Position estimation system and position estimation method
By setting up multiple communication machines on the vehicle, using wireless communication to obtain the signal reception strength and flight time correlation values of the mobile device, and using correction function to correct the flight time correlation values, solving the problem of position estimation error and data increase in the electronic key system. , achieving higher precision mobile device position estimation.
Patent Information
- Application Number
- CN202380034437.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-04-22
- Filing Date
- 2023-04-12
- Publication Date
- 2024-11-26
AI Technical Summary
When estimating the position of a mobile device relative to a vehicle, existing electronic key systems have problems such as increasing position error caused by signal diffraction and obstacles, and increasing data volumes lead to increased accuracy.
By setting up multiple communication machines on the vehicle, wireless communications are used to obtain the signal reception intensity and time-of-flight correlation values of the mobile device, and correct the time-of-flight correlation values to estimate the position of the mobile device, reducing the amount of data and improving accuracy.
It realizes more precise estimation of mobile device locations with less data, reduces position estimation errors, and improves system accuracy.
Smart Images

Figure CN119032292A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims the benefit of Japanese Patent Application No. 2022-070809 filed in Japan on April 22, 2022, the entire contents of which are incorporated herein by reference. Technical Field
[0003] The present disclosure relates to a system mounted on a vehicle for use, and is a position estimation system and a position estimation method for estimating the relative position of a mobile device with respect to the vehicle based on the reception status of a wireless signal transmitted from a mobile device carried by a user. Background Art
[0004] Conventionally, there is known an electronic key system that allows or executes a predetermined process for a vehicle, such as locking or unlocking the vehicle, or starting the engine, based on wireless communication between an onboard device and an electronic key. In such an electronic key system, the process is allowed or executed based on the position of the electronic key relative to the vehicle. To do this, it is necessary to estimate the position of the electronic key.
[0005] As a method for determining the position of the electronic key relative to the vehicle, there is a method using the received signal strength of the vehicle signal received by the electronic key. Specifically, the position of the electronic key is determined based on a combination of the received signal strength when the electronic key receives the vehicle signal transmitted from one vehicle-mounted antenna and the received signal strength when the electronic key receives the vehicle signal transmitted from another vehicle-mounted antenna. This is because the received signal strength becomes weaker as the distance from the transmission source increases, so it is related to the distance.
[0006] An example of a method for improving the accuracy of estimating the position of an electronic key is disclosed in Patent Document 1. In the electronic key system described in Patent Document 1, the individual difference in the reception sensitivity of the electronic key is taken into consideration, and the correlation between the reception sensitivity of the electronic key and the received signal strength is used to estimate the position of the electronic key with the corrected received signal strength.
[0007] Another method of determining the position of an electronic key is a distance measurement method using the time of flight (ToF) of a wireless signal.
[0008] Patent Document 1: Japanese Patent Application Publication No. 2019-85734
[0009] As a method of estimating the position of an electronic key, in a distance measurement method utilizing ToF, when there is an obstacle such as a vehicle body or a human body between an antenna mounted on a vehicle and the electronic key, there is a problem that the measured distance is prolonged due to diffraction of radio waves or shadows of obstacles, etc., and only reflected waves are detected instead of direct waves, and the error in position estimation increases.
[0010] When estimating the position of the electronic key, the position estimation accuracy is improved by using the measured data acquired in advance and comparing the measured data with the measured data. Although the more measurement points of the measured data, the higher the position estimation accuracy, the more data required. Summary of the invention
[0011] Therefore, the purpose of the disclosure is completed in view of the above-mentioned problems, and the purpose is to provide a position estimation system and a position estimation method that can better estimate the position of a mobile device based on the accuracy of a signal emitted from the mobile device even with a smaller amount of data.
[0012] In order to achieve the above-mentioned objectives, the present disclosure adopts the following technical means.
[0013] The position estimation system disclosed herein is a position estimation system for estimating the position of a mobile device relative to a vehicle by wirelessly communicating with a mobile device carried by a user, and includes: a plurality of communication devices configured to be able to wirelessly communicate with the mobile device and arranged at different respective positions of the vehicle; a communication control unit that controls the operation of the plurality of communication devices; and a position estimation unit that estimates the position of the mobile device, wherein the communication control unit executes: obtaining the reception strength of a signal from the mobile device in each of the plurality of communication devices; and obtaining a parameter different from the reception strength, namely a flight time correlation value, which directly or indirectly shows the flight time of an electric wave from each of the plurality of communication devices to the mobile device. The position estimation unit uses a correction function that outputs a corrected flight time correlation value by inputting the obtained reception strength and flight time correlation value, corrects the flight time correlation value, and estimates the position of the mobile device using the corrected flight time correlation value. The correction function is a function determined using teacher data including the positions of the plurality of mobile devices and the reception strengths and flight time correlation values at the positions of the plurality of mobile devices, and the amount of data is less than the teacher data.
