Driver estimation device, control method, program, and storage medium

Abstract

The present invention provides a driver estimation device that can estimate the driver without requiring any special operations or procedures. [Solution] The control unit 21 of the navigation device 2 receives arm motion data Da, which includes time-series measurement data of acceleration and angular velocity indicating the movement of the occupant's arm, from the arm-mounted terminal 1 worn by the occupant of the vehicle via the communication unit 27. Based on the arm motion data Da of the occupant received by the communication unit 27, the control unit 21 of the navigation device 2 estimates whether the occupant is the driver of the vehicle.

Description

Technical Field

[0001] The present invention relates to a technique for estimating a driver.

Background Art

[0002] Conventionally, a technique for performing driver authentication at the start of vehicle operation has been known. For example, Patent Document 1 discloses a technique for performing driver authentication at the start of vehicle operation by receiving an input of a driver's name using an input device such as a touch screen or performing biometric authentication such as fingerprint authentication.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The driving support device described in Patent Document 1 requires an input operation or a contact operation for driver authentication. These operations etc. may be troublesome for the driver and may prevent the spread of the device.

[0005] The present invention has been made to solve the above problems, and a main object thereof is to provide a driver estimation device capable of performing driver estimation without requiring special operations etc. for the driver.

Means for Solving the Problems

[0006] The invention described in the claims is a driver estimation device comprising: a receiving unit that receives time-series data relating to the movement of an arm from a first sensor worn by a vehicle occupant and which outputs a detection signal corresponding to the movement of the occupant's arm; and an estimation unit that estimates whether the occupant is the driver of the vehicle based on the time-series data, wherein the estimation unit detects a driving operation corresponding to the time-series data, determines whether the driving operation matches the driving state of the vehicle identified based on the output from a second sensor mounted on the vehicle, and estimates the occupant to be the driver if it determines that they match.

[0007] The invention described in the claim is a control method performed by a driver estimation device, comprising: a receiving step of receiving time-series data relating to arm movements from a first sensor worn by a vehicle occupant and outputting a detection signal corresponding to the occupant's arm movements; and an estimation step of estimating whether the occupant is the driver of the vehicle based on the time-series data, wherein the estimation step detects a driving operation corresponding to the time-series data, determines whether the driving operation matches the driving state of the vehicle identified based on the output from a second sensor mounted on the vehicle, and estimates the occupant to be the driver if it is determined that they match.

[0008] The invention described in the claims is characterized in that a receiving unit receives time-series data relating to the movement of the arm from a first sensor worn by a vehicle occupant and which outputs a detection signal corresponding to the movement of the occupant's arm, and a computer functions as an estimation unit that estimates whether the occupant is the driver of the vehicle based on the time-series data, the estimation unit detects driving operations corresponding to the time-series data, determines whether the driving operation matches the driving state of the vehicle identified based on the output from a second sensor mounted on the vehicle, and estimates the occupant to be the driver if it determines that they match. [Brief explanation of the drawing]

[0009] [Figure 1]This is an overview diagram of the driver estimation system. [Figure 2] This flowchart shows an example of the driver estimation process. [Figure 3] This is a schematic diagram of a driver estimation system when multiple arm-mounted devices are present in a vehicle. [Figure 4] This table shows the relationship between driving actions and the vehicle's driving state. [Figure 5] This flowchart shows an example of the procedure for driver estimation processing related to a modified example. [Modes for carrying out the invention]

[0010] In one preferred embodiment of the present invention, the driver estimation device includes a receiving unit that receives data relating to arm movements from a terminal worn by a vehicle occupant and which detects the occupant's arm movements, and an estimation unit that estimates whether the occupant is the driver of the vehicle based on the data relating to arm movements.

[0011] The driver estimation device described above comprises a receiving unit and an estimation unit. The receiving unit receives data on arm movements from a terminal worn by a vehicle occupant that detects the occupant's arm movements. The estimation unit estimates whether the occupant is the driver of the vehicle based on the arm movement data received by the receiving unit. In this embodiment, the driver estimation device receives data on arm movements from a terminal worn by the occupant and estimates whether the occupant is the driver based on this data. Generally, there is a clear difference in arm movements between the driver of a vehicle and an occupant who is not driving, due to factors such as whether or not they are performing driving operations. Therefore, in this embodiment, the driver estimation device can suitably estimate whether or not the occupant wearing the terminal is the driver without requiring any special operations.

