TW201724050A - Vehicle cooperative object positioning optimization method and vehicle cooperative positioning apparatus for extending a sensing range of a vehicle and improving the accuracy of positioning information so as to enhance vehicle driving safety - Google Patents

Abstract

A vehicle cooperative object positioning optimization method is disclosed, which comprises: receiving an information packet from a local vehicle, wherein the information packet includes an original vehicle coordinate and at least one original object coordinate provided by a neighbor vehicle, and respective positioning accuracies; performing a time delay compensation for the original vehicle coordinate and the original object coordinate to respectively acquire a vehicle coordinate and an object coordinate of the neighbor vehicle after the compensation; and performing an optimization procedure to respectively optimize the vehicle coordinate and the object coordinate so as to respectively obtain an optimized vehicle coordinate and an optimized object coordinate. Therefore, the optimized vehicle coordinate and the optimized object coordinate have higher accuracy than the coordinate information detected by a GPS receiver, so as to determine precisely a distribution of environmental objects and enhance driving safety.

Description

Vehicle coordinated object positioning optimization method and vehicle cooperative positioning device

This creation is an object positioning method, especially a cooperative object positioning (Cooperative positioning) optimization method.

Related technologies for configuring sensors in vehicles to detect the surrounding environment have been developed for a long time, such as using different sensors such as Global Positioning System (GPS), Radar (RADAR), LIDAR, and Travel Recorder. Provide diverse environmental information. However, for the vehicle itself, the environmental information obtained through the environmental information sensor will still be limited by many factors. For example, please refer to Figure 9, in the environment of the intersection, in the longitudinal lane A vehicle 101 can know the environmental information of the direction. For example, it can be seen that an object 200 (such as a pedestrian, a vehicle, or an animal) is rapidly rushing out of the intersection, but for the second vehicle 102 in the lateral lane, it is limited by The location thereof may be obscured by surrounding buildings to create a blind spot, even if the second vehicle 102 has a sensor, it is still unable to detect the sudden occurrence of the object 200 and collide with it. Therefore, if the environmental information is provided solely on the basis of the sensor of the vehicle itself, there is still a blind spot in actual application.

To this end, the vehicle cooperative positioning method has been developed. The concept of collaborative design means sharing the information sensed by surrounding neighboring vehicles or surrounding sensing devices (such as the road side unit, RSU). Having the vehicle receive information of other vehicles and amplifying its own sensing range. According to the example of FIG. 9 above, if the second vehicle 102 can receive the information provided by the first vehicle, the right intersection can be known. Some objects will suddenly appear and there will be enough time for emergency response.

However, the current collaborative positioning technology is faced with technical disadvantages of poor accuracy. Referring to the example shown in FIG. 10, the first vehicle 101 represents the vehicle, and the second vehicle 102 to the fourth vehicle 104 represent other vehicles in the vicinity. In the case of the second vehicle 102, it is equipped with a commercial GPS and a camera, and the positioning error of the general commercial GPS is about 5 to 15 meters, and the positioning error of the camera is about 5 meters. Therefore, the second vehicle 102 utilizes The position measured by the own GPS has a GPS error range A1. When the second vehicle 102 senses the position of the third vehicle 103 by the camera, the position of the third vehicle 103 measured by the camera has a camera error range A2. Therefore, if the second vehicle 102 relays the location information of the third vehicle that it senses to the first vehicle 101, the location information of the third vehicle 103 received by the first vehicle 101 is accumulated by error. Problems such as GPS and camera accumulation error range A3.

Further, if the first vehicle 101 transmits the position information of the third vehicle 103 twice to the fourth vehicle 104, the position error of the first vehicle 101 itself is further added, resulting in the fourth vehicle. The position information received by the vehicle 104 produces a greater amount of accumulated error. Therefore, when the information is transferred and shared many times, the accuracy of positioning the object will be significantly reduced, and even has no reference value.

In view of the problem that the existing collaborative positioning technology has poor positional information accuracy, the main purpose of this creation is to provide a vehicle coordinated object positioning optimization method, which can enhance the sensing range of the vehicle and improve the accuracy of the position information. The safety of the vehicle.

