POSITIONING DEVICE FOR MOBILE OBJECTS

Abstract

The present invention aims to provide a positioning device for mobile objects that positions a mobile object with high accuracy using a sensor.The positioning device (7) for mobile objects according to the present invention comprises: a sensor information acquisition unit that acquires a sensor value of a mobile object, wherein the sensor value is detected by a sensor; a swim angle estimation unit (12) that estimates the swim angle of the mobile object using the sensor value; and an inertial positioning unit (14) that performs an inertial positioning of the mobile object using the sensor value and the swim angle, wherein the swim angle estimation unit (12) estimates the swim angle based on the sensor value and a mixed model (f) obtained by obtaining a plurality of motion models (f1, f2) of the mobile object based on state variables of the mobile object and integrating the plurality of motion models.

Description

Technical field

[0001] The present invention relates to the positioning of a mobile object. State of the art

[0002] Automated driving of a mobile object requires controlling the mobile object so that the mobile object detects an area in which the mobile object should drive, generates a route that is a route along which the mobile object should drive, and drives along the generated route.

[0003] Conventional systems have been developed that detect the road using satellite positioning results from a global navigation satellite system (GNSS) and map data acquired with high accuracy. More recently, systems have been implemented that determine a vehicle's position with high accuracy using quasi-zenith satellites and network real-time kinematic positioning (RTK).

[0004] Patent document 1 describes a self-position estimating device comprising: a detection unit that detects the speed of a vehicle as measured by a speed sensor, the angular velocity of the vehicle as measured by an angular velocity sensor, and the position and attitude angle of the vehicle as measured by a positioning device; a second self-position estimating unit that estimates a second position, which is the position of the vehicle, based on a predetermined body-side drift angle, as well as the speed and angular velocity of the vehicle as measured by the detection unit;a calculation unit for the attitude angle correction value, which calculates an attitude angle correction value based on a deviation between the second position, estimated by the second estimation unit for its own position, and the position of the vehicle in question, as recorded by the detection unit, when the detection unit is recording the position of the vehicle in question;and an initial estimation unit for its own position, which estimates an initial position that is the position of the vehicle in question, based on the speed of the vehicle in question obtained from the detection unit, the angular velocity of the vehicle in question, the position and attitude angle of the vehicle in question previously detected by the detection unit, the attitude angle correction value calculated by the attitude angle correction value calculation unit, and the body roll angle if the detection unit does not detect the position of the vehicle in question. State of the art document (patent document)

[0005] Patent document 1: Japanese patent application disclosure JP 2020-112490A Summary Problem to be solved with the invention

[0006] In the configuration according to patent document 1, the position and attitude angle of a mobile object are estimated, taking into account the object's sideslip angle. This sideslip angle is estimated by referencing a predefined table based on the steering angle and vehicle speed, or by assuming that the mobile object is cornering in a stable state with a constant steering angle. Consequently, the accuracy in estimating the attitude angle may decrease if the steering angle changes instantaneously or suddenly.

[0007] The technology of the present invention was developed to solve the problem, and its purpose is to provide a positioning device for mobile objects that positions a mobile object with high accuracy using a sensor. Ways to solve the problem

[0008] One of the positioning devices for mobile objects of the present invention comprises: a sensor information acquisition unit for acquiring a sensor value of a mobile object, wherein the sensor value is detected by a sensor; a swim angle estimation unit for estimating the swim angle of the mobile object using the sensor value; and an inertial positioning unit for performing inertial positioning of the mobile object using the sensor value and the swim angle, wherein the swim angle estimation unit estimates the swim angle based on a mixed model and the sensor value, the mixed model being obtained by weighting a plurality of motion models of the mobile object based on state variables of the mobile object, and integrating the plurality of motion models. Effects of the invention

[0009] Since the positioning device for mobile objects according to the present invention estimates the swimming angle with high accuracy using the mixed model obtained by integrating a plurality of motion models, the positioning device for mobile objects can position a mobile object with high accuracy. The aims, features, aspects, and advantages of the present invention will become even clearer from the following detailed description and the accompanying drawings. Brief description of the drawings Fig. Figure 1 illustrates the overall structure of a vehicle to which a positioning device for mobile objects according to embodiment 1 is attached. Fig. Figure 2 is a block diagram illustrating a configuration of the positioning device for mobile objects according to embodiment 1. Fig. Figure 3 is a flowchart illustrating processes to be performed by the positioning device for mobile objects according to embodiment 1. Fig. Figure 4 is a schematic diagram illustrating a first model of movement. Fig. Figure 5 is a schematic diagram illustrating a second movement model. Fig. Figure 6 illustrates a weighting function. Fig. Figure 7 is a comparison diagram for the first motion model, the second motion model, and a mixed model. Fig. Figure 8 is a block diagram illustrating a configuration of a positioning device for mobile objects according to a modification of embodiment 1. Fig. Figure 9 is a block diagram illustrating a configuration of a positioning device for mobile objects according to a modification of embodiment 1. Fig. Figure 10 is a block diagram illustrating a configuration of a positioning device for mobile objects according to a modification of embodiment 1. Fig. Figure 11 illustrates a hardware configuration of the positioning device for mobile objects. Fig. Figure 12 illustrates a hardware configuration of the positioning device for mobile objects. Description of the embodiments A. Embodiment 1A-1. Structure

[0010] Fig. Figure 1 illustrates the overall structure of a vehicle 1 to which a positioning device 7 for mobile objects according to embodiment 1 is attached. The vehicle 1 is an example of a mobile object. As shown in Fig. As illustrated in Figure 1, vehicle 1 comprises the following: a steering wheel 2, a steering actuator 3, an antenna 5, a drive 6, a positioning device 7 for mobile objects, a sensor 8 for mobile objects, and a vehicle control device 9. Vehicle 1 also comprises a brake for braking vehicle 1, which is located in Fig. 1 is not shown.