[0014] Furthermore, the position estimation method disclosed herein is a position estimation method executed by at least one processor for estimating the position of a mobile device carried by a user, and includes: obtaining the reception strength of a signal from a mobile device from a plurality of communication devices, the plurality of communication devices being capable of wirelessly communicating with the mobile device and being arranged at respective different positions of a vehicle; obtaining a parameter, namely a flight time correlation value, which is different from the reception strength and directly or indirectly indicates the flight time of an electric wave from each of the plurality of communication devices to the mobile device; correcting the flight time correlation value using a correction function that outputs a corrected flight time correlation value by inputting the obtained reception strength and flight time correlation value; and estimating the position of the mobile device using the corrected flight time correlation value, the correction function being a function determined using teacher data including the positions of a plurality of mobile devices and the reception strengths and flight time correlation values at the positions of the plurality of mobile devices, and the amount of data being less than the teacher data.
[0015] According to such a position estimation system and position determination method, in order to estimate the position of a mobile device, the reception strength and flight time correlation value in each of a plurality of communication devices are used. The reception strength and the flight time correlation value are mutually different parameters, and there are cases where the causes of errors, etc. are different. Therefore, there are also cases where the estimated position of the mobile device when only the reception strength is used is inconsistent with the estimated position of the mobile device when only the flight time correlation value is used. Therefore, a correction function that outputs a corrected flight time correlation value by inputting the obtained reception strength and flight time correlation value is used to correct the flight time correlation value.
[0016] The correction function is a function determined by using teacher data including the positions of multiple mobile devices, the reception strength at the positions of multiple mobile devices, and the flight time correlation value. Moreover, the correction function is generated with less data than the teacher data. Thus, compared with saving all the teacher data and using the teacher data for correction, the amount of data can be reduced. In addition, since the flight time correlation value is corrected by using a correction function with two parameters, the correction accuracy can be improved compared with the configuration of using only one side's parameters for correction. Thus, using the corrected flight time correlation value, the position of the mobile device can be estimated with better accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 This is a block diagram showing the configuration of an electronic key system.
[0018] Figure 2 It is a diagram showing an example of arrangement of anchors.
[0019] Figure 3 is a flow chart showing a method of position estimation.
[0020] Figure 4It is a diagram for explaining a position estimation method.
[0021] Figure 5 is a diagram illustrating a neural network. DETAILED DESCRIPTION
[0022] (First Embodiment)
[0023] use Figure 1 to Figure 5 The first embodiment of the present disclosure is described. The electronic key system 100 of this embodiment is applied to a vehicle 101. Figure 1 The electronic key system 100 shown includes an in-vehicle system 102 and a mobile device 103. The in-vehicle system 102 is a system installed in a vehicle 101. The mobile device 103 is a device carried by a user of the vehicle 101. There may be multiple mobile devices 103. The electronic key system 100 is equivalent to a position estimation system, and estimates the position of the mobile device 103 relative to the vehicle 101 by wirelessly communicating with the mobile device 103.
[0024] First, the configuration that is the premise of the electronic key system 100 is described. The vehicle-mounted system 102 and the mobile device 103 are configured to be able to communicate with each other in short distance. The short distance communication here refers to communication in accordance with a prescribed short distance wireless communication standard with an actual communicable distance of, for example, about 5m to 100m. In the present embodiment, as the standard for short distance communication, for example, BLE (Bluetooth Low Energy: Bluetooth is a registered trademark) and UWB-IR (Ultra Wide Band-Impulse Radio: Ultra Wide Band Impulse Radio) are used.
[0025] The in-vehicle system 102 and the mobile device 103 are each configured to be able to perform wireless communication in accordance with the BLE standard. Therefore, the details of the communication method, such as communication connection and encrypted communication, are performed in accordance with the sequence specified in the BLE standard.
[0026] The following describes a case where the vehicle-mounted system 102 is set to act as the master in the communication with the mobile device 103, and the mobile device 103 acts as the slave. The slave in BLE communication is a device that intermittently sends advertising signals and performs data transmission and reception based on a request from the master. The master is a device that controls the communication connection state and communication timing with the slave.
[0027] An advertising signal is a signal used to notify other devices of its existence. The signal sent and received via BLE, such as an advertising signal, contains information about the source of transmission. The source of transmission information is, for example, inherent identification information assigned to the mobile device 103. For example, the device address, UUID (Universally Unique Identifier), etc. can be used as device identification information.
[0028] The mobile device 103 is a device that stores key information for using the vehicle 101 and uses the key information to function as an electronic key of the vehicle 101. The key information here is data used in the authentication process described later. The key information is used to prove that the person who wants to access the vehicle 101 is a regular user, in other words, it is data used to prove the legitimacy of the person who wants to access the vehicle 101. The key information can be called an authentication key, an encryption key, or a key code. The key information may be different for each mobile device 103. In the vehicle-mounted system 102, the key information of each mobile device 103 is stored and registered in a manner that establishes a correspondence with the device identification information.