[0012] In one embodiment of the driver estimation device described above, the driver estimation device further comprises an acquisition unit that acquires driving motion data, which is pre-created data relating to arm movements during driving, and the estimation unit estimates the driver based on the driving motion data and the data relating to arm movements. In this embodiment, the driver estimation device can suitably estimate whether or not the passenger wearing the terminal is the driver by receiving data relating to arm movements from the terminal.

[0013] In another embodiment of the driver estimation device described above, the estimation unit estimates whether the occupant is the driver of the vehicle based on data relating to the arm movements and the vehicle's driving state. Generally, the driving operations that a driver may perform are limited depending on the vehicle's driving state. Therefore, in this embodiment, the driver estimation device can more accurately estimate whether the occupant is the driver.

[0014] In another embodiment of the driver estimation device described above, the estimation unit estimates the driver of the vehicle when the vehicle is stopped for a predetermined time. The "predetermined time" is set based on experiments, for example, to the time required for a driver change. In this embodiment, the driver estimation device can accurately estimate the driver after a driver change by estimating the driver of the vehicle when there is a possibility that the driver has changed.

[0015] In another embodiment of the driver estimation device described above, the driver estimation device further comprises a door opening / closing detection unit that detects the opening and closing of the vehicle's doors, and the estimation unit estimates the vehicle's driver when the vehicle's doors are closed. In this embodiment, the driver estimation device estimates the vehicle's driver because there is a possibility that the driver has changed when the doors are opened or closed. This makes it possible to accurately estimate the driver after the change, even if the driver has been changed.

[0016] In another embodiment of the present invention, the driver estimation device includes a receiving unit that receives the result of an estimation from a terminal worn by a vehicle occupant, which detects the movement of the occupant's arm to estimate whether the occupant is the driver of the vehicle, and an estimation unit that estimates whether the occupant is the driver of the vehicle based on the result of the estimation, or estimates whether the occupant is the driver of the vehicle based on the result of the estimation and data relating to the arm movement received by the receiving unit along with the result of the estimation. In this embodiment, the driver estimation device can suitably estimate the driver from the estimation result received from a terminal that estimates whether the wearer is the driver of the vehicle. Furthermore, even if there is doubt about the reliability of the received estimation result, the driver estimation device can also perform its own estimation process based on the data relating to the arm movement received along with the estimation result.

[0017] In another embodiment of the present invention, the driver estimation device comprises a detection unit worn by a vehicle occupant and detecting the movement of the occupant's arm, and an estimation unit that estimates whether the occupant is the driver of the vehicle based on the arm movement detected by the detection unit. In this embodiment, the driver estimation device can suitably estimate whether the wearer is the driver without requiring any special operation.

[0018] In yet another embodiment of the present invention, the control method performed by the driver estimation device includes a receiving step of receiving data relating to arm movements from a terminal worn by a vehicle occupant and used to detect the occupant's arm movements, and an estimation step of estimating whether the occupant is the driver of the vehicle based on the data relating to arm movements. By performing this control method, the driver estimation device can estimate whether the occupant wearing the terminal is the driver without requiring any special operations.

[0019] In yet another embodiment according to the present invention, a program executed by a computer causes the computer to function as a receiving unit that receives data regarding the movement of an arm from a terminal worn by a passenger of a vehicle and that detects the movement of the arm of the passenger, and an estimation unit that estimates whether or not the passenger is the driver of the vehicle based on the data regarding the movement of the arm. By executing this program, the computer can estimate whether or not a passenger wearing the terminal is the driver without requiring a special operation or the like. Preferably, the above program is stored in a storage medium.

Example

[0020] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

[0021] [Configuration of Driver Estimation System] FIG. 1 is a schematic diagram of a driver estimation system according to the present embodiment. As shown in FIG. 1, the driver estimation system includes a wrist-worn terminal 1 worn on the arm of a passenger of a vehicle such as a driver, and a navigation device 2. The wrist-worn terminal 1 and the navigation device 2 perform data communication wirelessly or by wire (wireless in FIG. 1).

[0022] The wrist-worn terminal 1 is a wearable terminal that can be worn on the arm such as a smartwatch, and transmits data representing the movement of the wearer's arm (also referred to as "arm movement data Da") to the navigation device 2 based on a request from the navigation device 2. The wrist-worn terminal 1 may be a terminal manufactured specifically for this driver estimation system, or may be a general-purpose terminal installed with a program for conforming to this driver estimation system.

[0023] Navigation device 2 is a stationary in-vehicle unit or a portable terminal such as a PND (Portable Navigation Device) or smartphone, and performs route guidance to a destination specified by the user. In this embodiment, navigation device 2 focuses on the fact that there is a difference in arm movements between the driver and non-drivers of the vehicle, and performs a process (also called "driver estimation process") to determine whether the wearer of arm-mounted terminal 1 is the driver or not, based on arm motion data Da received from arm-mounted terminal 1.