[0011] The steering actuator 3 is mounted on the steering wheel 2, which operates the two wheels of the front wheels. The steering actuator 3 includes, for example, an electric power steering motor (EPS) and an electronic control unit (ECU). The steering actuator 3 operates according to a steering command from the vehicle control unit 9, so that the rotations of the steering wheel 2 and the front wheels can be controlled. The steering actuator 3 controls the steering according to a steering command value received from the vehicle control unit 9, so that the vehicle 1 travels along a road.

[0012] The drive unit 6 is mounted on the front axle of the vehicle 1 and propels the vehicle 1. The drive unit 6 comprises, for example, a drive motor, an ECU, and the brake. The drive unit 6 operates according to a command value for speed or acceleration received from the vehicle control unit 9, enabling the vehicle 1 to be braked and propelled. The drive unit 6 controls the braking and propulsion of the vehicle 1 according to the command value for speed or acceleration received from the vehicle control unit 9, thus allowing the speed of the vehicle 1 to adapt to a traffic situation.

[0013] The antenna 5 receives a satellite signal from a satellite 4 and transmits the received satellite signal to the positioning device 7 for mobile objects. The satellite 4 includes, for example, multiple satellites for the Global Positioning System (GPS). The satellite 4 is not limited to GPS satellites. Another positioning satellite, such as the Global Navigation Satellite System (GLONASS), can also be used as satellite 4.

[0014] The sensor 8 for mobile objects detects a state variable of the vehicle 1, such as the steering angle or the vehicle speed, and transmits the detected state variable to the positioning device 7 for mobile objects.

[0015] The positioning device 7 for mobile objects measures the position of the vehicle 1 based on the satellite signal received by the antenna 5 and the state variable of the vehicle 1 detected by the sensor 8 for mobile objects, and transmits the measured position of the vehicle 1 to the vehicle control device 9.

[0016] The vehicle control unit 9 outputs command values ​​to the steering actuator 3 and the drive 6, based on the position of the vehicle 1 measured by the positioning unit 7 for mobile objects, the state variable of the vehicle 1 detected by the sensor 8 for mobile objects, and a detection result, for example from a camera or millimeter wave radar, which are not shown.

[0017] Next, a configuration of the positioning device 7 for mobile objects according to embodiment 1 is described with reference to Fig. 2 described. Fig. Figure 2 is a block diagram showing the configuration of the positioning device 7 for mobile objects according to embodiment 1.

[0018] As in Fig. As illustrated in Figure 2, the positioning device 7 for mobile objects comprises the following: a receiving unit 10 for satellite positioning results, an inertial sensor 11, a swim angle estimation unit 12, a sensor correction unit 13, an inertial positioning unit 14, and a filter unit 15. Furthermore, the positioning device 7 for mobile objects comprises a sensor information acquisition unit, which is described in Figure 2. Fig. 2 is not shown. Furthermore, the antenna 5 and the sensor 8 for mobile objects are connected to the positioning device 7 for mobile objects.

[0019] The satellite positioning results receiver 10 converts the satellite signal, which is an output from the antenna 5, into data such as latitude, longitude, or azimuth, and outputs the data to the filter unit 15 as a satellite positioning result. The satellite positioning result is represented by a format available in the positioning device 7 for mobile objects, for example, a GPGGA format.

[0020] The inertial sensor 11 is an angular velocity sensor mounted on the positioning device 7 for mobile objects and outputs an angular velocity (a yaw rate) generated by the rotation or steering of the vehicle 1. In embodiment 1, the inertial sensor 11 is assumed to be mounted on the positioning device 7 for mobile objects. The inertial sensor 11 can also be mounted on the vehicle 1 outside the positioning device 7 for mobile objects, and it can input a detection result into the positioning device 7 for mobile objects. This can reduce the cost of the positioning device 7 for mobile objects.

[0021] The swim angle estimation unit 12 estimates the swim angle by weighting a plurality of motion models based on the vehicle speed and steering angle received from the moving object sensor 8 and the yaw rate received from the inertial sensor 11, and outputs an estimated swim angle result to the filter unit 15. The sensor correction unit 13 acquires sensor values ​​from the moving object sensor 8 and the inertial sensor 11, corrects any sensor error contained in each of the sensor values, such as a scaling factor or bias / pre-displacement, and outputs the sensor errors to the inertial positioning unit 14 and the filter unit 15. In other words, the sensor correction unit 13 acts as a sensor information acquisition unit that acquires sensor values ​​from the moving object sensor 8 and the inertial sensor 11.