[0029] The vehicle-mounted system 102 performs automatic authentication processing based on wireless communication with the mobile device 103. In addition, a system is implemented that implements vehicle control corresponding to the position of the user relative to the vehicle 101, based on successful authentication. The vehicle control here refers to locking and unlocking the doors, turning the power on and off, and starting the engine.
[0030] For example, when the vehicle-mounted system 102 confirms that the mobile device 103 is in a locking and unlocking area pre-set for the vehicle 101, it performs door locking and unlocking control based on the user's operation of the door button. The locking and unlocking area is set outside the vehicle compartment and near the driver's door, the passenger door, and the trunk door.
[0031] Furthermore, when the in-vehicle system 102 confirms the presence of the mobile device 103 in the vehicle cabin through wireless communication with the mobile device 103 , the in-vehicle system 102 executes engine start control based on the user's operation of the start button.
[0032] For example, the in-vehicle system 102 can authenticate the mobile device 103 by a challenge-response method. The authentication process is a process of comparing a response code generated by the mobile device 103 based on key information with a verification code stored in the vehicle 101 or generated dynamically.
[0033] Next, the mobile device 103 is described. The mobile device is a portable and general-purpose information processing terminal with a communication function. An application for making the electronic key of the vehicle 101 function, that is, a digital key application, is installed. As the mobile device 103, for example, a smartphone, a tablet terminal, a wearable device, etc. can be used. A wearable device is a device that is worn on the user's body for use, and can be a device in a variety of shapes such as a wristband type, a watch type, a ring type, a glasses type, and a headphone type.
[0034] like Figure 1 As shown, the mobile device 103 includes a terminal control unit 10, a portable BLE model 11, and a portable UWB model 12. In this embodiment, for convenience, the components of the mobile device 103 other than those related to wireless communication with the in-vehicle system 102 are briefly described.
[0035] The portable BLE model 11 is a communication model capable of short-range wireless communication in accordance with the above-mentioned BLE standard. The portable BLE model 11 is composed of, for example, an IC, an antenna, and a communication circuit. The portable BLE model 11 establishes a communication connection with the vehicle-mounted system 102 to perform short-range wireless communication.
[0036] The portable BLE model 11 periodically scans and receives the advertisement signal periodically transmitted from the vehicle-mounted system 102. The portable BLE model 11 receiving the advertisement signal transmits a connection request to the vehicle-mounted system 102. When the connection request is accepted, the communication connection between the portable BLE model 11 and the vehicle-mounted system 102 is established.
[0037] The portable UWB model 12 is a communication model capable of UWB-IR short-range wireless communication. Hereinafter, the UWB-IR short-range wireless communication is referred to as UWB communication. UWB communication can also be said to be ultra-wideband wireless communication. The portable UWB model 12 is the same as the portable BLE model 11, and is composed of, for example, an IC, an antenna, and a communication circuit.
[0038] The portable UWB model 12 performs UWB communication by sending and receiving pulse-shaped radio waves (hereinafter referred to as pulse signals). The pulse signal used in UWB communication refers to a signal with an extremely short pulse width. For example, a signal with a pulse width of 2ns. In addition, the pulse signal used in UWB communication refers to a signal with a bandwidth of more than 500MHz (that is, an ultra-wideband bandwidth). In addition, the frequency bands that can be used for UWB communication (hereinafter referred to as UWB bands) include 3.1GHz to 10.6GHz, 3.4GHz to 4.8GHz, 7.25GHz to 10.6GHz, and 22GHz to 29GHz. When the portable UWB model 12 receives a pulse signal sent from the vehicle-mounted system 102, it returns a response signal corresponding to the signal.
[0039] The terminal control unit 10 is implemented by a central processing unit (CPU) and controls the portable BLE model 11 and the portable UWB model 12. The terminal control unit 10 performs processing for realizing the functions of each part of the mobile device 103. The terminal control unit 10 can communicate with the vehicle-mounted system 102 by executing a program stored in the memory.
[0040] Next, the vehicle-mounted system 102 is described. Figure 1 As shown, the vehicle-mounted system 102 includes a smart ECU 20 and a plurality of anchors 30. ECU is an abbreviation of Electronic Control Unit, which refers to an electronic control device. The anchor 30 is equivalent to a communication device.
[0041] The smart ECU 20 is connected to each of the anchors 30 via the in-vehicle network 104. In addition, the smart ECU 20 is connected to other vehicle-mounted control devices, such as the power supply ECU and the body ECU, via the in-vehicle network 104 so as to be able to communicate with each other. The in-vehicle network 104 is a communication network constructed in the vehicle. As a standard of the in-vehicle network 104, for example, the Controller Area Network (CAN: registered trademark) is used.
