazardous state, and the position of the sole of the right foot of the seated person is "on the brake pedal." That is, since the position of the sole of the right foot of the seated person is "on the brake pedal," and the seated person is in a state in which the brake pedal 3B can be immediately depressed, the hazard prediction level is estimated to be "high."
On the other hand, the hazard prediction level estimating unit 26 estimates the hazard prediction level to be "low" when the vehicle is in a hazardous state, and the position of the sole of the right foot of the seated person is not "on the brake pedal." When the sole of the right foot of the seated person is not "on the brake pedal," time is required for the seated person to depress the brake pedal 3B, and therefore the hazard prediction level is estimated to be "low. "
The notification unit 28 of the seat controller 20 controls a notifier 53. The notifier 53 includes, for example, a display and a speaker, and provides notification to the seated person by image or sound. When the hazard prediction level estimating unit 26 estimates the hazard prediction level to be "low," the notification unit 28 notifies the seated person by image or sound that the hazard prediction level has decreased. The notification unit 28 may also prompt the seated person to take a rest.
When the hazard prediction level is equal to or lower than the prescribed determination value, the cause estimating unit 27 estimates, based on the body pressure distribution, the cause of the hazard prediction level being equal to or lower than the prescribed determination value. As shown in Figure 7, the body pressure distribution of the seat cushion 5 acquired by the first sensor 10A and the body pressure distribution of the seat back 6 acquired by the second sensor 10B vary depending on the seating position of the seated person (proper, forward, or rearward). As fatigue accumulates in the seated person, the seating position tends to shift forward. Therefore, the fatigue state of the seated person can be estimated based on the body pressure distribution. When the cause estimating unit 27 identifies fatigue as the cause of the hazard prediction level being equal to or lower than the determination value, the notification unit 28 may notify the seated person that the seated person is fatigued and may issue a message prompting the seated person to take a rest.
This concludes the description of the specific embodiment, but the present invention is not limited to the above embodiment and can be widely modified and implemented. In another embodiment, the number of monitoring areas may be changed as appropriate. For example, as shown in Figures 8 to 10, 6 monitoring areas B1 to B6 may be set. In this case, the first to seventh feature amounts may also be acquired for each of the monitoring areas B1 to B6. Accordingly, the number of feature amounts input to the machine learning model for estimating the position of the sole is 42 (6×7).
Some of the plurality of monitoring areas may be set so as not to include the ischial support portion 31. This prevents the pressure applied from the ischia from becoming dominant, and enables changes in pressure applied from the right thigh of the seated person to be accurately detected. For example, the monitoring areas B2 to B6 shown in Figures 9 and 10 are examples of the monitoring areas that do not include the ischial support portion 31. On the other hand, the monitoring areas A1 to A3 shown in Figure 4 and the monitoring area B1 shown in Figure 8 are examples of the monitoring areas that include the ischial support portion 31. The number of monitoring areas and the number of feature amounts can be changed as appropriate.
A plurality of ventilation holes penetrating in a thickness direction may be formed at appropriate positions of the first sensor 10A and the second sensor 10B. The plurality of ventilation holes may be arranged so as to avoid the detecting portion 11. For example, the ventilation holes may be arranged at the boundaries between adjacent detecting portions 11. Alternatively, some of the detecting portions 11 in the first sensor 10A and the second sensor 10B may be removed, and the ventilation holes may be provided in the removed portions. The plurality of ventilation holes may be arranged so as not to overlap with the monitoring area. The ventilation hole may be arranged opposite to a ventilation passage formed in the pad. The ventilation passage may be connected to a blower. For example, the detecting portion 11 to be removed may be a rectangular area defined by (X01, Y01) and (X04, Y12), or a rectangular area defined by (X01, Y01) and (X04, Y16). The ventilation holes may be provided in these areas. These areas are areas in the right half of the seat surface 7 of the seat cushion 5 that are closer to the accelerator pedal 3A than to the brake pedal 3B. Therefore, these areas have a small influence on the determination accuracy as to whether the sole of the foot is positioned on the brake pedal 3B. Accordingly, the ventilation holes can be provided without significantly reducing the determination accuracy in determining whether the sole of the foot is positioned on the brake pedal 3B, and further in determining the hazard prediction level.
The above embodiment may also be described as follows:
One embodiment provides a seat device 1, comprising: a seat main body 2 provided behind an operation pedal 3; a sensor 10 installed on a seat surface 7 of the seat main body 2 and configured to detect a body pressure distribution of a seated person; and a controller configured to estimate a position of a sole of a right foot of the seated person based on the body pressure distribution, and the controller is configured to estimate the position of the sole of the right foot based on the body pressure distribution in each of a plurality of monitoring areas set in different positions on the seat surface 7.