[0024] Next, referring to Figure 1, we will describe the configurations of the arm-mounted terminal 1 and the navigation device 2.

[0025] As shown in Figure 1, the arm-mounted terminal 1 mainly comprises a control unit 11, a communication unit 12, an acceleration sensor 13, a gyro sensor 14, and a storage unit 15. The communication unit 12 communicates data with the navigation device 2 based on the control of the control unit 11.

[0026] The acceleration sensor 13 is, for example, a 3-axis acceleration sensor, and generates detection signals corresponding to the wearer's arm movement, posture (tilt), and vibration. The gyro sensor 14 generates a detection signal indicating the angular velocity of the rotational movement when the wearer's arm rotates. The memory unit 15 stores programs for controlling the operation of the arm-mounted terminal 1 and holds information necessary for the operation of the arm-mounted terminal 1. The memory unit 15 also stores identification information unique to the arm-mounted terminal 1 (also called the "terminal ID").

[0027] The control unit 11 of the arm-mounted terminal 1 includes a CPU (not shown) and performs various controls on each component within the arm-mounted terminal 1. In this embodiment, the control unit 11 receives a request from the navigation device 2 to transmit arm motion data Da via the communication unit 12. In this case, the control unit 11 transmits the time-series data of acceleration and angular velocity measured by the acceleration sensor 13 and the gyro sensor 14 over a predetermined time period (also called "time-series measurement data") and the terminal ID stored in the storage unit 15 as arm motion data Da to the navigation device 2 via the communication unit 12.

[0028] Next, the configuration of the navigation device 2 will be described. As shown in Figure 1, the navigation device 2 mainly consists of a control unit 21, a positioning unit 22, a display unit 23 such as a display, an audio output unit 24 such as a speaker, an input unit 25, a storage unit 26, and a communication unit 27.

[0029] The positioning unit 22 is an autonomous positioning device such as a gyro sensor or distance sensor, or a GPS receiver, and performs positioning such as the vehicle's current location and direction of travel. The display unit 23 displays maps for route guidance, etc., based on the control of the control unit 21. The sound output unit 24 outputs voice guidance for route guidance, predetermined warnings, etc., based on the control of the control unit 21. The input unit 25 is a button, switch, touch panel, voice input device, etc., and supplies the input information to the control unit 21. The communication unit 27 transmits a request signal for arm movement data Da to the arm-mounted terminal 1 and receives arm movement data Da transmitted from the arm-mounted terminal 1, based on the control of the control unit 21. The communication unit 27 is an example of a "receiving unit" in this invention.

[0030] The memory unit 26 stores programs for controlling the operation of the navigation device 2 and holds information necessary for the operation of the navigation device 2. For example, the memory unit 26 stores map data, including road data with registered road types. The memory unit 26 also pre-stores a time-series set of output data from the acceleration sensor 13 and gyro sensor 14 (also called "driver operation data Dtmp") when an unspecified wearer of the arm-mounted terminal 1 performs driver-specific arm movements. Driver-specific movements include various driving actions such as turning the steering wheel, operating the shift lever, and operating the turn signals, and the memory unit 26 stores driver operation data Dtmp corresponding to each of these driving actions. Driver operation data Dtmp is, for example, time-series measurement data of acceleration and angular velocity measured when a person of average build performs driving actions in a vehicle of standard size.

[0031] The control unit 21, which includes a CPU (not shown), performs various controls on each component within the navigation device 2. In this embodiment, when the control unit 21 receives arm motion data Da from the arm-mounted terminal 1, it determines whether the wearer of the arm-mounted terminal 1 is the driver based on the time-series measurement data of acceleration and angular velocity included in the arm motion data Da. If the control unit 21 determines that the wearer of the arm-mounted terminal 1 is the driver, it stores the terminal ID included in the received arm motion data Da in the storage unit 26 as the terminal ID of the arm-mounted terminal 1 worn by the driver. Subsequently, for example, the control unit 21 transmits a control signal to the arm-mounted terminal 1 corresponding to the terminal ID stored in the storage unit 26, instructing it to perform a predetermined operation to assist the driver's driving. The control unit 21 is an example of the "estimation unit" and the "computer" that executes the program in this invention.