[0042] The smart ECU 20 estimates the position of the mobile device 103 by cooperating with the anchor 30 and the like. In addition, the smart ECU 20 realizes vehicle control corresponding to the estimated result of the position of the mobile device 103 by cooperating with other ECUs. The smart ECU 20 is implemented using a computer. That is, the smart ECU 20 has a processor 21, RAM, memory, I / O, and a bus connecting these components. Figure 1 Among them, only the processor 21 is shown in the figure.
[0043] The processor 21 is hardware for calculation processing combined with RAM (Random Access Memory). The processor 21 is, for example, a CPU. The processor 21 executes various processes for realizing the functions of each functional unit by accessing the RAM. RAM is a volatile storage medium. The memory is configured to include a non-volatile storage medium such as a flash memory. Various programs executed by the processor 21 are stored in the memory. The execution of a program by the processor 21 is equivalent to executing a method corresponding to the program, such as a position estimation method. I / O is a circuit model for communicating with other devices. I / O is implemented using analog circuit elements, ICs, etc.
[0044] The memory registers the device identification information of each mobile device 103. The memory also stores setting data indicating the mounting position of each anchor 30 in the vehicle 101. The memory corresponds to a storage device storing the setting data.
[0045] The smart ECU 20 includes a vehicle-mounted BLE model 22 and a vehicle-mounted UWB model 23. The vehicle-mounted BLE model 22 is a communication model capable of short-range wireless communication in accordance with the above-mentioned BLE standard. The vehicle-mounted BLE model 22 has the same structure as the above-mentioned portable BLE model 11. The vehicle-mounted BLE model 22 establishes a communication connection with the mobile device 103 for short-range wireless communication.
[0046] The vehicle-mounted UWB model 23 is a communication model capable of performing the above-mentioned UWB communication. The vehicle-mounted UWB model 23 has the same structure as the above-mentioned portable UWB model 12. The vehicle-mounted UWB model 23 performs UWB communication by sending and receiving pulse signals. When the vehicle-mounted UWB model 23 receives the pulse signal sent from the mobile device 103, it returns a response signal corresponding to the signal.
[0047] The plurality of anchors 30 are configured to be able to wirelessly communicate with the mobile device 103, and are arranged at different positions of the vehicle 101. The anchor 30 is a communication model including an anchor UWB model 31 and an anchor control unit 32. The anchor 30 performs UWB communication according to the instruction of the smart ECU 20. The anchor 30 is equivalent to an antenna. The anchor 30 is arranged at a plurality of positions inside and outside the vehicle 101. The vehicle 101 is provided with two or more anchors 30, including an anchor 30 arranged inside the vehicle and an anchor 30 arranged outside the vehicle. A unique identifier is set for each anchor 30.
[0048] Here, use Figure 2 An example of the arrangement of the anchor 30 according to this embodiment will be described. Figure 2As shown, there are four anchors 30 disposed outside the vehicle compartment and three disposed inside the vehicle compartment, for a total of seven. Figure 1 In order to facilitate understanding, only three anchors 30 are shown. The structures of the plurality of anchors 30 are identical to each other.
[0049] The four anchors 30 outside the vehicle cabin are specifically arranged near the left and right corners of the front end and the left and right corners of the rear end of the vehicle 101. The three anchors 30 inside the vehicle cabin are specifically arranged one inside the dashboard at the front center of the vehicle cabin and one on each side of the rear seat.
[0050] The anchor control unit 32 acquires a time-of-flight correlation value by sending and receiving a pulse signal. The time-of-flight correlation value is also called a ToF correlation value. The ToF correlation value is a parameter that directly or indirectly indicates the flight time of the radio wave from the anchor 30 to the mobile device 103. The distance from the anchor 30 to the mobile device 103 corresponds to the flight time (ToF) of the signal.
[0051] ToF is determined based on the phase difference between two frequencies and the round-trip time (RTT). The phase difference between two frequencies and RTT are equivalent to ToF correlation values. ToF correlation values can also be called distance correlation values. These ToF correlation values are parameters different from the reception intensity. The two-frequency phase difference here is the difference between the transmit and receive phase differences observed at two different frequencies. The two-frequency phase difference corresponds to the displacement of the phase angle caused by the change in frequency. The transmit-receive phase difference is equivalent to the phase difference between the transmitted CW signal and the received CW signal. The transmit-receive phase difference can also be simply called the phase angle.
[0052] In this embodiment, the round trip time is used as the ToF related value. Specifically, the anchor control unit 32 measures the elapsed time from the transmission of the pulse signal to the reception of the pulse signal as a response signal to the pulse signal, that is, the round trip time. The anchor control unit 32 outputs the measured round trip time together with its own identification information to the smart ECU 20. In addition, the method for obtaining the round trip time described here is only an example. As long as the round trip time used for ranging between the anchor 30 and the mobile device 103 can be obtained, other methods can also be used.
[0053] The anchor control unit 32 measures the strength of the signal received by the anchor UWB model 31. The anchor control unit 32 is configured to sequentially detect the strength of the signal received by the anchor UWB model 31. The signal indicating the reception strength or the measured value itself can also be called RSSI (Received Signal Strength Indicator / Indication). The anchor control unit 32 outputs the measured reception strength together with its own identification information to the smart ECU 20.