To achieve the foregoing objective, the method of the present invention is performed by a co-location device disposed inside the vehicle, the method comprising: receiving, by the vehicle, an information packet, the information packet including an original coordinate of the vehicle provided by the neighboring vehicle and at least The original coordinate of an object, the original coordinates of the vehicle and the original coordinates of the object respectively have their own positioning accuracy; time delay compensation is performed on the original coordinates of the vehicle and the original coordinates of the object to obtain a vehicle coordinate and an object compensated by the adjacent vehicle respectively. Coordinates; perform an optimization procedure, which includes: comparing the vehicle coordinates of the vehicle with the vehicle coordinates of the adjacent vehicle to determine which has higher positioning accuracy; preferentially performing an optimization operation on the vehicle coordinates with higher precision, and then The vehicle coordinates with lower positioning accuracy are optimized, wherein the optimization operation is performed: a) calculating a plurality of reference positions according to the vehicle coordinates of the vehicle and the vehicle coordinates of the adjacent vehicle; and b) weighting according to each reference position Value, calculate a vehicle optimization coordinate; optimize the object coordinates, compare the neighboring cars The difference between the original vehicle coordinates and the optimized coordinates of the vehicle, the object coordinates provided by the neighboring vehicle are compensated according to the difference amount to obtain an object optimization coordinate; and the original vehicle coordinates of the vehicle and the vehicle optimized coordinates of the vehicle are compared The amount of difference between the vehicles is compensated for the object coordinates provided by the vehicle to obtain an object optimization coordinate.

With the positioning optimization method of this creation, the optimized vehicle coordinates and object coordinates of the vehicle and the adjacent car can be obtained, and the optimized coordinates can be closer to the actual setting, which helps to accurately judge the surrounding environment during the driving process. The vehicle distribution status improves the safety of the vehicle.

Referring to FIG. 1 , the present invention utilizes a vehicle body sensor 15 provided on each vehicle, including a GPS receiver and various other sensors, such as radars, cameras, and the like to obtain environmental information of the coordinates of the vehicle and the surroundings of the vehicle. And through the wireless communication technology, the car coordinates and environmental information will be transferred to the surrounding neighboring vehicles. Therefore, in any vehicle, it is a two-way wireless communication that receives and transmits with surrounding vehicles.

The vehicle co-locating device 10 is disposed in the vehicle. The vehicle co-location device 10 includes a wireless transmission interface 11 , a delay correction module 12 , a position optimization module 13 , and a positioning comparison module 14 . The vehicle co-location device 10 performs a vehicle coordinated object positioning optimization method. The method is as shown in FIG. 2 and includes the following steps:

S10: receiving an information packet, where the information packet includes an original coordinate of a vehicle of the adjacent vehicle and at least one original coordinate of the object, and a positioning accuracy of the original coordinate of the vehicle and the original coordinate of the object;

S20: performing time delay compensation on the original coordinates of the vehicle and the original coordinates of the object to obtain a vehicle coordinate and an object coordinate compensated by the adjacent vehicle respectively;

S30: executing an optimization program;

S40: Coordinates are merged.

In step S10, the wireless transmission interface 11 is responsible for bidirectional transmission of data between the vehicle and the adjacent vehicle. In this embodiment, a short-range wireless communication interface (DSRC) is used to periodically send and receive data packets between vehicles. The data packet format is a Basic Safety Message (BSM) packet. Please refer to Figure 3. The message format of the data packet generally includes a msgID field, the first part 21 (Part I) and the second part 22 (Part II), wherein the first part 21 is defined as the necessary information, including the basic security message content, and the part that is inevitably included for each data packet, but the second part 22 is optional, which can be viewed by the user depending on the application requirements. The required information is added to the second part 22, which is a self-defined range.

In the first part 21 of the BSM data packet, there is necessarily a latitude and longitude information of the vehicle, that is, the original coordinates of the vehicle, indicating the position of the vehicle measured by the GPS receiver in the vehicle.

