TW201908765A - Real-time Precise Positioning System of Vehicle - Google Patents

Abstract

A real-time precise positioning system of vehicle, which provides a real-time precise positioning information for a vehicle, includes a cartographic material positioning module, a road vision sensing module, a vehicle-dynamic detective module, and a full-domain precisely operating module. The real-time precise positioning system of vehicle obtains a real-time high accuracy position of which the error is within a lane-level range by means of a data fusion regarding the cartographic material positioning information of the vehicle, the lane-vision information and the dynamic-state information executed by the full-domain precisely operating module.

Description

Vehicle accurate positioning system

The present invention relates to a positioning system, and more particularly to an instant precision positioning system related to an advanced driving assistance system (ADAS).

Advanced Driver Assistance Systems (ADAS) is a smart vehicle technology developed in recent years that collects and analyzes information on changes in the environment outside the vehicle to provide driver judgment and response. Specifically, the advanced driver assistance system uses the visual sensors and radar to detect environmental information (relative positioning information) around the vehicle to assist driving. However, advanced driver assistance systems are not able to detect future road and environmental conditions (absolute positioning information), such as providing a safe speed of the two-hundred-meter front interchange, thus creating a hidden risk factor in advanced driver assistance systems.

Of course, drivers can use the navigation machine to provide a wide range of road information (absolute positioning information). In detail, the navigation system receives the latitude, longitude (and sometimes height) information that is instantly returned by the satellite positioning, and cooperates with the map connected to the database, so that the driver knows the location and proximity of the current traffic vehicle. Road information for the area.

However, the existing parity navigation machine provides vehicle positioning with general grade accuracy, and cannot present more precise positioning and road information. Moreover, since the update frequency of the navigation machine is low (about once every three to three months), the map information of the database cannot be instantaneously affected by the change of the real environment, thereby affecting the way the navigation machine can be positioned. And the quality of the location information. In addition, in order to improve the quality of road and positioning information, it is generally thought to replace the wafer of the navigation machine, but the replacement cost of the wafer is high (higher than the cost of installing the navigation machine) and the navigation machine can be obtained by replacing the wafer. Positioning accuracy and quality improvement in providing road information are limited. Therefore, as far as the performance and cost of the navigation machine are concerned, it is not worth the loss.

Therefore, the object of the present invention is to provide an instant precise positioning system for a vehicle, which can obtain low-cost and high positioning accuracy map positioning even under the premise of using an inexpensive navigation machine (general accuracy level).

The technical means adopted by the present invention to solve the problems of the prior art provides a real-time accurate positioning system for a vehicle, which provides instant and accurate positioning information for a vehicle. The vehicle accurate positioning system includes: a positioning positioning module The group has a global positioning system and a high-precision electronic map, the global positioning system information obtained by the global positioning system, the high-precision electronic map has high-precision electronic map information; a road visual sensing module is configured to obtain The road visual information of the road on which the vehicle is located; a vehicle motion detection module configured to obtain motion information of the vehicle; and a global precision computing module, the signal is connected to the image positioning module, The road visual sensing module and the vehicle motion detecting module, the global precision computing module matches the global positioning system information and the high-precision electronic map information to obtain the map positioning information, and has an interactive a multi-model sub-module and a limited-loss non-destructive Kalman filter for transmitting the map positioning information transmitted by the map positioning module The road visual information transmitted by the road visual sensing module and the motion state information transmitted by the vehicle motion detecting module perform a data fusion process to obtain an instant and the error range is within one lane. High precision positioning.

In an embodiment of the invention, a vehicle accurate sensing system is provided, wherein the road vision sensing module has an image sensor and a light device, and the image sensor is configured to detect the The vehicle is located at the officially set road visual information, and the optical device is configured to measure the distance between the vehicle and the road visual information and correct the content of the road visual information. The global precision computing module And performing data fusion processing on the map positioning information and the road visual information to obtain a positioning correction information of the map positioning module.

In an embodiment of the present invention, a real-time accurate positioning system for a vehicle is provided, wherein the global precision computing module is provided with a map matching sub-module configured to provide global positioning system information and high precision. The electronic map information is matched to obtain the map positioning information, and the map positioning information, the road visual information and the motion state information are matched and transmitted to the interactive multi-model sub-module to perform the vehicle in the vertical direction. Motion, lateral motion, and oblique motion calculations.