[0054] Next, the control of the smart ECU 20 will be described. The smart ECU 20 transmits an advertising signal from the vehicle-mounted BLE model 22. For example, when the vehicle 101 is parked and a certain time has passed since all doors of the vehicle 101 were locked, the advertising signal is transmitted periodically.
[0055] The smart ECU 20 acquires information related to wireless communication with the mobile device 103 from the in-vehicle BLE model 22. When the portable BLE model 11 of the mobile device 103 that has received the advertisement signal establishes a connection with the in-vehicle BLE model 22, the smart ECU 20 acquires the information received by the in-vehicle BLE model 22.
[0056] The smart ECU 20 also functions as a communication control unit, and controls the operations of the plurality of anchors 30. The smart ECU 20 transmits a pulse signal from the vehicle-mounted UWB model 23. Similarly, the smart ECU 20 transmits a pulse signal from the anchor UWB model 31 of each anchor 30.
[0057] The smart ECU 20 sequentially transmits pulse signals from the plurality of anchors 30. For example, the smart ECU 20 sequentially transmits pulse signals from the anchors 30 at predetermined time intervals.
[0058] The smart ECU 20 starts transmitting the pulse signal when the connection with the portable BLE model 11 of the mobile device 103 is established or the connection is established. This can prevent wasteful transmission of the pulse signal even when there is no mobile device 103 around the vehicle 101.
[0059] The smart ECU 20 obtains information related to UWB communication received by the anchor 30 when the UWB model of the mobile device 103 performs UWB communication with the anchor 30. The smart ECU 20 obtains the round trip time output from each of the plurality of anchors 30 as a ToF related value. Therefore, the smart ECU 20 obtains the ToF related value from each of the plurality of anchors 30. In addition, the smart ECU 20 obtains the reception strength from each of the plurality of anchors 30. Therefore, the smart ECU 20 obtains the reception strength of the signal from the mobile device 103 in each of the plurality of anchors 30.
[0060] Next, the method for estimating the position of the mobile device 103 is described. Figure 3 The smart ECU 20 also functions as a position estimation unit to estimate the position where the mobile device 103 exists.
[0061] In step S1, the ToF correlation value and the reception intensity are obtained from each anchor 30, and the process proceeds to step S2. In step S2, the ToF correlation value is corrected using a correction function, and the process proceeds to step S3. The correction function outputs a corrected ToF correlation value by inputting the reception intensity and the ToF correlation value obtained in step S1.
[0062] In step S3, it is determined whether the mobile device 103 is inside or outside the vehicle, and the process proceeds to step S4. The determination of whether the mobile device 103 is inside the vehicle can be determined using the corrected ToF correlation value, the reception strength alone, or the ToF correlation value before correction and the reception strength.
[0063] In step S4, the vehicle compartment or the vehicle compartment outside determined in step S3 is subjected to detailed area determination, and the process proceeds to step S5. Detailed area determination refers to the process of determining which area is in the segmented area after the vehicle compartment or the vehicle compartment outside is segmented more finely. The area mentioned here can be an area with an area, or can be a process of specifying any one of a plurality of discrete points.
[0064] In step S5, the position of the mobile device 103 is estimated using the corrected ToF correlation value, and the process ends. In step S5, a specific point in the area determined in step S4 is used as a temporary position, and an operation is performed to obtain a position with better accuracy. The specific point in the area can be an arbitrary point, the center of gravity of the area, or any vertex of the area. In step S5, positioning is performed, for example, by nonlinear least squares method, but since the nonlinear least squares method is a method of repeatedly calculating and searching for the minimum point of error from the initial position, the initial position becomes an important factor. If the initial position is wrong, it converges to a local solution, resulting in a wrong judgment.
[0065] Therefore, in step S4, a detailed area determination is performed. Figure 4 The position estimation in step S4 and step S5 is described below. Figure 4 In order to facilitate understanding, the description is made using the distance measurement circles C1 to C3 of the three right-hand anchors 30 among the seven anchors 30. In addition, the actual position of the mobile device 103 is shown by a triangle.
[0066] Figure 4 The distance measurement circles C1 to C3 are circles with the distances determined using the ToF correlation values corrected in step S2 as radii h1 to h3 and with each anchor 30 as the center. Ideally, the distance measurement circles C1 to C3 should intersect at one point at the location of the mobile device 103. However, if Figure 4 As shown in FIG. 1 , the three distance measurement circles C1 to C3 may not intersect at one point. Various measurement errors may be the cause of this.
[0067] In step S5, in order to estimate the position of the mobile device 103 by the nonlinear least square method, an initial position needs to be set. Therefore, in step S4, although there is no position where the three ranging circles C1 to C3 intersect at one point, a temporary position where the mobile device 103 is likely to be located is set to set the initial position. The temporary position is a point within the detailed area set in step S4.