In the second part 22 of the BSM data package, the author adds two types of information. The first type of information is the sensor type (such as RTK, GPS) that outputs the original coordinates of the vehicle and the positioning accuracy of the sensor. The second type of information includes the original coordinates of the object, the type of sensor that produces the original coordinates of the object, or the accuracy of the sensor. The original coordinate of the object refers to the use of other sensors of the vehicle ( For example, radar, lidar, camera, etc.) sense information about objects other than the vehicle, which may be vehicles, pedestrians, moving objects or fixtures. Each vehicle can send the original coordinates of the vehicle and the original coordinates of the object to the surrounding neighboring vehicles for use; and the vehicle can also receive the original coordinates of the vehicle and the original coordinates of the object provided by the surrounding neighboring vehicles.

In step S20, the delay correction module 12 receives the BSM data packet transmitted by the neighboring vehicle through the wireless transmission interface 11, and obtains the original coordinate of the adjacent vehicle and the original coordinate of the object, and further, the position optimization module 13 further receives the The sensors provided by the vehicle provide various sensing results, such as vehicle coordinates (GPS) and object coordinates. The delay correction module 12 performs time delay compensation for the original coordinate of the vehicle provided by the adjacent vehicle and the original coordinate of the object. Referring to FIG. 4, a first vehicle 101 represents the vehicle, and a second vehicle 102 in the vicinity thereof represents the neighbor. In the vehicle, when the first vehicle 101 receives the data packet sent by the second vehicle 102, the first vehicle 101 can calculate the time according to the time when a packet is recorded in the packet and the time when the packet is received by the packet itself. A time delay amount n between the packet issuance time and the packet reception time. Since the second vehicle 102 continues to travel forward from the original position P _A to the position P _B after the packet is sent out, the original coordinates of the second vehicle 102 received by the first vehicle 101 are only representative of the original position P _A , and the delay correction mode The group 12 estimates a compensation distance moved by the second vehicle 102 according to the time delay amount n, and adds the compensation distance to know the instantaneous position P _B , and the calculation formula can be expressed as follows: P _B =P _A +V × n Where V represents the vehicle speed of the second vehicle 102.

When the delay correction module 12 estimates the instantaneous position P _B of the second vehicle 102, the compensation distance can be added to the original coordinate of the object provided by the second vehicle 102 according to the compensation distance of the two positions P _A , P _B . After the vehicle's original coordinates and the original coordinates of the object are time-compensated, a vehicle coordinate and an object coordinate are respectively obtained and provided to the position optimization module 13 for subsequent processing.

In step S30, the position optimization module 13 receives the vehicle coordinates and object coordinates of the time-compensated neighboring vehicle, and receives the vehicle coordinates of the vehicle and the object coordinates measured by the sensor of the vehicle, and performs an optimization. The program is shown in Figure 5. The optimization program includes the following processes S31~S33:

S31: Compare the positioning accuracy. Compare the vehicle coordinates of the vehicle with the vehicle coordinates of the adjacent car to determine which one has higher accuracy. For example, if the vehicle coordinates of the vehicle are coordinates obtained by using a real-time dynamic measurement (RTK) device, and the vehicle coordinates of the adjacent vehicle are received by a general GPS receiver, the coordinate provided by the RTK can be judged to have high precision. For example, if the vehicle and the neighboring car use the same level of GPS receiver to provide coordinates, the reliability of the two GPS signals can be used to determine which one has higher accuracy, for example, according to the GGA message included in the GPS signal. The reliability of the GPS signal is higher.

S32: Optimization of vehicle positioning. After determining which vehicle coordinates of the vehicle and the adjacent car are higher in accuracy, the operator will be optimized for the higher accuracy, and then the operator with lower accuracy will be optimized. Whether it is optimized for the position information of the vehicle or the neighboring car, the method is: a) calculating a plurality of reference positions according to the vehicle coordinates of the vehicle and the adjacent car; b) calculating a weight according to the weight values of the reference positions Vehicle optimization coordinates.