In an embodiment of the invention, a carrier accurate positioning system is provided, wherein the vehicle motion detection module has an inertial measurement unit and a vehicle dynamic sensor, and the inertial measurement unit is configured to detect the The vehicle's pitch sensor, the steering, the roll, and the steering rate are configured to detect the vehicle speed and the steering angle of the vehicle. The interactive multi-model sub-module is for the vehicle motion detection module. The motion state information, the map positioning information, and the map matching sub-module perform data fusion processing to obtain positioning correction information of the vehicle, and the positioning correction information of the vehicle is used as the limited-type lossless Kalman filter. A positioning operation limitation condition of the image positioning module.

In an embodiment of the present invention, a carrier accurate positioning system is provided, wherein the limited-type lossless Kalman filter is configured to perform data fusion according to the map matching sub-module and the interactive multi-model sub-module. And forming a limit sampling point to correct the positioning information of the image positioning module.

Through the technical means adopted by the present invention, the vehicle real-time precise positioning system obtains the latitude and longitude positioning of the position of the vehicle and the accuracy of the lane line level through an inexpensive global positioning system and a high-precision electronic map (Lane -Level) Highly accurate road map. At the same time, the vehicle's real-time precision positioning system detects the officially set road visual information (such as road traffic signs, markings and signs) of the vehicle's location by means of the road visual sensing module, and uses the vehicle motion detection. The test module detects the motion state information of the vehicle (ie, the pitch, steering, roll, steering rate, vehicle speed and steering angle of the vehicle). Moreover, the vehicle accurate positioning system performs data fusion processing for the image positioning module, the road visual sensing module and the vehicle motion detecting module through the global precision computing module to obtain the error range in one lane. High precision positioning.

Hereinafter, embodiments of the present invention will be described based on Figs. 1 to 4 . This description is not intended to limit the embodiments of the invention, but is an embodiment of the invention.

Referring to FIG. 1 , a vehicle instant precise positioning system 100 according to an embodiment of the present invention is configured on a carrier and provides instant accurate positioning information for the vehicle. The vehicle accurate positioning system comprises a picture positioning module 1, a road visual sensing module 2, a vehicle motion detecting module 3 and a global precision computing module 4. Moreover, the vehicle instant precise positioning system 100 of the embodiment of the present invention is suitable for an instant detection system of an advanced driving assistance system A (ADAS) to instantly determine the relative position and absolute position of the road where the vehicle is located, and improve the instant. The reliability of detection. Through the data fusion (Data Fusion) processing of the global precision computing module 4, the advanced driving assistance system A provides the service functions of "high-precision positioning A1", "vehicle dynamic behavior A2" and "map information A3". Therefore, the visual sensor of the advanced driving assistance system A is not affected by the weather, the line of sight and the road condition, and the detection of the road edge is invalid.

Referring to FIGS. 1 and 2 , the map positioning module 1 has a global positioning system 11 and a high-precision electronic map 12 . Specifically, the Global Positioning System (GPS) 11 is a navigation machine, as long as it is sufficient to provide global positioning system information (immediate latitude and longitude positioning information) of general level accuracy. Therefore, in the present invention, the global positioning system 11 can optionally use an inexpensive commercial navigation machine to reduce the installation cost of the overall system. The high-precision electronic map 12 is a pre-set high-precision positioning point database (with an error of 2 cm) to provide high-precision electronic map information. The high-precision electronic map information includes road information such as the starting and ending coordinates of the road section, the lane width, the number of lanes, the heading angle of the road, the curvature of the road, and the length of the road section. Moreover, the high-precision electronic map 12 is corrected by the RTK (Real-Time Kinematic) of the National Land Surveying Center, and is projected onto an absolute coordinate by coordinate conversion.

Referring to FIG. 1 , the road visual sensing module 2 has an image sensor 21 and a light reaching device 22 . The road visual sensing module 2 is configured to obtain road visual information of the road on which the vehicle is located. Specifically, the image sensor 21 can be a CCD (Charge Couple Device) image sensor, and is configured to detect that the position of the vehicle is officially set road visual information. The road visual information set by the official is a road traffic sign, a marking line or a sign. Further, the image sensor 21 is used for detecting and recognizing a lane line of the vehicle (for lateral correction of the vehicle), a stop line, and a road number (for longitudinal correction of the vehicle) ). The LiDAR (Light Detection And Ranging) device 22 is a device that measures a target with laser light. In the embodiment of the present invention, the optical device 22 is configured to measure the distance between the vehicle and the visual information of the road to assist in correcting the detection of the visual information of the road by the image sensor 21 due to the influence of illumination. error. In other words, the light reaching device 22 can measure the distance between the carrier and the environment without being affected by the illumination, thereby performing positional errors in the lateral and longitudinal directions of the carrier.