[0022] The inertial positioning unit 14 performs an inertial positioning calculation, for example, for the position, attitude, and speed, which are the positioning results of the vehicle 1, using the sensor values ​​that have been corrected by the sensor correction unit 13, and outputs an inertial positioning result to the filter unit 15.

[0023] The filter unit 15 estimates - using the result of the inertial positioning received from the inertial positioning unit 14 - an error between the result of the inertial positioning and the sensor value output by the sensor 8 for mobile objects. A-2. Operation

[0024] Next, a procedure of processes to be carried out by the positioning device 7 for mobile objects according to embodiment 1 is described with reference to the flowchart in Fig. 3 described.

[0025] As soon as the positioning device 7 for mobile objects begins operation, the sensor correction unit 13 acquires inertial sensor values ​​from the inertial sensor 11, and it acquires sensor values ​​for mobile objects from the sensor 8 for mobile objects, specifically in step S101. Here, the inertial sensor values ​​include the current angular velocity and the current acceleration of the vehicle 1. The angular velocities include yaw rate, pitch rate, and roll rate, and the accelerations include longitudinal acceleration, lateral acceleration, and vertical acceleration. Additionally, the sensor values ​​for mobile objects include the current steering angle and the current vehicle speed of the vehicle 1.

[0026] Next, in step S102, the sensor correction unit 13 acquires a sensor correction value, calculated in a previous cycle, from the filter unit 15. If the previous sensor correction value is not available for any reason, for example immediately after commissioning, the sensor correction unit 13 uses a predefined initial value (0 or a value at delivery).

[0027] Next, in step S103, the sensor correction unit 13 corrects the error between the inertial sensor value and the sensor value for mobile objects obtained in step S101, using the sensor correction value obtained in step S102, and outputs the corrected error to the filter unit 15.

[0028] Next, in step S104, the swim angle estimation unit 12 calculates the swim angle of vehicle 1 based on the steering angle and vehicle speed of vehicle 1 received from the moving object sensor 8, and the yaw rate of vehicle 1 received from the inertial sensor 11, and outputs the swim angle to the inertial positioning unit 14.

[0029] Here, a method for estimating the sideslip angle using the sideslip angle estimator 12 is described. The sideslip angle estimator 12 calculates a state of the vehicle 1 using a plurality of vehicle motion models and a weighting function. In embodiment 1, it is described that the sideslip angle estimator 12 calculates a state of the vehicle 1 using two vehicle motion models and a weighting function. However, the sideslip angle estimator 12 can also use three or more vehicle motion models or a plurality of weighting functions to calculate a state of the vehicle 1.

[0030] Fig. Figure 4 schematically illustrates a first motion model. This first motion model is a dynamic model that uses an equation of motion in the transverse direction and the rotation of a vehicle. Since this model allows the calculation of vehicle motion according to the force generated by the tires, it can express, with high accuracy, vehicle motion, especially at high speeds, where lateral acceleration is generated when turning or cornering. Initially, a vehicle state variable x and an input u are specified in a first motion model f1 as shown below. [Mathematical expression 1] x=[X, Y, θ, V, γ, β, δ, ] u=[ω, jx]T

[0031] In equation (1), X, Y, and θ denote the position of the center of gravity and the azimuth angle of vehicle 1 in an inertial coordinate system, V denotes the vehicle speed, γ denotes the yaw rate, β denotes the sideslip angle, δ denotes the front wheel steering angle, and a x denotes the longitudinal acceleration. In equation (2), ω denotes the front wheel steering angular velocity, and j x This refers to a longitudinal jerk or longitudinal over-acceleration.

[0032] The first motion model f1 is expressed by the equation below, using variables in equations (1) and (2). [Mathematical expression 2] x˙=f1(x,u)=[Vcos(θ+β)Vsin(θ+β)γax2I(IfYf−IrYr)−γ+2MV(Yf+Yr)ωjx]

[0033] Here, M denotes the mass of the vehicle, V denotes the vehicle speed, I denotes the yaw moment of the vehicle, l rdenotes the distance from the vehicle's center of gravity to the front axle, and l r Y denotes the distance from the vehicle's center of gravity to the rear axle. f and Y r denote the lateral forces of the front and rear wheels, and they are determined by the equations below using the lateral stiffnesses K. f and K r expressed as the front wheels or the rear wheels. [Mathematical expression 3] Yf=−Kf(β+LfVγ−δ) Yr=−Kr(β−LrVγ)

[0034] Rearranging equation (3) using equations (4) and (5) leads to the following equation, which is a dynamic motion model f. [Mathematical expression 4] x˙=f1(x, u) = |

[0035] Fig. Figure 5 schematically illustrates a second motion model. This second motion model is a geometric model calculated from the vehicle's geometric relationship. Unlike the first motion model, this model can accurately represent the vehicle's motion at low speeds, where the vehicle steers in the direction of the tires without considering the force generated by the tires. A vehicle state variable x and an input u in the second motion model are identical to those of the first motion model. A second motion model f2 is expressed by the equation below, using the variables in equations (1) and (2). [Mathematical expression 5] x˙=f2(x,u)=[Vcos(θ+β)Vsin(θ+β)γaxγkm−γτβkm−βτωjx]

