JP6401559B2 - Crew state estimation device - Google Patents

Description

  The present invention relates to an occupant state estimation device that estimates an occupant's driving state based on operation data on a steering wheel of an occupant of a vehicle.

  In recent years, in a vehicle such as an automobile, for example, when there is a risk of collision with another vehicle or an obstacle, there is a system that performs driving support such as alarm output, forced braking operation, automatic avoidance steering, etc. Has been developed. In such a system, it is necessary to estimate the driving state of the driver and determine the optimum driving support including an alarm, but in order to estimate the driving state of the driver, accelerator operation, brake operation, steering wheel operation, etc. It is necessary to acquire a huge number of parameters and perform complicated processing.

  For this reason, conventionally, various techniques for estimating the driving state of the driver using relatively simple parameters related to steering have been proposed. For example, Patent Literature 1 includes a seat-like pressure sensor built in a steering wheel, and based on a gripping force calculated by the pressure sensor and a reaction force calculated from a steering torque detected by the torque sensor, Technology for predicting the occurrence of a driver's hand slip in advance and issuing an alarm is disclosed.

JP 2005-119539 A

  However, in Patent Document 1, only the grip force data is acquired as a unique parameter by the driver's own steering. Therefore, the steering wheel operation at a corner, the steering operation with little change in a traffic jam, It is difficult to estimate the driver's driving state such as slow steering operation. For this reason, it is difficult to accurately estimate the driving state of the driver from the steering state of the steering wheel and perform optimal driving support.

  The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide an occupant state estimation device capable of accurately estimating a driving state of a driver from a steering state of a steering wheel and enabling optimal driving support. Yes.

An occupant state estimation device according to the present invention is an occupant state estimation device that estimates a driving state of an occupant based on operation data on a steering wheel of a vehicle occupant, and includes a steering angle detection unit that detects a steering angle of the steering wheel, Handle pressure detection unit for detecting pressure distribution due to gripping of the handle occupant, handle temperature detection unit for detecting temperature distribution due to gripping of the handle occupant, and data relating to the steering angle, pressure distribution and temperature distribution of the handle A data collection unit that collects a group in time series, and an initial data group collected at the data collection unit at the start of operation and a data group during operation are compared, and the data group during operation becomes the initial data group If outside the range for, from the steering angle and the pressure distribution and the temperature distribution of the handle, the strength and grasp grip position and grip area of the handle, the occupant of the handle A gripping state estimation unit for estimating a gripping state, based on a change in the holding state of the handle which is estimated by the grip state estimation unit, the motion direction and the steering speed of the data group and the occupant's hand representing the grip how the handle A data group representing a change, a contact pattern recognition unit that recognizes each extracted data group as a contact pattern for the steering wheel in association with the steering angle of the steering wheel, and a contact recognized by the contact pattern recognition unit And a driving state estimating unit that estimates a driving state of the occupant based on the pattern.

  According to the present invention, the driving state of the driver can be accurately estimated from the steering state of the steering wheel, and optimal driving support can be made possible.

Configuration diagram of occupant state estimation device Passenger state estimation process flowchart Explanatory drawing which shows the operation state of a handle Explanatory drawing showing an example of contact pattern data

  Embodiments of the present invention will be described below with reference to the drawings.

  The occupant state estimation device according to the present embodiment is a device that estimates a driving state of a driver from an operation state of a steering wheel (steering wheel) of the driver (occupant) of the vehicle, and requires various driving assistance based on the estimated driving state. No can be determined. As shown in FIG. 1, such an occupant state estimation device is configured by an electronic control unit (ECU) 10 composed of a microcomputer or the like, and is configured as a dedicated device connected to an in-vehicle network or a part of a vehicle control device. Has been.