Referring to FIG. 1 , the vehicle motion detection module 3 is configured to obtain motion state information of the vehicle, and has an inertial measurement unit 31 and a vehicle motion sensor 32 . In detail, the Inertial Measurement Unit (IMU) 31 is a device for measuring an object's triaxial attitude angle (or angular rate) and acceleration. In the embodiment of the present invention, the inertial measurement unit 31 is configured to detect The pitch, the Yaw, the Roll, and the Yaw Rate of the vehicle under motion are measured. In an embodiment of the invention, the vehicle dynamics sensor 32 is configured to detect the wheel speed and steering angle of the vehicle. Specifically, the vehicle dynamics sensor 32 measures the wheel speed of the vehicle with a rotational angle encoder while detecting the steering angle of the vehicle (ie, the yaw angle of the vehicle) with an angle sensor.

Referring to FIG. 1 to FIG. 2 , the global precision computing module 4 has a map matching sub-module 41 , an interactive multi-model sub-module 42 and a restricted lossless Kalman filter 43 . The global precision computing module 4 signal is connected to the image positioning module 1, the road visual sensing module 2 and the vehicle motion detecting module 3 for transmitting the image positioning module 1 The data positioning information, the road visual information transmitted by the road visual sensing module 2, and the motion state information transmitted by the vehicle motion detecting module 3 are performed to perform a data fusion process. The positioning correction information of the image positioning module 1 is obtained, and a high-precision positioning with an immediate error range within one lane is obtained.

In the embodiment of the present invention, the map matching sub-module 41 performs a map matching calculation process on the received information. Referring to FIG. 2 and FIG. 3 , first, the map matching sub-module 41 searches for the candidate road segment N according to the current position of the vehicle, that is, the map matching sub-module 41 is obtained by the global positioning system 11 . The GPS information is matched to the high-precision electronic map information in the high-precision electronic map 12, thereby obtaining the map positioning information.

The map matching sub-module 41 then performs a "full search O" to perform an operation step P. Specifically, the operation step P (1) locates the projection point of the current position on all candidate segments, and the global positioning is performed. The system 11 obtains the latitude and longitude information of the location of the vehicle in real time, and simultaneously performs the road visual information of the road visual sensing module 2, the information of the latitude and longitude of the global positioning system 11, and the high-precision electronic map 12. The coordinates are defined and converted to ascertain that the vehicle is currently in the position displayed by the high precision electronic map 12. (2) calculating the distance between the positioning position and the determined projection point, calculating the latitude and longitude information of the global positioning system information, the road visual information around the vehicle, and the position information of the vehicle on the high-precision electronic map information. . (3) The best candidate link is confirmed based on the result of (2), that is, the optimum position of the carrier on the high-precision electronic map 12 is confirmed. And (4) confirming the driving direction according to the elevation angle of the vehicle and the elevation angle of the road (following the road section or the reverse direction along the road section), referring to the map matching sub-module 41 referring to the high-precision electronic map information, the road visual information And the motion state information of the vehicle to confirm the driving direction of the vehicle on the road section.

The map matching sub-module 41 executes "confirm map matching result Q" to confirm "consistency R of map history data and new map matching information". Specifically, in the step of confirming the "consistency R of the map history data and the new map matching information", first (1) confirming the number of the matching road segment, based on the vehicle information in the global positioning system information, the high precision electronic The location displayed by the map information and the visual information of the road around the vehicle are matched to know the road segment where the vehicle is located. (2) Confirming the elevation angle of the matching road segment, matching the elevation information of the road segment presented by the high-precision electronic map information and the pitch detected by the inertial measurement unit 31 to the vehicle. (3) Confirming the approximate value between the projection point (reference node) and the positioning point, that is, confirming that the difference between the global positioning system information, the high-precision electronic map information, the road visual information, and the motion state information of the vehicle is in accordance with the standard .

The map matching sub-module 41 continues the "pass consistency check S". When the confirmation result is "matching section T", the map matching sub-module 41 "stops the initial stage and inputs the next stage U" to perform another stage of map matching work. If it is confirmed that the matching result is no (that is, the global positioning system information, the high-precision electronic map information, the road visual information, and the difference between the motion state information of the vehicle is too large), the map matching sub-module 41 is re-established. Perform "Comprehensive Search O" to perform the operation step P again.