[0036] Here, τ denotes the time constant of the yaw rate γ and the sideslip angle β. Different values ​​of τ can be used for the yaw rate γ and the sideslip angle β. Furthermore, γ denotes km and β km The yaw rate and the sideslip angle can be calculated from a two-wheeled model using a geometric relation, and they are expressed by the respective equations below. [Mathematical expression 6] γkm=Vlf+lrtan(δ) βkm=tan−1(lrlf+lrtan(δ))

[0037] Rearranging equation (7) using equations (8) and (9) leads to the following equation, which is a dynamic motion model f2. [Mathematical expression 7] x˙=f2(x, u)=[Vcos(θ+β)Vsin(θ+β)γax−γτ+1τVlf+lrtan(δ)−βτ+1τtan−1(lrlf+lrtan(δ))ωjx]

[0038] Although the first motion model f1, expressed by equation (6), and the second motion model f2, expressed by equation (10), have the same vehicle state variable x and the same input u, the first motion model f1 and the second motion model f2 differ by the differential equation with respect to the yaw rate γ and the sideslip angle β. The first motion model and the second motion model are not restricted to the models mentioned above if they have the same vehicle state variable x and the same input u and have different differential equations with respect to at least one vehicle state variable.

[0039] The second motion model can, for example, be a dynamic model that uses an equation of motion in the transverse direction and the rotation of the vehicle when cornering in a stable state. Although this model, unlike the first motion model, cannot express transient motion, it can express vehicle motion at low vehicle speeds with high accuracy. The second motion model, which uses this second model, is expressed by the equation below. [Mathematical expression 8] x˙=[Vcos(θ+β)Vsin(θ+β)γaxγsst−γτβsst−βτωjx]

[0040] Here, γ denotes sst and β sst The yaw rate or sideslip angle during cornering in a stable state, and they are expressed by the equations below. [Mathematical expression 9] γsst=II+AV2VIδ βsst=I+m lf2llrkrV2I+AV2lrlδ

[0041] Here, A is referred to as the stability factor and is expressed by the equation below. [Mathematical expression 10] A=−m2l2lfKf−lrKrKfKr

[0042] In embodiment 1, a model that is accurate at high vehicle speeds is used as the first motion model, whereas a model that is accurate at low vehicle speeds is used as the second motion model. This allows for the use of models with different characteristics as the first and second motion models, and the weighting of the first and second motion models using a weighting function described later, which can lead to advantages with respect to both the first and second motion models.

[0043] The weighting function is a function that weights the first and second motion models, and it is predefined with a value between 0 and 1. Since the first motion model in embodiment 1 includes a division by the vehicle speed, the value diverges near 0 km / h, and the accuracy is poor near 0 km / h. In contrast, the second motion model is a model in which the yaw rate γ and the sideslip angle β are generated by a steering angle, and the force generated in the vehicle is not taken into account. Consequently, the accuracy of the second motion model is poor at high vehicle speeds, where a centrifugal force is generated during steering.Consequently, the equation below expresses a weighting function as a function of velocity, where the weighting of the first motion model is greater at high velocities and the weighting of the second motion model is greater at low velocities. [Mathematical expression 11] α=V2V2+Vs2

[0044] V s is a parameter of a weighting function. If V=V s If this is true, then the weighting of the first motion model is equal to that of the second motion model. Fig. Figure 6 illustrates an example weighting function at this speed. A new vehicle motion model f is created using the first motion model, the second motion model, and this weighting function. [Mathematical expression 12] x˙=f(x,u)=αf1(x,u)+(1−α)f2(x,u)

[0045] Rearranging this leads to the following equation. [Mathematical expression 13] x˙=f(x, u)=[Vcos(θ+β)Vsin(θ+β)γaxV2V2+VS2(−2IV(KfIf2+KrIr2)γ−2I(KfIf−KrIr)β+2IKfIfδ)+VS2V2+VS2(−γτ+1τVIf+I rtan(δ))V2V2+VS2(−(1+2(KfIf−KrIr)MV2)γ−2MV(Kf+Kr)β+2KfMVδ)+VS2V2+VS2(−βτ+1τtan−1(IrIf+Irtan(δ))ωjx]

[0046] This vehicle motion model f is created by a differential equation that has the same vehicle state variable x and the same input u as those of the first motion model f1 and the second motion model f2. This vehicle motion model f is subsequently referred to as a hybrid model. As in Fig. As shown in Figure 7, by creating a vehicle motion model in this way, a vehicle state variable is calculated by the first motion model f1 when the vehicle speed is high, and a vehicle state variable is calculated by the second motion model f2 when the vehicle speed is low.

[0047] Consequently, the vehicle state variable can be calculated with high accuracy at any vehicle speed. Since the first motion model f1 and the second motion model f2 transition seamlessly into each other via the weighting function α, the vehicle state variable can be calculated without a jump. Furthermore, because the number of vehicle state variable elements x and the number of control input elements u are consistent between the first motion model f1 and the second motion model f2, an increase in the calculated value can be suppressed more effectively than in a high-accuracy motion model where one vehicle state variable has been increased. In other words, the swim angle estimation unit 12 generates a blended model by weighting a plurality of motion models based on the vehicle speed 1.

[0048] In embodiment 1, the weighting function α is a function where the denominator and numerator each contain quadratic terms representing the vehicle speed V. However, the weighting function α can also be any function, such as a polynomial function or an exponential function, as long as the weighting function α is specified with a value between 0 and 1.