[0068] exist Figure 4 In the example shown, there are three intersection points of the distance measurement circles C1 to C3 at six locations. Figure 4 The intersection of the triangle positions is set as the initial position. Therefore, as described later, the initial position is set by using a statistical model learned using teacher data. Figure 4 In the example shown, as shown by the prism, the point where the two distance measuring circles C1 and C3 intersect is used as the initial position. And in step S5, starting from this initial position, a mathematical search is performed for the position where the total distance from each distance measuring circle C1 to C3 is the smallest. Figure 4 In the initial position shown, the distances to the two distance measuring circles C1 and C3 are 0, and the distance to the distance measuring circle C2 is indicated by an arrow L2.
[0069] Next, the correction function used in step S2 is described. The correction function is a function determined using teacher data including the positions of multiple mobile devices 103 and the reception strength and ToF related values at the positions of multiple mobile devices 103, and the amount of data is less than the teacher data. The teacher data is, for example, the measured data of the ToF value and the reception strength at the measurement positions at 16530 points inside and outside the vehicle.
[0070] In this embodiment, machine learning that can process a large amount of data is used to perform high-precision correction. Data of the ToF correlation value and reception strength of each anchor 30 and the mobile device 103 are acquired in advance around the vehicle 101 and learned. If learning is performed in a manner that covers the surrounding positions of the vehicle 101, a large amount of data will be processed, but the logic is required to be executed by a small-capacity smart ECU 20 such as the ECU installed in the vehicle 101, so the method uses a neural network that can learn efficiently and can reduce the ROM capacity of the smart ECU 20.
[0071] Compared with pattern matching, neural networks can reduce the amount of data used. For example, assume that the measurement positions at 16,530 points inside and outside the car are used to obtain each ToF correlation value and reception strength as teacher data. In this case, the ToF correlation value and reception strength based on seven anchors 30 are obtained at one measurement position. Therefore, there are two parameters and seven antennas, and the measurement positions are 16,530 points, so 231,420 data are required. Therefore, the amount of teacher data is about 1,850KB.
[0072] When pattern matching is performed to correct the measurement values obtained by each anchor 30, a large amount of data is preferably used, so the teacher data is directly used and correction is performed by interpolation and extrapolation. In this way, the smart ECU 20 needs to store the teacher data.
[0073] In contrast, Figure 5 As shown in FIG. 1 , consider the case of a neural network using a multilayer perceptron. When the neural network has, for example, an input layer, two hidden layers, and an output layer, the input layer and the output layer correspond to seven anchors 30, and seven nodes are required. Moreover, fifty nodes are set for each of the two hidden layers. In this case, the threshold, gain, and weight of each layer become the numbers shown in Table 1. The number of gains and weights can also be said to be the number of paths from each layer to the next layer.
[0074] [Table 1]
[0075] Input Layer Hidden Layer Hidden Layer Output Layer Threshold 7 50 50 7 Gain and Weighting 7 350 2500 350
[0076] Therefore, in the neural network, a total of 3321 data of threshold, gain and weight is required. This data amount is about 26.5KB. Therefore, it is about 1 / 70 of the teacher data. The neural network performs a specified operation on the input value to output the value. Therefore, the neural network can be said to be a correction function.
[0077] Next, Figure 3The other determination methods in step S3 are described. In the case of determining whether it is inside the vehicle in step S3, a statistical model can also be used for determination. The statistical model outputs whether the mobile device 103 is inside the vehicle or outside the vehicle by inputting the reception strength and ToF related value obtained in step S1. Such a statistical model uses the same teacher data as above to enable it to be pre-learned to determine whether it is inside the vehicle. Learning can be done by, for example, machine learning. As machine learning, for example, deep learning can be used.
[0078] Next, Figure 3 The other determination methods of step S4 are described below. The detailed area determination of step S4 may also be performed using other statistical models instead of the corrected ToF correlation value. Moreover, different statistical models may be used for the situation inside the vehicle and the situation outside the vehicle.
[0079] The statistical model for the vehicle cabin uses the reception strength and ToF related values obtained in step S1 to output a statistical model of which of the multiple pre-set areas in the vehicle cabin the mobile device 103 exists, thereby determining the area where the mobile device 103 exists. In addition, the statistical model for the vehicle cabin uses the reception strength and ToF related values obtained by input to output a statistical model of which of the multiple pre-set areas outside the vehicle cabin the mobile device 103 exists, thereby determining the area where the mobile device 103 exists.
[0080] Such a statistical model uses the teacher data inside the car among the above-mentioned teacher data to learn the statistical model used inside the car, and uses the teacher data outside the car to learn the statistical model outside the car. The trends of the reception intensity and ToF correlation values are different inside and outside the car, so the area can be determined with better accuracy.