Referring to FIG. 2, the interactive multi-model sub-module 42 is configured to use an interactive multiple model (IMM) algorithm to locate the information, the road visual information, and the motion state information. Performing an operation to obtain a positioning correction information about the image positioning module 1. In the embodiment of the present invention, the global sensor system 11, the inertial measurement unit 31, and the original sensor information of the vehicle dynamic sensor 32 are processed by the signal processing of "confirm update and process noise B", and the global positioning is performed. The sensor information of the system information and the motion state information of the vehicle is transmitted to the interactive multi-model sub-module 42. Moreover, the map matching sub-module 41 also transmits the real-time positioning feedback data of the map positioning information to the interactive multi-model sub-module 42. Thereby, the interactive multi-model sub-module 42 is capable of performing state estimation of the dynamic behavior of the vehicle. That is, the vehicle is calculated in the longitudinal direction, the lateral (ie, lateral), and the oblique (Slope).

Specifically, the interactive multi-model algorithm used by the interactive multi-model sub-module 42 is mainly composed of an interaction, a model probability update, and an estimation fusion. The estimated mathematical expression of the interaction of the interactive multi-model algorithm is as follows: [Math 1] among them For mixing probability, For Markov transfer probability (ie the probability of converting from model j to model i), For the probability of model i at time k-1, It is normalizing constants. The mathematical formula of the normalization constant is as follows: [Math 2] The mathematical expression of the state of each model is as follows: [Math 3] And, the initial value calculation mathematical formula of each model covariance matrix is as follows: [Math 4] among them The estimated result for each vehicle model at time k-1.

The probability model update in the interactive multi-model algorithm is processed according to the innovation error generated by the measured value, and the error is assumed to have Gaussian statistical properties. Probability equation for the measured value As shown below: [Math 5] among them In order to update the vector, the mathematical expression is as follows: [Math 6] and Then it is the covariation matrix of the update vector, and its mathematical expression is as follows: [Math 7] Therefore, the update formula of the model probability in the interactive multi-model algorithm is as follows: [Math 8] Where c is a normalization constant, and the mathematical expression is as follows: [Math 9]

The estimation and combination phase of the interactive multi-model algorithm is based on the probability of each model calculated by the above mathematical formula, and further estimates the true state using the state estimated by each model and its covariation matrix. Value and its covariation matrix. The mathematical formula of the estimated value of the real state is as follows: [Math 10] The covariation matrix of the estimated values of the real state is as follows: [Math 11]

Finally, the interactive multi-model sub-module 42 performs the dynamic calculation of the vehicle according to the mathematical formula of the interactive multi-model described above. Wherein, the lateral dynamic mathematical formula of the vehicle is as follows: [Math 12] The horizontal longitudinal mathematical formula of the vehicle is as follows: [Math 13] And, the dynamic equation of the slope change of the vehicle is as follows: [Math 14] Where X and Y are the coordinate positions of the vehicle, For the heading angle of the vehicle, For speed, For the yaw rate, For the vehicle's oblique height, It is the elevation angle of the vehicle.

Referring to FIG. 2 and FIG. 4, in the embodiment of the present invention, the road attribute sensed by the road visual sensing module 2 (ie, road number, stop line, ...), the map matching submodule The group 41 calculates the map positioning information of the instant positioning and the state estimation of the dynamic behavior of the vehicle carried by the interactive multi-model sub-module 42 to further cooperate with the "restriction timing I" to form the longitudinal direction of the vehicle. Horizontal restrictions." The Constraint Unscented Kalman Filter (CUKF) 43 receives the "longitudinal and lateral constraints" of the vehicle and the sensor status transmitted by "confirm update and process noise B". For positioning operations and data fusion. The limited-type lossless Kalman filter 43 transmits the calculated optimal estimated value to the map matching sub-module 41 to obtain the map update information of the map matching sub-module 41 for instant update (the instant update) The map location information is then transmitted to the interactive multi-model sub-module 42). Specifically, the limited-type lossless Kalman filter 43 forms a limit sampling point according to the motion state information of the map matching sub-module 41 and the vehicle motion detecting module 3 to correct the image positioning mode. Group 1 location information. Referring to FIG. 4, the operation mode of the limited-type lossless Kalman filter 43 is a numerical range formed by a plurality of sampling points K, and the limited sampling point formed by the "limit timing I" is added. L and the constraint M are used to confirm the estimated value J.