[0032] [Driver estimation process] Next, the details of the driver estimation process performed by the control unit 21 of the navigation device 2 will be described. In general terms, the control unit 21 estimates the wearer of the arm-mounted terminal 1 to be the driver if it determines, based on the time-series measurement data of acceleration and angular velocity included in the arm movement data Da, that the wearer of the arm-mounted terminal 1 has performed a driving operation.

[0033] In this case, the control unit 21 uses known pattern recognition technology to determine whether the driver operation data Dtmp for each predetermined driving operation is similar to the time-series measurement data of acceleration and angular velocity received from the arm-mounted terminal 1. The control unit 21 then estimates that the wearer of the arm-mounted terminal 1, which is the source of the arm operation data Da, is the driver if the time-series measurement data is similar to or nearly identical to the driver operation data Dtmp corresponding to any of the driving operations. In this case, instead of directly comparing the driver operation data Dtmp and the time-series measurement data of acceleration and angular velocity received from the arm-mounted terminal 1, the control unit 21 may extract characteristic feature quantities of a predetermined dimension using known pattern recognition technology and perform the similarity determination by comparing the extracted feature quantities.

[0034] Preferably, the control unit 21 stores the time-series measurement data of acceleration and angular velocity used in the driver estimation process in the storage unit 26, and uses the stored time-series measurement data in place of the driver operation data Dtmp in the driver estimation process. In this case, if the control unit 21 estimates that the wearer of the arm-mounted terminal 1 is the driver based on the driver estimation process, it stores the time-series measurement data of acceleration and angular velocity used in the driver estimation process in the storage unit 26. Then, for example, if the control unit 21 has stored the time-series measurement data of acceleration and angular velocity more than a predetermined number of times, it uses the stored data in place of the driver operation data Dtmp to estimate the driver. In this case, the control unit 21 may determine for each driving operation whether or not stored data exists and how many times it has been stored, and then determine whether or not to use it in place of the driver operation data Dtmp.

[0035] Preferably, when the control unit 21 uses the stored data instead of the driver operation data Dtmp, it is preferable to perform known learning on the data stored for each driving operation to model the time-series changes in acceleration and angular velocity for each driving operation. In this case, after receiving the arm movement data Da, the control unit 21 determines whether the wearer of the arm-mounted terminal 1 is the driver by determining whether the modeled time-series changes in acceleration and angular velocity for each driving operation are similar to the time-series measurement data included in the arm movement data Da.

[0036] Here, we will provide a supplementary explanation regarding the effects of accumulating and utilizing time-series measurement data of acceleration and angular velocity. Generally, arm movements based on driving actions differ depending on the type of vehicle being driven (e.g., passenger car, large vehicle, bus), even for the same driving action, and there are also individual differences among drivers. On the other hand, the combination of the navigation device 2 and the vehicle, and the combination of the navigation device 2 and the driver are generally fixed or limited. Taking the above into consideration, the control unit 21 accumulates time-series measurement data of acceleration and angular velocity and utilizes it as a sample (template) for driver estimation. As a result, the navigation device 2 can perform driver estimation processing accurately and in a relatively short time, according to the type of vehicle and the individuality of the driver's driving actions.

[0037] [Processing flow] Figure 2 is a flowchart showing an example of the driver estimation process procedure according to this embodiment.

[0038] First, the navigation device 2 determines whether or not it has detected that the engine switch of the vehicle in which the navigation device 2 is installed has been turned on (step S101). For example, the navigation device 2 determines that the vehicle's engine switch has been turned on when power is supplied from the vehicle, or when it receives information from the vehicle's control unit (ECU: Electronic Control Unit) via CAN (Controller Area Network) or the like that the engine switch has been turned on.

[0039] Then, if it detects that the engine switch has been turned on (step S101; Yes), the navigation device 2 sends a request signal for arm motion data Da to the arm-mounted terminal 1 (step S102). In this case, the navigation device 2 may send the above request signal unspecified by broadcast, or it may send it to an arm-mounted terminal 1 if one has already established communication with that arm-mounted terminal 1.

[0040] When the arm-mounted terminal 1 receives a request signal for arm motion data Da from the navigation device 2, it measures arm motion using the acceleration sensor 13 and the gyro sensor 14 (step S103). The arm-mounted terminal 1 then transmits the arm motion data Da, which includes the time-series measurement data from the acceleration sensor 13 and the gyro sensor 14, and the terminal ID of the arm-mounted terminal 1, to the navigation device 2 (step S104). The navigation device 2 then receives the arm motion data Da (step S105).