In this embodiment, it is assumed that the vehicle and the adjacent car are equipped with a GPS receiver and other sensors, and the vehicle optimized coordinates are calculated according to the four reference positions, and when the vehicle coordinates of the vehicle H and the neighboring vehicles are judged After the coordinates of the vehicle, it is known that the vehicle coordinates of the vehicle H have a high degree of precision. Therefore, with the vehicle H as the center, the vehicle coordinates of the vehicle are optimized first, and the steps are as follows. First, please refer to FIG. 6A, the vehicle and the adjacent car are respectively represented by H and R, and the two vehicles can know their own vehicle coordinates according to their own GPS receivers, wherein the vehicle H is known according to the GPS receiver. The coordinate is used as the first reference position H1, and the difference between the first reference position H1 and the actual position of the vehicle H is caused by the error amount of the GPS receiver, and the GPS receiver sensed by the GPS receiver of the neighboring vehicle R serves as a reference position. R1.

Referring to FIG. 6B, the sensor of the neighboring vehicle R can know the existence of the vehicle H, so that the relative distance D1 between the neighboring vehicle R and the vehicle H and the relative angle θ1 can be known. The relative coordinates of the car H. The neighboring vehicle R converts the relative coordinates of the vehicle H into a latitude and longitude coordinate according to the distance D1 and the angle θ1 with the reference position R1 of the own vehicle as the reference reference, and the latitude and longitude coordinates serve as the second reference position H2. The data packet sent by the neighboring vehicle R, that is, the latitude and longitude coordinates of the second reference position H2.

Referring to FIG. 6C, the sensor of the vehicle H can know the relative distance D2 between the vehicle H and the neighboring vehicle R and the relative angle θ2 because the presence of the neighboring vehicle R can be perceived. After the vehicle H obtains the relative distance D2 and the relative angle θ2 from the neighboring vehicle R, the latitude and longitude coordinates of the vehicle H are reversely derived using the reference position R1 of the neighboring vehicle R as a reference reference, and a third reference position H3 is obtained.

Referring to FIG. 6D, the adjacent vehicle R and the vehicle H transmit data packets by wireless signals. Therefore, the relative distance D3 between the two vehicles can be estimated according to the intensity attenuation between the wireless signals. For example, the wireless signal power emitted by the neighboring vehicle R is preset to -10 dBi, and the wireless signal power received by the vehicle H has become -30 dBi, that is, the distance between the two vehicles is shown to attenuate the wireless signal by 20 dBi because of the attenuation amplitude and The distance is proportional to and a decay relationship table can be established in advance, so that the relative distance D3 between the two vehicles can be estimated or looked up based on the attenuation amount of the 20dBi. On the other hand, since the vehicle coordinates detected by the host vehicle H and the neighboring vehicle R using the GPS receiver are known, that is, the first reference position H1 and the reference position R1 are known, and can be extended according to the direction between the two positions. The line infers the direction of the vehicle H relative to the neighboring car R. Therefore, based on the reference position R1 of the neighboring vehicle R, a fourth reference position H4 is calculated based on the relative distance D3 and the direction.

Referring to FIG. 7, after the first reference position H1 to the fourth reference position H4 are obtained, the first coverage range M1 to the fourth coverage range M4 may be determined by using the reference positions H1 to H4 as the center, each coverage area. The size of the sensor is based on the accuracy of the source sensor of the reference position. If the GPS receiver of the vehicle H has the best accuracy, the first reference position H1 has the smallest coverage and the higher accuracy. The reference position will also have a higher weight value ω, where the GPS receiver will provide its own accuracy. If other sensors are unable to provide their own accuracy, the weight value may be determined according to the distance of the position after the sensor is inferred. The farther the distance is, the lower the weight is. The present invention calculates a vehicle optimization coordinate H(x, y) in the intersection area of the first coverage range M1 to the fourth coverage area M4 (as indicated by the oblique line area).

In step b), a vehicle optimization coordinate H(x, y) is calculated according to the weight value ω of each reference position, and the calculation manner can be expressed as follows:

,among them,

In the above formula, m denotes the number of reference positions, so m=4 in the present embodiment; (x _i , y _i ) denote coordinates of the first to fourth reference positions H1 to H4, respectively; one of the weight values ω _i The calculation method can adopt Adaboost algorithm or other algorithms.