[0049] Although the weighting function α is a function whose variable is the vehicle speed V, the weighting function α can also be a function with a different variable that depends on the first or second motion model. Furthermore, the weighting function α can be used that differs for each vehicle state variable to be calculated. For example, suppose that V s1 the speed for switching the weighting function of the yaw rate γ is denoted and V s2If the speed for switching the weighting function of the swim angle β is denoted, then the mixed model f is expressed as below. [Mathematical expression 14] x˙=f(x,u)=[Vcos(θ+β)Vsin(θ+β)γaxV2V2+VS12(−2IV(KfIf2+KrIr2)γ−2I(KfIf−KrIr)β+2IKfIfδ)+VS2V2+VS12(−γτ+1τV If+Irtan(δ))V2V2+VS22(−(1+2(KfIf−KrIr)MV2)γ−2MV(Kf+Kr)β+2KfMVδ)+VS2V2+VS2(−βτ+1τtan−1(IrIf+Irtan(δ))ωjx]

[0050] Therefore, integrating variations of the swim angle β obtained from the mixing model f can estimate the current swim angle β.

[0051] This allows the skew angle β to be continuously maintained, and z to be constant from the time when vehicle 1 stops until when vehicle 1 is traveling at high speed, and vehicle 1 can be positioned with high accuracy.

[0052] Reference is made again to Fig. 3 taken. After step S104, the inertial positioning unit 14 calculates variations of the position and attitude of the vehicle 1 using the sensor correction value received from the sensor correction unit 13, updates the variations and outputs the variations as a result of the inertial positioning to the filter unit 15 in step S105.

[0053] Next, the filter unit 15 estimates the error between the inertial positioning result, the output from the inertial sensor and the output from the mobile object sensor, using the inertial positioning result and the satellite positioning result output from the satellite positioning results receiver unit 10, and outputs the estimated error in steps S106 to S108.

[0054] More precisely: Filter unit 15 determines whether the satellite positioning result has been updated in step S106 of this calculation cycle.

[0055] In general, the calculation cycle for satellite positioning is equivalent to, and slower than, that of the inertial sensor 11 and the sensor 8 for mobile objects. Consequently, if the satellite positioning result is not updated, the filter unit 15 also does not update the sensor correction values ​​for the inertial sensor 11 and the sensor 8 for mobile objects, but instead outputs a positioning calculation result based on the sensor correction value that was previously estimated, the inertial sensor value, the sensor value for mobile objects, and the estimated result of the swimming angle (step S107).

[0056] When the satellite positioning result has been updated for the calculation of filter unit 15, filter unit 15 calculates a sensor correction value using the satellite positioning result, the inertial sensor value, and the sensor value for mobile objects (step S108).

[0057] More precisely: The filter unit 15 applies a filter to determine the sensor correction values ​​of the inertial sensor 11 and the sensor 8 for mobile objects to the error between the satellite positioning result and the inertial positioning result, which is called loose coupling, using a filter that includes an extended Kalman filter.

[0058] As in Fig. 4 and Fig. As illustrated in Figure 5, the direction of travel ψ of the vehicle is expressed by the equation below, using the azimuth angle θ and the sideslip angle β of the vehicle. [Mathematical expression 15] ψ=θ / β

[0059] Since the azimuth angle obtained by satellite positioning is determined by the amount and speed of movement of the antenna, the azimuth angle θ of the vehicle is expressed by the equation (19) below. [Mathematical expression 16] θ=ψ−β

[0060] The inertial positioning unit 14 calculates a result of the inertial positioning, for example, by defining a state variable, as expressed below, using the azimuth angle θ of the vehicle 1, taking into account the sideslip angle β, which has been obtained as described above. [Mathematical expression 17] yd=[λd ϕd hd θd]T y dIn equation (21), denotes a state vector at inertial positioning, obtained by combining the state variables over inertial positioning. Furthermore, λ denotes d the latitude, φ d denotes longitude, and h d denotes the ellipsoidal height, all of which were obtained through inertial positioning calculations. θ d denotes the azimuth angle taking into account the skew angle β, which has been obtained by equation (20).

[0061] A vector obtained by differentiating y over time a The result obtained in equation (21) is given below. [Mathematical expression 18] y˙d=g(yd,u)

[0062] The following input is expressed as a vector u. [Mathematical expression 19] [V^,γ^]T [Mathematical expression 20] V^,γ^ denote the vehicle speed V and the yaw rate γ, between which an error has been corrected by means of the sensor correction unit 13.

[0063] Substituting y d and the input u in the directly preceding step in [Mathematical expression 21] g(yd,u) and a momentary integration of this quantity makes it possible to obtain a result of inertial positioning.

[0064] The filter unit 15 estimates the error between a state variable and a sensor value using a Kalman filter, taking into account the result of the inertial positioning, considering the float angle β, which has been obtained in the manner described above.

[0065] More precisely: Filter unit 15 performs the estimation using the following sensor models. [Mathematical expression 22] V=(I+sv)Vt γ=(I+sγ)(γt+bγ)

[0066] Equation (22) is a model in which a true value Vt of the vehicle speed is multiplied by a scaling factor S. v is multiplied by the vehicle speed. Equation (23) is a model in which a preload or pre-displacement b γ a yaw rate sensor the true value γ t The yaw rate is superimposed and the resulting value is scaled with a factor s. γ The yaw rate is multiplied.