  The ECU 10 examines the pressure distribution and temperature distribution of the handle when the driver grips the handle, recognizes a contact pattern based on the handle gripping state, and estimates the driving state of the driver from the recognized contact pattern. For this reason, a steering angle sensor 6 for detecting the steering angle of the handle 5, a pressure sensor 7 for detecting the pressure applied to the handle 5, and a temperature sensor 8 for detecting the temperature of the handle 5 are connected to the ECU 10. The ECU 50 executes processing for realizing a function as an occupant state estimation device based on signals from these sensors and vehicle information such as vehicle speed input via the in-vehicle network.

  The functions of the ECU 10 as an occupant state estimation device include a steering angle detection unit 11, a handle pressure detection unit 12, a handle temperature detection unit 13, a data collection unit 14, a gripping state estimation unit 15, a contact pattern recognition unit 16, and a driving state estimation unit. Represented by 17. In the present embodiment, the ECU 10 includes a control intervention determination unit 18 that determines whether or not a control system intervention including an alarm is necessary according to the driving state estimated by the driving state estimation unit 17. Is output to the vehicle control system 20.

  The steering angle detector 11 detects the steering angle of the handle 5 based on the signal from the steering angle sensor 6. Note that the steering angle information based on the signal from the steering angle sensor 6 may be transmitted to the ECU 10 from another control device such as a vehicle behavior control device connected to the in-vehicle network. The function as the steering angle detection unit 11 of the device is provided in the vehicle behavior control device or the like.

  The handle pressure detector 12 detects the pressure distribution of the handle 5 due to the grip of the driver based on the signal from the pressure sensor 7. The pressure sensor 7 is provided over the entire circumference of the handle 5, and can be configured as a tactile sensor using, for example, MEMS (Micro Electro Mechanical Systems) technology. By using such a tactile sensor, it is possible to detect the force in the compression direction and the force in the sliding direction on the handle surface, and to detect the transition of the grip position, grip time, and grip strength of the handle, It is also possible to detect movements of hands and fingers.

  The handle temperature detector 13 detects the temperature distribution of the handle 5 due to the grip of the driver based on the signal from the temperature sensor 8. The temperature sensor 8 is provided over the entire circumference of the handle 5, and the handle temperature detection unit 13 changes the grip position of the handle 5, changes in the grip time, and grip strength based on a signal from the temperature sensor 8. The temperature distribution corresponding to the change in the height is detected.

  The data collection unit 14 stores a data group related to the steering angle of the steering wheel detected by the steering angle detection unit 11, the pressure distribution of the steering wheel detected by the steering wheel pressure detection unit 12, and the temperature distribution of the steering wheel detected by the steering wheel temperature detection unit 13. Collect serially. A group of data collected during a certain period at the start of driving of the vehicle is stored as initial data specific to each driver and is compared with data collected during driving.

  Initial data is basically collected and discarded for each driving cycle, but if the individual driver can be specified separately, the long-term driving tendency of the individual driver can be grasped while maintaining and updating the initial data. You may make it do.

  The gripping state estimation unit 15 compares the initial data group at the start of operation collected by the data collection unit 14 with the data group during operation, and estimates the gripping state of the handle by the driver. If the data group showing the pressure distribution and temperature distribution of the steering wheel during operation is within the set range with respect to the initial data group, it is judged as a normal operation state and data collection is continued, and it is out of the set range. In this case, it is determined that the driving state is different from the normal driving state, and the gripping state of the driver such as the grip position and area of the steering wheel and the grip strength is estimated from the steering angle, pressure distribution, and temperature distribution of the steering wheel.

  The contact pattern recognition unit 16 recognizes a contact pattern with respect to the handle based on the change in the handle grip state estimated by the grip state estimation unit 15. For example, a group of data representing how to grip a handle by extracting data related to handle grip strength, grip area, difference in grip strength between left and right, and temperature rise of the grip portion from changes in handle pressure distribution and temperature distribution In addition, data relating to the direction and speed of pressure change on the steering wheel is extracted to obtain a data group representing changes in hand movement direction and steering speed. These data groups are associated with the steering angle of the steering wheel and acquired as a contact pattern with respect to the steering wheel during driving.