In detail, the limited-type lossless Kalman filter 43 of the present invention is based on the calculation result of the UKF (Unscented Kalman Filter), and the limiting condition M is added to obtain the precision of the present invention. The result of the positioning operation. Herein, the "restricted timing" refers to a point in time when the limiting type lossless Kalman filter 43 imposes a restriction condition. For example, when the vehicle is in a transient behavior (eg, lane change), the restriction is not applied. Conversely, when the vehicle is in a steady state behavior (eg, maintained on the same lane), the constraint is imposed. The constraint condition M refers to a physical limit in a real environment. For example, when the lane width on the road section is 3.4 meters, and the vehicle is maintained in the lane, the lane width is the constraint condition. . Therefore, the estimation obtained by the limited-type lossless Kalman filter 43 is worth falling within the range defined by the constraint, and a high-precision positioning of an error range in one lane is obtained.

The Unscented Kalman Filter (UKF) has operational stages such as initialization, calculation of sampling points and weights, system status and system output update, and measurement update. The initialization mathematics is as follows: [Math 15] among them, For the state vector, A matrix of state vectors. The mathematical formula for calculating the sampling points and weights of the lossless Kalman filter is as follows: [Math 16] among them, To provide a special degree of freedom to fine tune the higher moments, To control the parameters of the sigma point distribution, L is the dimension, Usually take zero, W is the weight value, A variable is distributed for the variable. The mathematical expressions of system status and system output update are as follows: [Math 17] among them, For the vector state of the converted sigma point, for Mean, The mean and covariance of the state prediction. [Math 18] among them To measure the state, Then Mean. The mathematical expression of the measurement update of the lossless Kalman filter is as follows: [Math 19] among them To measure the variance matrix, Covariance matrix for the measured value and the state vector. [Math 20] among them, For the gain of the lossless Kalman filter, To update the state vector and the common variance matrix.

The vehicle real-time positioning system of the embodiment of the present invention obtains the latitude and longitude information of the position of the vehicle in an inexpensive manner through the global positioning system 11 of the map positioning module 1 in the above manner, and cooperates with the height. The accuracy electronic map 12 presents a highly accurate road map with a lane-level accuracy, thereby providing the driver with an accurate electronic map (absolute positioning information). At the same time, the image real-time positioning system of the embodiment of the present invention detects the location of the vehicle by the image sensor 21 of the road visual sensing module 2 as the officially set road visual information (ie, the road traffic sign) , the marking line and the number), and the longitudinal and lateral position correction between the vehicle and the officially set road visual information is obtained by the light reaching device 22, thereby obtaining instantaneous and regional positioning information (relative positioning information). Moreover, the vehicle accurate positioning system of the embodiment of the present invention detects the pitch, steering, roll and steering rate of the vehicle by the inertial measurement unit 31 of the vehicle motion detection module 3, and uses the vehicle dynamic sensor The vehicle speed and the steering angle of the vehicle are detected to match the image positioning information of the image positioning module 1 to improve the positioning effect. Specifically, the vehicle accurate positioning system 100 is configured by the global precision computing module 4 for the positioning information of the image positioning module 1 , the road visual information of the road visual sensing module 2 , and the road visual information The motion state information of the motion detection module 3 of the vehicle is subjected to data fusion processing. The map matching sub-module 41 matches the map positioning information, the road visual information, and the motion state information, and transmits the map information to the interactive multi-model sub-module 42 for calculation. The interactive multi-model sub-module 42 determines the motion behavior of the vehicle by using the information of the global positioning system 11, the inertial measurement unit 31, the vehicle dynamic sensor 32, and the map matching sub-module 41. It is also possible to instantly detect the movement behavior of the wearer in the longitudinal, lateral or oblique (climbing) as a model probability index. Then, the restricted lossless Kalman filter 43 is used to calculate the probability index of the model to obtain a high-precision positioning of the error range in one lane and correspondingly correct the positioning information of the image positioning module 1, thereby enabling advanced driving. The auxiliary system (ADAS) can assist driving by accurately detecting the future road and environment conditions by the vehicle's real-time precise positioning system to achieve improved driving safety.

The above description and description are only illustrative of the preferred embodiments of the present invention, and those of ordinary skill in the art can make other modifications in accordance with the scope of the invention as defined below and the description above, but such modifications should still be It is within the scope of the invention to the spirit of the invention.