[0081] As described above, according to the position estimation system and the position determination method of the present embodiment, the reception intensity and the ToF correlation value in each of the plurality of anchors 30 are used to estimate the position of the mobile device 103. The reception intensity and the ToF correlation value are mutually different parameters, and there are cases where the causes of errors etc. are different. Therefore, there are cases where the estimated position of the mobile device 103 when only the reception intensity is used is inconsistent with the estimated position of the mobile device 103 when only the ToF correlation value is used. Therefore, the ToF correlation value is corrected using a correction function that outputs a corrected ToF correlation value by inputting the acquired reception intensity and ToF correlation value.
[0082] The correction function is a function determined using teacher data including the positions of multiple mobile devices 103, the reception strength at the positions of multiple mobile devices 103, and ToF related values. Moreover, the correction function is generated with less data than the teacher data. Thus, compared with saving all the teacher data and using the teacher data for correction, the amount of data can be reduced. In addition, since the correction is performed by using a correction function with two parameters, the correction accuracy can be improved compared with the configuration in which only one side's parameters are used for correction. Thus, the corrected time-of-flight related value can be used to estimate the position of the mobile device 103 with better accuracy.
[0083] In addition, in this embodiment, Figure 3 As described in step S5 of , the temporary position of the mobile device 103 set using the corrected ToF correlation value is set as the initial value of the nonlinear least square method, and the position with the minimum distance from the corrected time-of-flight correlation value is estimated as the position of the mobile device 103 by searching from the initial value. By using the corrected ToF correlation value, the initial value can be set with good accuracy, so the accuracy of position estimation can be improved.
[0084] In this way, in the present embodiment, not only the ToF correlation value of each anchor 30 but also the information of the radio wave intensity is used as a parameter for estimating the position of the mobile device 103. In addition, the ToF correlation value is corrected using the actual measurement data measured in advance at the plurality of anchors 30. Machine learning is used for correction, but in order to achieve a capacity that can be installed in the microcomputer mounted on the vehicle 101, a neural network that can learn efficiently is used. By applying the neural network, it is possible to reduce the capacity that can be installed in the smart ECU 20.
[0085] In addition, in order to prevent the key from being locked in the car or to ensure the engine start area, it is first determined whether it is inside or outside the car, and then the detailed area determination and position estimation are performed. By using a neural network, it is also possible to make high-precision determinations inside and outside the car. In addition, by combining a neural network to select the best initial position, the positioning accuracy of the nonlinear least squares method can be improved.
[0086] (Other Embodiments)
[0087] As mentioned above, although the preferred embodiment of the present disclosure has been described, the present disclosure is not limited to the above-mentioned embodiment, and can be implemented with various modifications within the scope not departing from the gist of the present disclosure.
[0088] The configurations of the above-described embodiments are merely illustrative, and the scope of the present disclosure is not limited to the scope described therein. The scope of the present disclosure is indicated by the description of the claims, and further includes all modifications within the meaning and scope equivalent to the description of the claims.
[0089] In the first embodiment described above, Figure 3 After determining the inside and outside of the vehicle in step S3, a detailed area determination is performed in step S4, but the composition of the detailed area is not limited to being determined in two stages. For example, a statistical model that outputs a plurality of pre-set areas inside and outside the vehicle 101 by inputting the received strength and ToF related values obtained by input can be used to determine the area where the mobile device 103 exists. The statistical model is the same as other statistical models and can be obtained by learning from teacher data. Figure 3 The processing of step S3 and step S4 is combined into one.
[0090] In the first embodiment described above, Figure 3 In step S4, a point in the area determined by the detailed area determination is set as the initial value, but it is not limited to such a configuration. For example, a ranging circle can be generated based on the corrected ToF correlation value, and a point equidistant from the intersection of multiple ranging circles can be set as a temporary position, or the center of the area where the intersection of multiple ranging circles gathers can be set as a temporary position.
[0091] In the first embodiment described above, the round trip time is used as the ToF related value, but it is not limited to such a configuration. For example, the two-frequency phase difference of each frequency combination can also be used as the ToF related value. In BLE communication, there are more than two frequencies provided for communication, so it is possible to obtain the two-frequency phase difference of more than two different frequency combinations. The device distance is estimated based on the two-frequency phase difference.
[0092] In the configuration using multi-frequency phase differences as ToF correlation values, the communication for distance measurement can be understood as communication for determining the transmission / reception phase difference for every two or more frequencies. Transmission / reception of CW signals at multiple frequencies can be equivalent to the communication for distance measurement.
[0093] In the first embodiment described above, the position of the mobile device 103 is estimated using the reception strength and the ToF correlation value, but other position estimation methods may also be combined. For example, the smart ECU 20 may also obtain the direction of arrival of the signal as information indicating the reception status of the signal from the mobile device 103. For example, the estimation of the direction of arrival of the signal may be implemented by various methods such as the MUSIC method and the ESPRIT method. The reception strength, phase, direction of arrival, etc. may be referred to as the characteristics of the received signal.