[0041] Next, the navigation device 2 determines whether it can be estimated that the wearer of the arm-mounted terminal 1, which is the source of the arm movement data Da, is the driver, based on the time-series measurement data of acceleration and angular velocity included in the received arm movement data Da (step S106). Specifically, the navigation device 2 determines whether any driving operation has been performed by determining the similarity between the driver operation data Dtmp for each driving operation and the time-series measurement data of acceleration and angular velocity. In this case, if the navigation device 2 has accumulated time-series measurement data of acceleration and angular velocity that was stored in the storage unit 26 when it was estimated that the wearer was the driver in a previous driver estimation process, it may determine whether any driving operation has been performed based on the accumulated data.

[0042] Then, if the navigation device 2 can estimate that the wearer of the arm-mounted terminal 1, which is the source of the arm movement data Da, is the driver (step S106; Yes), it considers the wearer of the arm-mounted terminal 1, which is the source of the arm movement data Da, to be the driver and stores the terminal ID included in the arm movement data Da in the storage unit 26 (step S107). On the other hand, if the navigation device 2 cannot estimate that the wearer of the arm-mounted terminal 1, which is the source of the arm movement data Da, is the driver (step S106; No), it returns to step S102, receives the arm movement data Da, and determines whether the source of the arm movement data Da is the driver's terminal. Note that if the navigation device 2 cannot estimate that the wearer of the arm-mounted terminal 1 is the driver after a predetermined number of attempts in step S106, it may determine that the wearer of the arm-mounted terminal 1 is not the driver and terminate the flowchart processing without executing step S107.

[0043] In the flowchart of Figure 2, the arm-mounted terminal 1 sent the arm motion data Da to the navigation device 2 in step S103 each time there was a request to send the arm motion data Da in step S102. Alternatively, after the arm-mounted terminal 1 has received one request to send the arm motion data Da in step S102, it may repeatedly generate and send the arm motion data Da until it receives notification from the navigation device 2 that the arm motion data Da is no longer needed.

[0044] [If multiple arm-worn devices are present] Next, we will provide supplementary information regarding the case where multiple arm-mounted terminals 1 are present in the vehicle (i.e., when multiple passengers are wearing arm-mounted terminals 1). Figure 3 shows an overview of the driver estimation system when multiple arm-mounted terminals 1 (1A, 1B, ...) are present in the same vehicle together with the navigation device 2.

[0045] In the case of Figure 3, when arm-mounted terminals 1A and 1B receive a request signal for arm motion data Da from the navigation device 2, they transmit arm motion data Da, including a terminal-specific terminal ID, to the navigation device 2. The navigation device 2 then determines whether the wearer of arm-mounted terminal 1A is a driver based on the arm motion data Da received from arm-mounted terminal 1A, and also determines whether the wearer of arm-mounted terminal 1B is a driver based on the arm motion data Da received from arm-mounted terminal 1B. If the navigation device 2 determines that the wearer of either arm-mounted terminal 1A or 1B is a driver, it stores the terminal ID of the arm-mounted terminal 1 corresponding to the wearer determined to be a driver in the storage unit 26.

[0046] In this way, even when multiple arm-mounted terminals 1 are present in the vehicle, the navigation device 2 determines whether the wearer of each arm-mounted terminal 1 is the driver based on the arm movement data Da of each arm-mounted terminal 1. As a result, the navigation device 2 can accurately estimate the driver even when multiple arm-mounted terminals 1 are present in the vehicle.

[0047] As described above, the control unit 21 of the navigation device 2 according to this embodiment receives arm motion data Da, which includes time-series measurement data of acceleration and angular velocity indicating the movement of the occupant's arm, from the arm-mounted terminal 1 worn by the occupant of the vehicle via the communication unit 27. The control unit 21 of the navigation device 2 then estimates whether the occupant is the driver of the vehicle based on the arm motion data Da received by the communication unit 27. Generally, there is a clear difference in arm movements between the driver of a vehicle and an occupant who is not driving, due to the presence or absence of driving actions. Therefore, in this embodiment, the navigation device 2 can suitably estimate whether the occupant wearing the arm-mounted terminal 1 is the driver or not.

[0048] [Differentiation] The following describes suitable modifications of the above-described embodiments. The following modifications may be applied to the above-described embodiments in any combination.

[0049] (Variation 1) The navigation device 2 may further consider the vehicle's driving conditions when a driving operation is performed and execute driver estimation processing.

[0050] Figure 4 is a table showing the relationship between driving actions and the vehicle's driving state. In the table in Figure 4, each driving action is associated with the expected vehicle driving state when that driving action is performed. For example, when turning the steering wheel, the expected driving states are turning right or left and changing lanes; when operating the shift lever, the expected driving states are moving forward and backward; and when operating the turn signal, the expected driving states are turning right or left and changing lanes. The memory unit 26 stores in advance a table showing the relationship between driving actions and the vehicle's driving state as shown in Figure 4.