[0067] In this example, filter unit 15 estimates values ​​s as a sensor error. ve , s γe and b γe from s V , s γ or b γ The sensor correction unit 13 corrects sensor values ​​of the sensor 8 for mobile objects and the inertial sensor 11 by the equations below, using the estimated values ​​generated by the filter unit 15. [Mathematical expression 23] Ve−1(1+sVe)V γe=11+sγλ−bγe

[0068] Filter unit 15 defines a state vector as specified by the equation below. [Mathematical expression 24] x=[λ ϕ h θ sv sγ bγ]T

[0069] Assume the scaling factor S v the vehicle speed and the scaling factor s γ If the yaw rate is very small, then the true value Vt of the vehicle speed and the true value γ can be determined. t The yaw rate can be approximated by the respective equations below. [Mathematical expression 25] Vt=(I+sV)V γt=(I−sγ)γ−bγ

[0070] Dynamic models of the scaling factor S v the vehicle speed, the scaling factor s γ the yaw rate and the bias b γThe yaw rate sensor's behavior is expressed by the equations below. The dynamic models are driven by a first-order Markov process to predict the next state from the current state. [Mathematical expression 26] s˙V=(−sV+wsV) / τsV s˙γ=(−sγ+wsγ) / τsy b˙γ−(−bγ+wbγ) / τbγ

[0071] In equations (29) to (31): [Mathematical expression 27] s˙V denotes the time derivative of s V ; [Mathematical expression 28] s˙γ denotes the time derivative of s γ ; and [Mathematical expression 29] b˙γ denotes the time derivative of b γ Furthermore, τ denotes sV a model parameter value of a vehicle speed scaling factor. W sVdenotes disturbances in the temporal transition of the vehicle speed scaling factor. τ sγ denotes a model parameter value of the yaw rate scaling factor. W sγ denotes disturbances in the time transition of the yaw rate scaling factor. τ bγ denotes a model parameter value of a yaw rate bias or pre-displacement. w bγ refers to disturbances in the temporal transition of the yaw rate bias.

[0072] Rearranging equations (29) to (31) expresses the equation of the state by the following equation (32). [Mathematical expression 30] x˙=f(x,u)

[0073] In equation (32) [Mathematical expression 31] x˙ denotes a vector obtained by differentiating a state vector x over time. Furthermore, u denotes the following input vector. [Mathematical expression 32] [V,γ]T

[0074] Next, an observation value from a Kalman filter, obtained from the satellite positioning results receiver 10, is described. The satellite positioning results receiver 10 outputs coordinate information, such as latitude, longitude, and antenna 5 altitude. Observation values ​​from a GNSS sensor are subsequently described using (λ m , φ m , h m , ψ m These coordinate information elements are obtained from the result of the inertial positioning. However, the result of the inertial positioning is the coordinates of a vehicle at the navigation center. Consequently, the observed values ​​of the GNSS sensor are predicted using an offset value from the vehicle's navigation center to the position of antenna 5.

[0075] In other words, assuming (Δx, Δy, Δz) denotes the offset value from the vehicle's navigation center to antenna 5, expressed in a vehicle navigation coordinate system, then the predicted observation values ​​of the GNSS sensor (λp, φp, hp, ψp) are determined from an inertial positioning value y as given by equation (33) below. d and an offset value v (Δx, Δy, Δz) using a coordinate conversion function c (y d , v). [Mathematical expression 33] [λpϕphpψp]=c(yd,v)

[0076] This allows a state to be estimated using the state equation from equation (32) and the observation equation from equation (33) with the aid of an extended Kalman filter. This yields an estimated value of the state vector from equation (26) and allows the determination of a position error and a sensor error.

[0077] The inertial positioning unit 14 uses an azimuth angle taking into account the float angle, as described above, so that the accuracy of the inertial positioning result and the accuracy of the state variable and sensor error estimated by the filter unit 15 are improved.

[0078] As described above, the positioning device 7 for mobile objects according to embodiment 1 comprises the following: the sensor information acquisition unit, which acquires sensor values ​​including the steering angle, vehicle speed and angular velocity of the vehicle 1, which are detected by the sensor 8 for mobile objects and the inertial sensor 11, which are sensors attached to the vehicle 1, which is a mobile object; the inertial positioning unit 14, which performs inertial positioning of the vehicle 1 using the sensor values; and the sideslip angle estimation unit 12, which estimates the sideslip angle of the vehicle 1 using the sensor values.

[0079] Then the swim angle estimation unit 12 estimates the swim angle based on the sensor values ​​and the mixed model f, which is obtained by weighting the plurality of motion models f1 and f2, which have different differential equations, with respect to at least one of the state variables of vehicle 1, based on the state variables of vehicle 1, and by integrating the motion models f1 and f2.

[0080] Consequently, the positioning device 7 for mobile objects can estimate the yaw angle of vehicle 1 even when vehicle 1 is not in a stable state, and can estimate the position and attitude of vehicle 1 with high accuracy. For example, when the positioning device 7 for mobile objects is used for an application such as automated driving, it improves the control safety and driving comfort of vehicle 1.