According to this aspect, since the controller estimates the position of the sole of the right foot based on the body pressure distribution for each of a plurality of monitoring areas on the seat surface 7, the position of the foot can be accurately estimated while suppressing the influence of pressure applied from the buttocks. Accordingly, it is possible to provide the seat device 1 capable of accurately estimating the position of the sole.
In the above embodiment, preferably, the seat main body 2 includes a seat cushion 5 configured to support buttocks of the seated person from below, and a seat back 6 that extends upward from a rear portion of the seat cushion 5 and is configured to support a back of the seated person from behind, the plurality of monitoring areas includes a first monitoring area, and the first monitoring area includes at least a central portion in a front-and-rear direction on a right side of the seat surface 7 of the seat cushion 5.
According to this aspect, since the first monitoring area is arranged at a position corresponding to the right thigh of the seated person, pressure changes related to the position of the sole of the right foot can be accurately captured, further improving the accuracy of estimating the sole position.
In the above embodiment, preferably, the controller is configured to: calculate a feature amount related to the body pressure distribution in each of the plurality of monitoring areas based on the body pressure distribution in each of the plurality of monitoring areas; and estimate the position of the sole of the right foot based on the feature amount.
In the above embodiment, preferably, the feature amount includes a centroid position G in each of the plurality of monitoring areas.
According to this aspect, by using the centroid position G in each monitoring area as the feature amount, it is possible to capture the change in the body pressure distribution concisely and accurately, and improve the accuracy of estimating the position of the sole of the foot.
In the above embodiment, preferably, the feature amount includes a distance between a centroid position G in each of the plurality of monitoring areas and a reference point set on the seat surface 7.
According to this aspect, by including the distance between the centroid position G and the reference point as the feature amount, the spatial relative relationship of the position of the sole can be clearly reflected, enabling more accurate estimation.
In the above embodiment, preferably, the feature amount includes an angle on the seat surface 7 of a centroid position G in each of the plurality of monitoring areas, the angle being centered on a reference point set on the seat surface 7.
According to this aspect, by including the angle of the centroid position G centered on the reference point as the feature amount, it becomes possible to estimate the sole position while reflecting the orientation and movement tendency of the foot, thereby improving the reliability of the estimation.
In the above embodiment, preferably, the feature amount includes, in each of the plurality of monitoring areas, a ratio of an area in which pressure is equal to or higher than a determination value.
According to this aspect, by including the ratio of the area in which the pressure is equal to or higher than the determination value as the feature amount, it becomes possible to quantitatively evaluate the intensity and spread of the pressure distribution, thereby improving the accuracy of estimating the sole position.
In the above embodiment, preferably, the controller is configured to estimate the position of the sole of the right foot based on the feature amount using a machine learning model that receives the feature amount as input and outputs the position of the sole of the right foot.
According to this aspect, by determining whether the sole is located at a position facing the pedal using the machine learning model, it becomes possible to perform a highly accurate determination that comprehensively takes a plurality of feature amounts into account.
In the above embodiment, preferably, the seat main body 2 is installed in a vehicle, the operation pedal 3 is a brake pedal 3B of the vehicle, and the controller is configured to determine a hazard prediction level of the seated person based on the position of the sole of the right foot when the controller determines a risk of collision between the vehicle and an obstacle.
According to this aspect, when there is a risk of collision, the hazard prediction level of the seated person can be determined based on the sole position. Accordingly, it is possible to provide the seat device 1 that accurately evaluates the driver's reaction readiness and contributes to safe driving support.
In the above embodiment, preferably, the controller is configured to, when the hazard prediction level is equal to or lower than a prescribed determination value, estimate a cause of the hazard prediction level being equal to or lower than the prescribed determination value based on the body pressure distribution.
According to this aspect, when the hazard prediction level is low, the cause thereof can be estimated from the body pressure distribution, thereby making it possible to identify the driver's lack of attention or inappropriate posture, and to provide more appropriate driving assistance and warning control.