In addition to the Adaboost algorithm, a way to calculate the weight value is provided here. First, assuming that the error values of the first reference position H1 to the fourth reference position H4 are 3, 6, 4, and 5 meters, respectively, an error inverse ratio algorithm can be used to calculate four different weight values, and the calculation manner is as follows: Amount of error, Calculate the difference between the total error amount and each error value separately. Calculate the total amount of difference, The four weight values are: Vehicle optimization coordinates of the vehicle H , where (x _i , y _i ) represent the coordinates of the four reference positions, respectively; H(x, y) represents the vehicle optimized coordinates.

After the optimization of the vehicle coordinates of the vehicle H is completed, the vehicle coordinates of the neighboring vehicle R are optimized based on the neighboring vehicle R. The calculation process is the same as above, but the role between the vehicle and the adjacent vehicle is exchanged. In other words, the adjacent car data is regarded as the vehicle data in the above calculation, and the vehicle data is regarded as the adjacent car data in the above calculation. Therefore, the vehicle optimized coordinates R(x, y) representing the neighboring cars can also be obtained.

S33: Object positioning optimization, that is, the object coordinates measured by the sensor of the neighboring vehicle R and the vehicle H are optimized. In the H part of the car, since the vehicle optimized coordinate H(x, y) after the vehicle H optimization has been known, the original vehicle coordinates of the vehicle H can be compared according to the object coordinates measured by the vehicle H sensor. And the difference between the vehicle's optimized coordinates H(x, y), when calculating the object optimization coordinate of the object measured by the vehicle H sensor, that is, the difference measured by the vehicle H sensor according to the difference amount The quantity is compensated, and an object optimization coordinate is obtained to complete the optimization operation of the object coordinates measured by the vehicle H. In the same way, in the R part of the neighboring car, since the optimized coordinates of the vehicle after the optimization of the neighboring vehicle R have been known, the original vehicle coordinates of the neighboring vehicle R and its vehicles can be compared according to the coordinates of the object measured by the neighboring vehicle R sensor. Optimizing the amount of difference between the coordinates, and recalculating the object optimization coordinates of the object measured by the neighboring vehicle R sensor, that is, compensating the difference measured by the neighboring vehicle R sensor according to the difference amount, and Get an object optimization coordinate.

In step S40, the author can use the positioning comparison module 14 to fuse multiple coordinates from different neighboring cars R. For example, referring to FIG. 8, for the first vehicle 101, any other vehicle 102-104 will sense surrounding objects and share it with the first vehicle 101, and the first vehicle 101 will receive more. The coordinates of the pen about the same object (for example, the second vehicle 102) are temporarily stored in a buffer, so that the positioning comparison module 14 in the first vehicle 101 takes the coordinates of the similar object from the buffer, and They are combined into a single location data, and different object coordinates are added separately. One of the fusion methods is to calculate the average value of multiple coordinates by K-Means grouping algorithm, calculate the vehicle representative coordinates of the same vehicle on average, and calculate the object representative coordinates of the same object on average. coordinate.

In summary, the collaborative vehicle object positioning optimization method of the present invention has the following advantages:

1. Using the information exchanged between the vehicle and the neighboring car, the sensing range of each vehicle can be amplified to obtain more environmental information.

2. By re-correcting the coordinate data of the vehicle and surrounding objects through the co-location device, the creation can obtain more precise coordinates compared to the coordinate data measured by the GPS receiver alone, whether it is applied to the automatic driving system or Driving safety can be improved by warning the driver of the surrounding environment in advance.

3. Using the collaborative positioning method of this creation, after the vehicle completes the optimization calculation, the vehicle positioning data, the adjacent vehicle positioning data and one or more object positioning data after optimization can be obtained, and the three materials will be BSM data. The format of the packet is transmitted to the surrounding neighboring car. When the neighboring car receives the BSM data packet, the optimization operation of the present invention can also be performed separately. Therefore, the surrounding vehicles can be distributed and calculated, and the vehicles accumulate with each other over time. The positioning data between them will gradually improve the accuracy.