  The driving state estimation unit 17 estimates the driving state of the driver based on the contact pattern recognized by the contact pattern recognition unit 16. The driving state of this driver is a state estimated from the operation state of the steering wheel, for example, a rapid steering operation when entering a corner, a steering operation with little change in traffic jam, a slow steering operation in a fatigue state, etc. The driving state is estimated for.

  The control intervention determination unit 18 determines whether or not the driving state of the driver estimated by the driving state estimation unit 17 satisfies a condition for intervening vehicle control by the system, and controls the vehicle control system 20 according to the determination result. By giving instructions, we will provide driving assistance that improves safety, safety and comfort. Whether the driving state of the driver satisfies the conditions for control intervention by the system is determined, for example, by setting a threshold value for control intervention based on the characteristics of the vehicle and each parameter indicating the driving state. Judgment can be made. It is also possible to input each parameter representing the driving state into a learning model such as a neural network, and use the output of the learning model as a determination result of necessity of control intervention.

  The functions of the above units are realized by program processing in the ECU 10. Next, the occupant state estimation program process executed by the ECU 10 will be described with reference to the flowchart of FIG. The flowchart in FIG. 2 shows an example of occupant state estimation processing for determining control intervention of the system for safe driving.

  The occupant state estimation process in FIG. 2 is a program process that is executed prior to engine start when the ignition switch is turned on. When this process starts, first, in steps S1 to S4, data on the steering angle, pressure, and temperature of the steering wheel is collected based on signals from the steering angle sensor 6, the pressure sensor 7, and the temperature sensor 8, and the data on the steering wheel is collected. The transition of the grip position of the driver is calculated.

  Specifically, as shown in FIG. 3, the steering angle φ of the steering wheel is obtained as an angle from the reference position O (step S1), and the temperature at the gripping positions H1 and Hr of the steering wheel of the left and right hands of the driver is obtained. Pressure is detected (steps S2 and S3). Then, from the steering angle of the steering wheel and the pressure and temperature of the gripping position, a transition of the hand moving on the steering wheel while receiving a reaction force against the steering force (transition of the gripping position) is calculated (step S4).

  In this case, immediately after the ignition switch is turned on, the steering wheel temperature data is collected in advance at multiple points around the entire circumference of the steering wheel, and the average value is compared with the temperature at the position held by the driver's hand. It is used as reference data. In addition, for a predetermined period (for example, 5 minutes) at the start of operation, rudder angle, pressure, and temperature data are collected as data specific to the driver, and are estimated from the data and collected data collected during this period. Gripping strength and area of the steering wheel, hand movements of the steering wheel (steering direction and steering speed), and gripping state such as left and right hand holding, as well as the actual steering angle of the steering wheel (including the steering direction), as an initial data group of the individual driver It is held until the operation ends.

  After collecting the steering wheel steering angle, gripping position pressure and temperature data and calculating the transition of the gripping position, in Steps S5 to S7, these data groups are compared with the initial data group to grip the handle of the driver. Estimate the state. The temperature at the handle gripping position is compared with the average handle temperature (step S5), and if the difference between the two is within the set range, data collection is continued assuming that there is no particular change from the normal driving state.

  On the other hand, when the difference between the handle temperature at the gripping position and the average handle temperature is outside the set range, it is further checked whether the handle gripping pressure is greater than the initial data (step S6). When the difference between the gripping pressure and the initial data is within the set range, data collection is continued assuming that there is no particular change from the normal driving state, and when the difference between the two is outside the set range, the driver makes the handle stronger than usual. Assuming that the driver may be gripping (or weak) and that the driver may be unnaturally stressed (or lack of concentration), the transition on the handle of the driver's hand It is checked whether or not the range is exceeded (step S7).