[0051] Next, we will explain how to use the table in Figure 4. The control unit 21 of the navigation device 2, similar to the embodiment, detects any driver-specific driving action based on the time-series measurement data of acceleration and angular velocity included in the arm movement data Da, and further recognizes the vehicle's driving state at the time of acceleration and angular velocity measurement. Next, the control unit 21 refers to the table in Figure 4, and if the combination of the detected driving action and the recognized vehicle's driving state is recorded in the table shown in Figure 4, it estimates that the wearer of the arm-mounted terminal 1 is the driver. In other words, even if the control unit 21 detects any driver-specific driving action based on the time-series measurement data of acceleration and angular velocity, if the recognized vehicle's driving state is not associated with the detected driving action in the table in Figure 4, it considers that the driving action has been misdetected.

[0052] Here, we will explain a specific example of a method for recognizing the driving status in order to refer to the table in Figure 4.

[0053] For example, the control unit 21 recognizes the vehicle's driving state, such as turning left or right, moving forward or backward, or changing lanes, based on the output of the positioning unit 22, which includes a gyro sensor that outputs a signal corresponding to a change in the vehicle's direction of travel, a vehicle speed sensor or acceleration sensor for detecting the vehicle's forward or backward movement, and a camera for detecting lane changes, etc. At this time, the control unit 21 may take into account the time lag between the time when a driving operation is detected and the time when the driving operation is actually performed, and store the output of the positioning unit 22 for a predetermined period of time in a buffer such as the storage unit 26, and when a driving operation is detected, recognize the vehicle's driving state at the time the driving operation was performed based on the output of the positioning unit 22 from the predetermined time prior.

[0054] In other examples, the control unit 21 may estimate the vehicle's driving state by obtaining information such as steering angle, shift lever position, and turn signal position from the vehicle's control unit (ECU) via the communication unit 27.

[0055] The combination of driving actions used in the driver estimation process and the vehicle's driving state at the time of the driving action is not limited to the combinations shown in Figure 4. For example, if the navigation device 2 can receive arm motion data Da from the arm-mounted terminal 1 and perform driver estimation processing even when the vehicle's engine is off, the navigation device 2 may estimate that the wearer of the arm-mounted terminal 1 is the driver when it detects an engine start operation while the vehicle is stopped. In this case, the navigation device 2 detects the action of inserting the key and turning the engine start switch based on the received arm motion data Da, and if it determines from the output of the positioning unit 22 that the vehicle was stopped when these actions were detected, it estimates that the wearer of the arm-mounted terminal 1 is the driver. Similarly, the navigation device 2 may estimate that the wearer of the arm-mounted terminal 1 is the driver when it detects the wearer getting into the vehicle. In this case, the navigation device 2 estimates that the wearer of the arm-mounted terminal 1 is the driver when it detects the action of opening the door, sitting in the seat, and fastening the seat belt based on the received arm motion data Da.

[0056] (Modification 2) If the arm-mounted terminal 1 does not have a gyro sensor 14, the navigation device 2 may perform driver estimation processing based solely on the output of the acceleration sensor 13. In this case, the arm-mounted terminal 1 transmits the time-series measurement data of the acceleration sensor 13 to the navigation device 2 as arm motion data Da, and the navigation device 2 performs driver estimation processing based on pre-measured driver motion data Dtmp related to acceleration for each driving operation and the received time-series measurement data of the acceleration sensor 13.

[0057] (Variation 3) Navigation device 2 may perform the driver estimation process again even after the driver has been authenticated, if predetermined conditions are met that indicate there is a possibility that the driver has been replaced.

[0058] For example, if the navigation device 2 detects the opening or closing of a vehicle door, it may determine that the driver has changed. In this case, the navigation device 2 detects the opening or closing of the vehicle door by receiving a signal related to the opening or closing of the door from the vehicle's control unit (ECU) via a protocol such as CAN. When the opening or closing of the vehicle door is detected, the navigation device 2 receives the arm motion data Da by sending a request signal for arm motion data Da to the arm-mounted terminal 1. Then, in step S106 of Figure 2, if the navigation device 2 determines that the wearer of the arm-mounted terminal 1 with a terminal ID different from the terminal ID previously recorded in step S107 is the driver, it rewrites the recorded terminal ID of the arm-mounted terminal 1 worn by the driver. Note that in step S106, if the navigation device 2 could not determine that the wearer of the arm-mounted terminal 1 with a terminal ID previously recorded in step S107 is the driver, it may erase the previously recorded terminal ID in step S107.