10‧‧‧Vehicle co-location device
11‧‧‧Wireless transmission interface
12‧‧‧Delay correction module
13‧‧‧Location Optimization Module
14‧‧‧ Positioning comparison module
15‧‧‧Body Sensor
21‧‧‧Part 1
22‧‧‧Part II
101‧‧‧First vehicle
102‧‧‧Second vehicle
103‧‧‧ Third vehicle
104‧‧‧fourth vehicle
200‧‧‧ objects
A1‧‧‧GPS error range
A2‧‧‧ camera error range
A3‧‧‧GPS and camera accumulator error range
H‧‧‧Car
R‧‧‧ neighboring car
H1~H4‧‧‧first reference position to fourth reference position
R1‧‧‧ reference position
M1~M4‧‧‧First Coverage~Fourth Coverage

Figure 1: Architecture block diagram of the authoring co-location device. Figure 2: Flow chart of the collaborative vehicle positioning optimization method for the author vehicle. Figure 3: Schematic diagram of the BSM data packet format of this creation. Figure 4: Schematic diagram of the implementation of time delay compensation for the original position of the vehicle. Figure 5: Flow chart of the optimization program executed by the creative location optimization module. FIG. 6A is a schematic diagram of the first reference position H1 obtained by the present creation. FIG. 6B is a schematic diagram of the first reference position H2 obtained by the present creation. FIG. 6C is a schematic diagram of the first reference position H3 obtained by the present creation. FIG. 6D is a schematic diagram of the first reference position H4 obtained by the present creation. Figure 7: This creation uses a plurality of reference locations to calculate a schematic diagram of the vehicle's optimized coordinates. Figure 8: Schematic diagram of information sharing for multiple vehicles in this creation. Figure 9: Schematic diagram of a vehicle passing through an intersection. Figure 10: Schematic diagram of the cumulative error of the contracted positioning.

H‧‧‧Car

H1~H4‧‧‧first reference position~fourth reference position

M1~M4‧‧‧First Coverage~Fourth Coverage

Claims (10)