  As a result, when the hand transition on the steering wheel is within the range of the initial data, the data collection is continued as being within the range of the normal driving state. On the other hand, when the transition state of the hand on the handle exceeds the transition speed or range of normal driving, it is estimated that the driver is operating the handle in a grip state different from the normal state, and the contact pattern is determined from the handle grip state. Is recognized (step S8).

  For example, as shown in FIG. 4, as the driver's steering wheel operation during turning of the vehicle, the pressure (gripping strength) P of the gripping portion suddenly increases, and the steering angle φ peaks at the peak of the pressure P. Thus, when it is estimated from the transition of the grip position on the handle (the movement of the hand on the handle) that the reaction force F received from the handle suddenly increases with respect to the steering force of the handle, Is recognized as a contact pattern when entering the corner.

  Based on the recognized contact pattern, it is estimated whether the driving state of the driver is a sudden steering operation with entry to the corner, a steering operation with little change in traffic congestion, a slow steering operation in a fatigued state, etc. Then, it is determined whether or not the operating state requires control intervention by the system (step S9). As a result, if control intervention is required, the vehicle control system 20 is instructed to perform control intervention (step S10).

  For example, when the driving state of the driver is in a driving state where safety cannot be ensured with respect to the driving environment of the vehicle recognized by a captured image of a laser radar or an in-vehicle camera, switching to automatic driving including alarm, brake control, and steering control An instruction to start control intervention such as an instruction is transmitted to the vehicle control system 20. On the other hand, if it is determined that the driving state of the driver is different from the normal driving state but is not sufficient to implement the control intervention, for example, the driver is informed that the driving state is different from the normal driving state. (Step S11), the process is exited.

  As described above, in this embodiment, the grip state of the handle is estimated from the relationship with the actual steering angle by examining the pressure and temperature distribution of the part where the driver on the handle actually contacts, and the change in the grip state The driving state of the driver is estimated from the contact pattern recognized based on the above. This makes it possible to accurately estimate the driver's driving state, such as sudden steering operation at corners, steering operation with little change in traffic jams, and slow steering operation in fatigued conditions, and optimal driving including warnings Support can be made possible.

DESCRIPTION OF SYMBOLS 5 Handle 6 Steering angle sensor 7 Pressure sensor 8 Temperature sensor 10 Electronic controller 11 Steering angle detection part 12 Handle pressure detection part 13 Handle temperature detection part 14 Data collection part 15 Grasping state estimation part 16 Contact pattern recognition part 17 Driving | running state estimation part 17 18 Control Intervention Determination Unit 20 Vehicle Control System

Claims (2)

An occupant state estimation device that estimates an occupant's driving state based on operation data for a steering wheel of a vehicle occupant,
A steering angle detector for detecting a steering angle of the steering wheel;
A handle pressure detector for detecting a pressure distribution due to gripping of an occupant of the handle;
A handle temperature detecting unit for detecting a temperature distribution by gripping an occupant of the handle;
A data collection unit for collecting a data group related to the steering angle, pressure distribution, and temperature distribution of the steering wheel in time series;
When the initial data group collected at the data collection unit and the initial data group at the start of driving are compared with the initial data group, the steering angle of the steering wheel the pressure distribution and the temperature distribution, the intensity and grasp grip position and grip area of the handle, the grip state estimation unit for estimating a gripping state by the occupant of the handle and,
Based on the change in the gripping state of the handle estimated by the gripping state estimation unit, a data group that represents how to grip the handle and a data group that represents a change in the movement direction and steering speed of the passenger's hand are extracted and extracted. A contact pattern recognition unit that recognizes each data group as a contact pattern with respect to the steering wheel in association with the steering angle of the steering wheel ,
An occupant state estimation device comprising: a driving state estimation unit that estimates a driving state of the occupant based on the contact pattern recognized by the contact pattern recognition unit. Furthermore, the occupant state estimating apparatus according to claim 1, wherein the operating state of the occupant estimated by the driving state estimation unit comprises a control intervention determination unit that determines whether or not the condition for the intervention of the vehicle control.

Images