[0094] In the first embodiment described above, the functions implemented by the smart ECU 20 may be implemented by hardware and software different from the above, or by a combination thereof. The smart ECU 20 may communicate with other control devices, for example, and the other control devices may execute part or all of the processing. When the smart ECU 20 is implemented by an electronic circuit, the smart ECU may be implemented by a digital circuit including many logic circuits, or an analog circuit.
[0095] The smart ECU 20 includes a method of using one or more ICs to implement a certain function as hardware. As a processor (computing core), a CPU, an MPU, a GPU, a DFP (Data Flow Processor), etc. can be used. In addition, multiple types of computing processing devices can be combined to implement part or all of the functions of the processor. It is also possible to use a system-on-chip (SoC), FPGA, ASIC, etc. to implement part or all of the functions of the processor. FPGA is the abbreviation of Field-Programmable Gate Array. ASIC is the abbreviation of Application Specific Integrated Circuit.
[0096] Various computer programs executed by the control device can also be stored as instructions executed by the computer in a non-transitory tangible storage medium that can be read by the computer. As a storage medium for the program, an HDD (Hard-disk Drive), an SSD (Solid State Drive), a flash memory, etc. can be used.
[0097] In the above-mentioned first embodiment, the position estimation device is used in the vehicle 101 , but it is not limited to the state of being mounted on the vehicle 101 , and at least a part of it may not be mounted on the vehicle 101 .
Claims
1. A position estimation system for estimating the position of a mobile device (103) carried by a user relative to a vehicle (101) by wirelessly communicating with the mobile device (103), wherein: Include: A plurality of communication devices (30) configured to be capable of wirelessly communicating with the mobile device and arranged at different locations of the vehicle; A communication control unit (20) for controlling the operation of the plurality of communication devices; and A position estimation unit (20) estimates the position of the mobile device. The communication control unit performs: obtaining the reception strength of the signal from the mobile device in each of the plurality of communication devices; and acquiring a flight time correlation value which is a parameter different from the reception intensity and directly or indirectly indicates the flight time of the radio waves from each of the plurality of communication devices to the mobile device, The position estimation unit outputs a correction function of the corrected flight time correlation value using the received intensity and the flight time correlation value obtained by input, corrects the flight time correlation value, and estimates the position of the mobile device using the corrected flight time correlation value. The correction function is a function determined using teacher data including the positions of the plurality of mobile devices and the reception strengths and flight time correlation values at the positions of the plurality of mobile devices, and the amount of data is less than that of the teacher data.
2. The position estimation system according to claim 1, wherein: The position estimation unit sets the temporary position of the mobile device set using the corrected time-of-flight correlation value as the initial value of the nonlinear least squares method, starts searching from the initial value, and estimates the position with the shortest distance determined by the corrected time-of-flight correlation value as the position of the mobile device.
3. The position estimation system according to claim 1, wherein: The position estimation unit uses the reception strength and the flight time correlation value obtained through input to output a statistical model of which area among multiple areas pre-set inside and outside the vehicle the mobile device is located in, determines the area where the mobile device is located, sets a point in the determined area as an initial value of the nonlinear least squares method, starts searching from the initial value, and estimates the position with the shortest distance determined by the corrected flight time correlation value as the position of the mobile device.
4. The position estimation system according to claim 1, wherein: The above position estimation unit performs: Determine whether the mobile device is in a vehicle compartment by using at least one of the received strength and the flight time correlation value; In the case of being in the vehicle compartment, using the received strength and the flight time correlation value obtained by input, a statistical model of which area of the plurality of pre-set areas in the vehicle compartment the mobile device is present is output, and the area where the mobile device is present is determined; When the mobile device is outside the vehicle, the receiving strength and the flight time correlation value obtained by input are used to output a statistical model of which area of the plurality of pre-set areas outside the vehicle is the mobile device, so as to determine the area where the mobile device is present; A point in the determined region is set as an initial value of the nonlinear least squares method; as well as The search is started from the initial value, and the position with the shortest distance determined by the corrected time-of-flight correlation value is estimated as the position of the mobile device.
5. A location estimation method is a location estimation method executed by at least one processor for estimating the location of a mobile device carried by a user, wherein: Include: Acquiring reception strength of a signal from the mobile device from a plurality of communication devices, the plurality of communication devices being capable of wirelessly communicating with the mobile device and being disposed at respective different positions of the vehicle; Acquiring a flight time correlation value which is a parameter different from the reception intensity and directly or indirectly indicates the flight time of the radio wave from each of the plurality of communication devices to the mobile device; Correct the time-of-flight correlation value by using a correction function that outputs a corrected time-of-flight correlation value by inputting the received intensity and the time-of-flight correlation value; as well as Using the corrected time-of-flight correlation value, estimating the position of the mobile device, The correction function is a function determined using teacher data including the positions of the plurality of mobile devices and the reception strengths and flight time correlation values at the positions of the plurality of mobile devices, and the amount of data is less than that of the teacher data.