100‧‧‧Cars Instant Accurate Positioning System

1‧‧‧Map positioning module

11‧‧‧Global Positioning System

12‧‧‧High-precision electronic map

2‧‧‧Road Visual Sensing Module

21‧‧‧Image Sensor

22‧‧‧Light device

3‧‧‧ Vehicle motion detection module

31‧‧‧ inertial measurement unit

32‧‧‧Vehicle Dynamic Sensor

4‧‧‧Global Precision Computing Module

41‧‧‧Map Matching Submodule

42‧‧‧Interactive multi-model sub-module

43‧‧‧Restricted lossless Kalman filter

A‧‧‧Advanced Driver Assistance System

A1‧‧‧High precision positioning

A2‧‧‧ Vehicle dynamic behavior

A3‧‧‧Map Information

B‧‧‧Confirm updates and handling of noise

I‧‧‧Restricted timing

J‧‧‧ Estimated value

K‧‧ ‧ sampling point

L‧‧‧Restricted sampling points

M‧‧‧Restrictions

O‧‧‧Comprehensive search

P‧‧‧ arithmetic steps

Q‧‧‧Confirm map match results

R‧‧‧Confirm the consistency of map historical data and new map matching information

S‧‧‧Confirmed by consistency

T‧‧‧ matching sections

U‧‧‧Stop the initial phase and enter the next phase

1 is a schematic diagram showing a system architecture of a vehicle instant precise positioning system according to an embodiment of the present invention; FIG. 2 is a schematic diagram showing an operation flow architecture of a vehicle instant precise positioning system according to an embodiment of the present invention; The figure shows the flow chart of the map matching algorithm. The fourth figure shows the operation of the restricted type lossless Kalman filter to impose the limit sampling point.

Claims (5)

A real-time accurate positioning system for a vehicle provides instant and accurate positioning information for a vehicle. The instant positioning system of the vehicle includes: a positioning positioning module, a global positioning system and a high-precision electronic map, the world Global positioning system information obtained by the positioning system, the high-precision electronic map has high-precision electronic map information; a road visual sensing module is configured to obtain road visual information of the road where the vehicle is located; The module is configured to obtain information on the motion state of the vehicle; and a global precision computing module, the signal is connected to the image positioning module, the road visual sensing module and the vehicle motion detecting module The global precision computing module matches the global positioning system information and the high-precision electronic map information to obtain map positioning information, and has an interactive multi-model sub-module and a limited-loss non-destructive Kalman filter. The map positioning information transmitted by the map positioning module, the road visual information transmitted by the road visual sensing module, and The motion state information transmitted by the vehicle motion detection module performs a data fusion process to obtain a high-precision positioning with an immediate and error range within one lane. The instant visual positioning system of the present invention, wherein the road visual sensing module has an image sensor and a light device, the image sensor being configured to detect the vehicle The position is an officially set road visual information, and the optical device is configured to measure a distance between the vehicle and the road visual information and correct the content of the road visual information, wherein the global precision computing module is The map positioning information and the road visual information are subjected to data fusion processing to obtain positioning correction information of the map positioning module. The instant precise positioning system of the vehicle of claim 1, wherein the global precision computing module is provided with a map matching sub-module configured to provide global positioning system information and a high-precision electronic map. The information is matched to obtain the map positioning information, and the map positioning information, the road visual information and the motion state information are matched and transmitted to the interactive multi-model sub-module to perform longitudinal movement of the vehicle. The operation of lateral motion and oblique motion. The vehicle of claim 3, wherein the vehicle motion detection module has an inertial measurement unit and a vehicle motion sensor, the inertial measurement unit configured to detect the vehicle The vehicle's dynamic sensor is configured to detect the wheel speed and steering angle of the vehicle, and the interactive multi-model sub-module is configured for the motion detection module of the vehicle. The status information, the map positioning information, and the map matching sub-module perform data fusion processing to obtain positioning correction information of the vehicle, and the positioning correction information of the vehicle is used as the limited-type lossless Kalman filter. A positioning operation limitation condition of the positioning module. The instant accurate positioning system of the vehicle of claim 3, wherein the limited-type lossless Kalman filter is configured to perform data fusion according to the map matching sub-module and the interactive multi-model sub-module, and form A limit sampling point is used to correct the positioning information of the image positioning module.