[0059] In another example, the navigation device 2 may determine that the driver may have changed if it detects that the vehicle has started moving after stopping in a parking lot (including highway service areas and parking areas) registered in the map data, or if it detects that the vehicle has started moving after stopping for a predetermined time in a location other than a parking lot, and may perform the driver estimation process again.

[0060] Thus, according to this modified example, the navigation device 2 can recognize the driver who has been replaced, even if the driver changes between the time the engine switch is turned on and the time the engine switch is turned off again. The control unit 21 is an example of the "door opening / closing detection unit" in the present invention.

[0061] (Modification 4) Navigation device 2 may receive arm movement data Da via a communication unit (not shown, also called "vehicle communication unit") under the control of the vehicle's control unit. In this case, the arm-mounted terminal 1 and the vehicle communication unit, and the navigation device 2 and the vehicle communication unit each establish communication, and the vehicle communication unit forwards the transmission request for arm movement data Da received from the navigation device 2 to the arm-mounted terminal 1, and forwards the arm movement data Da received from the arm-mounted terminal 1 to the navigation device 2.

[0062] (Variation 5) In the example shown in Figure 3, the arm-mounted terminal 1 transmitted arm motion data Da in step S104 based on a request for transmission of arm motion data Da from the navigation device 2. Alternatively, the arm-mounted terminal 1 may generate arm motion data Da and transmit it to the navigation device 2 when it recognizes that the wearer has boarded the vehicle, based on communication with the vehicle's control unit or based on the output of a sensor (not shown).

[0063] (Experimental variation 6) The vehicle control unit (ECU) may perform the driver estimation process that the navigation device 2 performed in this embodiment, instead of the navigation device 2.

[0064] In this case, the vehicle control unit, for example, uses the vehicle communication unit described in Modification 4 to exchange arm motion data Da, etc. with the arm-mounted terminal 1, and stores driver motion data Dtmp, etc. necessary for driver estimation processing in a storage unit (not shown). The vehicle control unit then executes the processes in steps S101 to S102 and S105 to S107 of Figure 2. In this case, the vehicle communication unit is an example of the "receiving unit" in the present invention, and the vehicle control unit is an example of the "estimation unit" and the "computer" that executes the program in the present invention.

[0065] (Example 7) The driver estimation system consists only of the arm-mounted terminal 1, and the processing of the navigation device 2 may be performed by the arm-mounted terminal 1.

[0066] Figure 5 is an example flowchart showing the procedure for driver estimation processing when the arm-mounted terminal 1 performs the processing of the navigation device 2 in the embodiment.

[0067] First, the arm-mounted terminal 1 determines whether or not it has detected the wearer getting into the vehicle (step S201). For example, the arm-mounted terminal 1 determines that the wearer has gotten into the vehicle when it receives predetermined information from the vehicle's control unit, such as that the engine switch has been turned on or that a door has been opened or closed, or when it detects predetermined user input. If the arm-mounted terminal 1 detects that the wearer has gotten into the vehicle (step S201; Yes), it measures arm movements using the acceleration sensor 13 and the gyro sensor 14 (step S202). Then, similar to step S106 in Figure 3, the arm-mounted terminal 1 refers to the driver operation data Dtmp etc. stored in the memory unit 15 and determines whether or not the wearer can be presumed to be the driver (step S203). If the arm-mounted terminal 1 can be presumed to be the driver (step S203; Yes), it switches to a mode that assumes the wearer is the driver (step S204). In this case, the arm-mounted terminal 1 outputs, for example, guidance or warnings to assist driving.

[0068] (Variation 8) The arm-mounted terminal 1 may, instead of the navigation device 2, determine whether the wearer of the terminal is the driver, and if it determines that the wearer is the driver, it may transmit the terminal ID and other information of the arm-mounted terminal to the navigation device 2.

[0069] In this case, the arm-mounted terminal 1, similar to the modification 7, estimates whether the wearer is a driver, and if it estimates that the wearer is a driver, it transmits the estimation result information along with the terminal ID to the navigation device 2. When the navigation device 2 receives the estimation result information from the arm-mounted terminal 1, it refers to this information, considers the wearer of the arm-mounted terminal 1 to be a driver, and stores the terminal ID of the arm-mounted terminal 1 in the storage unit 26.