A vehicle coordinated object positioning optimization method is implemented by using a co-location device disposed inside the vehicle, the method comprising: receiving, by the vehicle, an information packet, where the information packet includes an original coordinate of the vehicle provided by the neighboring vehicle and at least The original coordinate of an object, the original coordinates of the vehicle and the original coordinates of the object respectively have their own positioning accuracy; time delay compensation is performed on the original coordinates of the vehicle and the original coordinates of the object to obtain a vehicle coordinate and an object compensated by the adjacent vehicle respectively. Coordinates; perform an optimization procedure, which includes: comparing the vehicle coordinates of the vehicle with the vehicle coordinates of the adjacent vehicle to determine which has higher positioning accuracy; preferentially performing an optimization operation on the vehicle coordinates with higher positioning accuracy, and then Optimizing the operation of the vehicle coordinates with lower positioning accuracy, wherein the optimization operation is performed: a) calculating a plurality of reference positions according to the vehicle coordinates of the vehicle and the vehicle coordinates of the adjacent vehicle; and b) according to each reference position Weight value, calculate a vehicle optimization coordinate; optimize the object coordinates, Comparing the difference between the original vehicle coordinates of the adjacent car and the optimized coordinates of the vehicle, the object coordinates provided by the adjacent car are compensated according to the difference amount to obtain an object optimization coordinate; and the original vehicle coordinates of the vehicle and the vehicle are compared The amount of difference between the coordinates of the vehicle is optimized, and the object coordinates provided by the vehicle are compensated according to the difference amount to obtain an object optimization coordinate. The vehicle coordinated object positioning optimization method according to claim 1, after performing the optimization program, further comprises: a target pair fusion step, comparing the plurality of vehicle optimization coordinates obtained after performing the optimization program on different neighboring cars, The pen object optimizes coordinates, calculates a vehicle representative coordinate on average for the vehicle optimization coordinates of the same vehicle, and calculates an object representative coordinate on average for the object optimization coordinates of the same object. The vehicle coordinated object positioning optimization method according to claim 1 or 2, wherein the vehicle coordinates of the vehicle and the original coordinates of the vehicle of the adjacent vehicle are respectively output by a GPS receiver disposed in the vehicle and the adjacent vehicle. The vehicle coordinated object positioning optimization method according to claim 3, wherein in comparing the vehicle coordinates of the vehicle with the vehicle coordinates of the adjacent vehicle, in the step of determining which has higher positioning accuracy, comparing the vehicle and the neighbor The positioning accuracy of the car's GPS receiver. The vehicle coordinated object positioning optimization method according to claim 4, wherein, in the step of performing time delay compensation, the neighboring vehicle is calculated according to the time delay between the sending time and the receiving time of the information packet. A compensation distance moved within the time delay amount is obtained by adding the compensation distance to the original coordinate of the vehicle of the adjacent vehicle. The vehicle coordinated object positioning optimization method according to claim 5, wherein the adjacent vehicle uses the original coordinate of the vehicle output by the GPS receiver as a reference position, and the plurality of reference positions include: a first reference position, which is the vehicle a vehicle coordinate output by the GPS receiver; a second reference position, the neighboring vehicle senses a relative coordinate of the vehicle by one of the sensors, and then uses the reference position as a reference point to reverse the relative coordinate Calculating the latitude and longitude of the vehicle, obtaining the second reference position; a third reference position, the vehicle uses one of its own sensors to sense the relative distance and angle of the adjacent car, and then using the reference position as a reference point, The relative distance and angle are inversely calculated to know the latitude and longitude of the vehicle, and the third reference position is obtained; a fourth reference position is estimated based on the attenuation degree of the data packet from the neighboring vehicle to the vehicle after receiving the vehicle, and estimating the vehicle and the neighboring vehicle. Relative distance, and according to the straight line of the first reference position and the reference position, the direction of the vehicle relative to the adjacent car is obtained, according to the relative distance and the direction of the opposite direction That the latitude and longitude of the vehicle, obtains the fourth reference position. The vehicle coordinated object positioning optimization method according to claim 6, wherein the vehicle optimized coordinate system is calculated by: ,among them, m represents the number of reference positions, and (x _i , y _i ) denote coordinates of the first to fourth reference positions, respectively, and the weight value of each reference position is ω _i . The vehicle coordinated object positioning optimization method according to claim 7, wherein the data packet is a basic security message (BSM) packet. The vehicle coordinated object positioning optimization method according to claim 7, wherein the weight value of each reference position is set according to the accuracy of the sensor used when generating each reference position. A vehicle co-location device includes: a wireless transmission interface, which receives a data packet, wherein the information packet includes a vehicle original coordinate of the adjacent vehicle and at least one original object coordinate, and a positioning of the original coordinate of the vehicle and the original coordinate of the object Accuracy; a delay correction module receives the data packet sent by the neighboring car through the wireless transmission interface, and compensates the original coordinate of the vehicle and the original coordinate of the object for time delay to obtain a vehicle coordinate and an object after time delay compensation. a coordinate optimization module, the vehicle coordinate and the object coordinate of the adjacent vehicle provided by the delay correction module, and receiving the vehicle coordinates of the vehicle and the object coordinates measured by the sensor of the vehicle, the position optimization mode The group executes an optimization program, which comprises: comparing the vehicle coordinates of the vehicle with the vehicle coordinates of the adjacent vehicle to determine which has higher positioning accuracy; preferentially performing an optimization operation on the vehicle coordinates with higher positioning accuracy, Optimize the coordinates of the vehicle with lower positioning accuracy, where the The calculation is performed: a) calculating a plurality of reference positions according to the vehicle coordinates of the vehicle and the coordinates of the vehicle of the adjacent vehicle; and b) calculating a vehicle optimized coordinate according to the weight value of each reference position; optimizing the coordinates of the object, comparing The difference between the original vehicle coordinates of the adjacent vehicle and the optimized coordinates of the vehicle, the object coordinates provided by the neighboring vehicle are compensated according to the difference amount to obtain an object optimization coordinate; and the original vehicle coordinates of the vehicle and the vehicle of the vehicle are compared The amount of difference between the coordinates is optimized, and the object coordinates provided by the vehicle are compensated according to the difference amount to obtain an object optimization coordinate.