LIST OF REFERENCE NUMERALS

1: seat device
2: seat main body
3: operation pedal
3A: accelerator pedal
3B: brake pedal
5: seat cushion
6: seat back
7: seat surface
10: sensor
10A: first sensor
10B: second sensor
11: detecting portion
20: seat controller
25: sole position estimating unit
26: hazard prediction level estimating unit
27: cause estimating unit
28: notification unit
31: ischial support portion
40: vehicle control system
44: travel control unit
45: obstacle recognition unit
46: hazard predicting unit
51: external environment recognizer
53: notifier

Claims

A seat device, comprising:
a seat main body provided behind an operation pedal;

a sensor installed on a seat surface of the seat main body and configured to detect a body pressure distribution of a seated person; and

a controller configured to estimate a position of a sole of a right foot of the seated person based on the body pressure distribution, wherein

the controller is configured to estimate the position of the sole of the right foot based on the body pressure distribution in each of a plurality of monitoring areas set in different positions on the seat surface.
The seat device according to claim 1, wherein the seat main body includes a seat cushion configured to support buttocks of the seated person from below, and a seat back that extends upward from a rear portion of the seat cushion and is configured to support a back of the seated person from behind,
the plurality of monitoring areas includes a first monitoring area, and

the first monitoring area includes at least a central portion in a front-and-rear direction on a right side of the seat surface of the seat cushion.
The seat device according to claim 2, wherein the controller is configured to:
calculate a feature amount related to the body pressure distribution in each of the plurality of monitoring areas based on the body pressure distribution in each of the plurality of monitoring areas; and

estimate the position of the sole of the right foot based on the feature amount.
The seat device according to claim 3, wherein the feature amount includes a centroid position in each of the plurality of monitoring areas.
The seat device according to claim 3, wherein the feature amount includes a distance between a centroid position in each of the plurality of monitoring areas and a reference point set on the seat surface.
The seat device according to claim 3, wherein the feature amount includes an angle on the seat surface of a centroid position in each of the plurality of monitoring areas, the angle being centered on a reference point set on the seat surface.
The seat device according to claim 3, wherein the feature amount includes, in each of the plurality of monitoring areas, a ratio of an area where pressure is equal to or higher than a determination value.
The seat device according to claim 3, wherein the controller is configured to estimate the position of the sole of the right foot based on the feature amount using a machine learning model that receives the feature amount as input and outputs the position of the sole of the right foot.
The seat device according to claim 1, wherein the seat main body is installed in a vehicle,
the operation pedal is a brake pedal of the vehicle, and

the controller is configured to determine a hazard prediction level of the seated person based on the position of the sole of the right foot when the controller determines a risk of collision between the vehicle and an obstacle.
The seat device according to claim 9, wherein the controller is configured to, when the hazard prediction level is equal to or lower than a prescribed determination value, estimate a cause of the hazard prediction level being equal to or lower than the prescribed determination value based on the body pressure distribution.