[0081] The positioning device 7 for mobile objects further comprises: the filter unit 15, which estimates a sensor error contained in each of the sensor values, using the result of the inertial positioning and the swimming angle; and the sensor correction unit 13, which corrects the sensor error contained in the sensor value; and the inertial positioning unit 14, which can perform the inertial positioning using the sensor value corrected by the sensor correction unit 13. A-3. Modifications

[0082] In the configuration according to Fig. 2. The sensor values ​​from sensor 8 for mobile objects and inertial sensor 11 are entered into the swim angle estimation unit 12. However, the sensor values ​​that are corrected by the sensor correction unit 13 can also be entered into the swim angle estimation unit 12, as shown in Fig. Figure 8 illustrates this. In other words, the swim angle estimation unit 12 can estimate the swim angle based on the sensor values, which are corrected by the sensor correction unit 13. This reduces the error contained in the input to the swim angle estimation unit 12 and improves the accuracy of estimating the swim angle.

[0083] In the configuration according to Fig. 2. The estimated result of the swim angle is output to the filter unit 15 by the swim angle estimator 12. However, the estimated result of the swim angle by the swim angle estimator 12 can also be output to the inertial positioning unit 14, as shown in Fig. Figure 9 illustrates this. In other words, the inertial positioning unit 14 can perform inertial positioning taking into account the float angle, using the float angle.

[0084] More precisely: The inertial positioning unit 14 performs an inertial positioning calculation taking into account the swim angle β by changing the observed values ​​(λm, φm, hm, ψm) of the GNSS sensor into values ​​that take the swim angle β into account, i.e. (λ m , φ m , h m , ψ m - β). This improves the accuracy of estimating the azimuth angle using inertial positioning.

[0085] Furthermore, the inertial positioning unit 14 can contain the float angle estimation unit 12, as shown in Fig. Figure 10 illustrates this. In other words, the inertial positioning unit 14 can perform the inertial positioning calculation using the model that is used by the swim angle estimation unit 12 to estimate the swim angle.

[0086] In other words, the positioning device 7 for mobile objects may comprise: the sensor information acquisition unit, which acquires sensor values ​​including the steering angle, vehicle speed, and angular velocity of the vehicle 1, detected by the sensor 8 for mobile objects and the inertial sensor 11, which are sensors attached to the vehicle 1; the sensor correction unit 13, which corrects any sensor error contained in each of the sensor values; the inertial positioning unit 14, which performs inertial positioning of the vehicle 1 using the sensor value corrected by the sensor correction unit 13; and the filter unit 15, which estimates the sensor error contained in the sensor value using the result of the inertial positioning and the sideslip angle of the vehicle 1.

[0087] Furthermore, the inertial positioning unit 14 can estimate the sideslip angle based on the sensor values ​​and the mixed model f, which is obtained by weighting the plurality of motion models f1 and f2 with different differential equations with at least one of the state variables of the vehicle 1, based on the state variables of the vehicle 1, and integrating the motion models f1 and f2.

[0088] More precisely: The front wheel steering angle δ can be added to the input, which is expressed below through a Kalman filter. [Mathematical expression 34] [V^,γ^]T

[0089] Furthermore, a model of the azimuth angle θ of vehicle 1 in an equation of a state, as expressed below, which is to be used for a Kalman filter, can be changed into a more detailed model that is not limited to the integration of yaw rates, but is the one expressed by equation (18). [Mathematical expression 35] x˙=f(x,u)

[0090] Thus, by weighting and estimating a plurality of motion models using the inertial positioning unit 14, the accuracy of the motion models to be used for inertial positioning is increased and the estimation accuracy is improved.

[0091] In embodiment 1, the latitude, longitude, altitude, and azimuth obtained through positioning calculations using the GNSS sensor are used as the GNSS observation values. Since some GNSS sensors can output raw data, such as a pseudo-range, Doppler data, or carrier phases, close coupling can be employed using this data as the observation values. This may occasionally necessitate adding, for example, the drift of the GNSS sensor's receiver time to the state variable and estimating the receiver time error. However, even if the number of visible satellites is small, for example, only one, the close coupling can still generate the GNSS observation values. This enables highly accurate positioning, even with a low number of visible satellites.

[0092] Although the number of antennas 5 in embodiment 1 is one, a structure with a plurality of antennas can also be used, enabling the calculation of the azimuth using the positions and velocities of the antennas. In other words, the filter unit 15 estimates the direction of travel of the vehicle 1 using satellite signals received from the satellite 4 by the plurality of antennas 5 attached to the vehicle 1, and estimates the sensor error based on the result of the inertial positioning, the yaw angle, and the direction of travel of the vehicle 1. This allows the filter unit 15 to calculate the motion or rotation of the vehicle from the relative positions of the antennas 5 and to estimate the azimuth with higher accuracy. A-4. Hardware configuration

[0093] A processing circuit 81 in Fig. 11 implements the satellite positioning results receiver 10, the swim angle estimator 12, the sensor correction unit 13, the inertial positioning unit 14, and the filter unit 15 in the above positioning device 7 for mobile objects. In other words, the processing circuit 81 includes the satellite positioning results receiver 10, the swim angle estimator 12, the sensor correction unit 13, the inertial positioning unit 14, and the filter unit 15 (hereinafter referred to as "the swim angle estimator 12, etc."). This processing circuit 81 can be dedicated hardware, or it can be a processor that executes a program stored in memory. The processor could be, for example, a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, or a digital signal processor (DSP).