[0070] In this case, preferably, the arm-mounted terminal 1 transmits the time-series measurement data of acceleration and angular velocity used in the driver estimation process to the navigation device 2 along with the estimation result information, and causes the navigation device 2 to perform driver estimation processing as appropriate. In this case, for example, if the driver cannot be identified even with the received driver estimation result, the navigation device 2 performs driver estimation processing in the same manner as in the embodiment based on the time-series measurement data of acceleration and angular velocity received from the arm-mounted terminal 1. Then, if the navigation device 2 can estimate that the wearer of the arm-mounted terminal 1 is the driver, it stores the terminal ID of the arm-mounted terminal 1 in the storage unit 26. The above-mentioned "cases where the driver cannot be identified" include, for example, cases where the driver estimation results received from multiple arm-mounted terminals 1 indicate that the wearer of each arm-mounted terminal 1 is the driver, making it impossible to identify a single driver. In this case, the navigation device 2 identifies one driver from among multiple candidate drivers by determining which of the time-series measurement data of acceleration and angular velocity received from each arm-mounted terminal 1 is most similar to the driver operation data Dtmp or the measurement data stored in the storage unit 26 during driver authentication.

[0071] In modifications 7 and 8, the arm-mounted terminal 1 is an example of the "driver estimation device" in the present invention, and the control unit 11 is an example of the "detection unit" and "estimation unit" in the present invention.

[0072] (Extreme variation 9) Navigation device 2 may be an in-vehicle device used for purposes other than route guidance (for example, a drive recorder). [Explanation of symbols]

[0073] 1. Arm-worn terminal 2 Navigation system 11, 21 Control Unit 12, 27 Communications Department 13. Accelerometer 14. Gyroscope sensor 15, 26 Storage section 23 Display section 24. Sound output section 25 Input section

Claims

1. A receiving unit receives time-series data relating to the arm movements from a first sensor worn by a vehicle occupant and which outputs a detection signal corresponding to the occupant's arm movements. An estimation unit that estimates whether the passenger is the driver of the vehicle based on the aforementioned time-series data, Equipped with, The driver estimation device is characterized in that the estimation unit detects driving operations corresponding to the time-series data, determines whether the driving state of the vehicle identified based on the output from a second sensor mounted on the vehicle matches the driving operations, and estimates the occupant to be the driver if it determines that they match.

2. In the driver estimation device according to claim 1, The estimation unit maintains a table showing the correspondence between the driving operation and the vehicle's driving state. A driver estimation device characterized by determining whether or not it matches the driving state by referring to the table when detecting the aforementioned driving operation.

3. In the driver estimation device according to claim 2, The table showing the aforementioned correspondence includes: (i) Items that associate the driving action being steering with the driving state being a right turn, a left turn, or a lane change, and (ii) an item that associates the driving operation being a shift operation with the driving state being forward or reverse, including at least one of the above. A driver estimation device characterized by the following features.

4. In the driver estimation device according to claim 1, The output of the second sensor includes a vehicle signal obtained from the vehicle's control unit. The vehicle signal includes at least one of the steering angle, shift position, and turn signal activation status. The estimation unit determines whether the driving state of the vehicle identified based on the vehicle signal matches the driving operation. A driver estimation device characterized by the following features.

5. In the driver estimation device according to any one of claims 1 to 4, When the estimation unit detects that the vehicle's doors have been opened or closed, or that the vehicle has started moving after being stopped for a predetermined period of time, it re-executes the process of determining whether the vehicle's driving state matches the driving operation. A driver estimation device characterized by the following features.

6. A control method performed by a driver estimation device, A receiving step of receiving time-series data relating to the arm movements from a first sensor worn by a vehicle occupant and which outputs a detection signal corresponding to the occupant's arm movements, An estimation step of estimating whether the passenger is the driver of the vehicle based on the aforementioned time-series data, It has, The control method is characterized in that the estimation step detects a driving operation corresponding to the time-series data, determines whether the driving operation matches the driving state of the vehicle identified based on the output from a second sensor mounted on the vehicle, and estimates the occupant to be the driver if a match is determined.

7. A receiving unit receives time-series data relating to the arm movements from a first sensor worn by a vehicle occupant and which outputs a detection signal corresponding to the occupant's arm movements. Based on the aforementioned time-series data, the computer is made to function as an estimation unit that estimates whether the passenger is the driver of the vehicle. The estimation unit is characterized by detecting driving operations corresponding to the time-series data, determining whether the driving operation matches the vehicle's driving state identified based on the output from a second sensor mounted on the vehicle, and estimating the occupant to be the driver if a match is determined.

8. A computer-readable storage medium storing the program described in claim 7.

Images