[0094] If the processing circuit 81 is dedicated hardware, then the processing circuit 81 corresponds, for example, to a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an application-specific integrated circuit (ASIC), or a field-programmable gate array (FPGA), or any combination thereof. The functions of each of the units, i.e., the float-angle estimation unit 12, etc., can be implemented by a plurality of processing circuits 81, or the functions of the units can be implemented collectively by a single processing circuit.

[0095] If the processing circuit 81 is a processor, the functions of the swim angle estimation unit 12, etc., are implemented by some combination of software, etc. (software, firmware, or software and firmware). For example, the software is described as a program and stored in memory. As in Fig. As shown in Figure 12, a processor 82, which is to be used as the processing circuit 81, implements the functions of each of the units by reading and executing a program stored in a memory 83. In other words, the positioning device 7 for mobile objects has the memory 83 for storing a program which, when executed by the processing circuit 81, results in the execution of the functions of each of the units.

[0096] In other words, this program causes a computer to execute procedures or processes for the float angle estimation unit 12, etc. Examples of memory 83 here include: non-volatile or volatile semiconductor memory, such as selective access memory (RAM), read-only memory (ROM), flash memory, erasable programmable read-only memory (EPROM), and electrically erasable programmable read-only memory (EEPROM), hard disk drive (HDD), magnetic disk, flexible disk, optical disk, compact disk, minidisc, digital versatile disc (DVD), a drive mechanism therefor, and furthermore, any storage medium to be used in the future.

[0097] The configuration for implementing the functions of the swim angle estimation unit 12, etc., using hardware and software, etc., is described above. However, the configuration is not limited to this; part of the swim angle estimation unit 12, etc., can be implemented using dedicated hardware, and another part can be implemented using software, etc.

[0098] The embodiments can be freely combined, suitably modified, or omitted. The above description is purely illustrative. It is understood that numerous modifications can be used, which have not been described here as examples.

[0099] Although vehicle 1 is described as an example of a mobile object, the concept of a mobile object is not limited to vehicles. Examples of mobile objects include various types of inspection robots and personal mobility devices. Reference symbol list 1 vehicle 2 Steering wheel 3 Steering actuator 4 satellites 5 antenna 6 Drive 7 Positioning device for mobile objects 8 Sensor for mobile objects 9 Vehicle control unit 10 Receiving unit for satellite positioning results 11 Inertial Sensor 12 Swimming angle estimation unit 13 Sensor correction unit 14 Inertial positioning unit 15 filter units. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2020- 112 490 A

[0005]

Claims

[1] Positioning device for mobile objects, comprising: a sensor information acquisition unit for acquiring a sensor value of a mobile object, the sensor value being detected by a sensor; a float angle estimation unit for estimating the float angle of the mobile object using the sensor value; and an inertial positioning unit for performing inertial positioning of the mobile object using the sensor value and the float angle, wherein the slip angle estimation unit estimates the slip angle based on a mixed model and the sensor value, the mixed model being obtained by weighting a plurality of motion models of the mobile object based on state variables of the mobile object and integrating the plurality of motion models. [2] Positioning device for mobile objects, comprising: a sensor information acquisition unit for acquiring a sensor value of a mobile object, the sensor value being detected by a sensor; a sensor correction unit for correcting a sensor error included in the sensor value; an inertial positioning unit for performing inertial positioning of the mobile object using the sensor value corrected by the sensor correction unit; and a filter unit for estimating the sensor error contained in the sensor value using a result of the inertial positioning and the float angle of the mobile object, wherein the inertial positioning unit estimates the slip angle based on a mixed model and the sensor value and performs the inertial positioning using the slip angle, wherein the mixed model is obtained by weighting a plurality of motion models of the mobile object based on state variables of the mobile object and integrating the plurality of motion models. [3] A positioning device for mobile objects according to claim 1, comprising: a filter unit for estimating a sensor error included in the sensor value using a result of the inertial positioning and the float angle; and a sensor correction unit for correcting the sensor error contained in the sensor value, wherein the inertial positioning unit performs the inertial positioning using the sensor value corrected by the sensor correction unit. [4] The mobile object positioning device according to claim 3, wherein the slip angle estimation unit estimates the slip angle using the sensor value corrected by the sensor correction unit. [5] Positioning device for mobile objects according to one of claims 1, 3 and 4, wherein the sideslip angle estimation unit generates the mixed model by weighting the plurality of motion models based on the speed of the mobile object. [6] A mobile object positioning device according to any one of claims 2 to 4, wherein the filter unit estimates the traveling direction of the mobile object using satellite signals received from a satellite through a plurality of antennas mounted on the mobile object, and estimates the sensor error based on the result of the inertial positioning, the float angle and the traveling direction of the mobile object. [7] The mobile object positioning device according to claim 2, wherein the inertial positioning unit generates the mixed model by generating the plurality of motion models based on the speed of the mobile object.

Images