A system and method for improving positioning accuracy of Internet of Vehicles based on dead reckoning collaboration

Abstract

The present invention relates to the field of high-precision collaborative positioning technology, and specifically to a method and system for improving the positioning accuracy of a vehicle network based on dead reckoning collaboration, which can improve the GPS positioning accuracy. The method for improving the positioning accuracy of a vehicle network based on dead reckoning collaboration includes obtaining the GPS positions of a target vehicle and adjacent vehicles; correcting the GPS positions of the target vehicle and adjacent vehicles using a DR algorithm; and correcting the initial position deviation in the DR algorithm using an IPC algorithm. The system of the present invention is implemented based on the above method. The present invention corrects the initial position deviation in the DR algorithm through an IPC algorithm, and combines the IPC algorithm with the DR algorithm to effectively reduce the GPS positioning error. The IPC algorithm does not require any additional information, such as a digital city map. In an actual urban environment, the average positioning accuracy is improved by 15% using this method.

Description

Technical Field

[0001] The present invention relates to the technical field of high-precision collaborative positioning, and in particular to a system and method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration. Background Art

[0002] With the development of economy and the advancement of technology, more and more people tend to buy cars, which leads to the emergence of various traffic problems, such as traffic congestion and frequent accidents. In view of this, intelligent transportation system came into being to improve the vehicle driving and operation environment, among which vehicle driving safety is very important. In order to ensure driving safety, anti-collision warning has been widely used. Studies have shown that if drivers can know the location of other vehicles nearby in advance, they can avoid dangerous accidents.

[0003] The existing solution is that each vehicle continuously broadcasts its location through vehicle wireless communication technology to inform nearby vehicles, and the positioning accuracy of the vehicle depends largely on the Global Positioning System (GPS) device. The Global Positioning System is a space-based satellite navigation system that has been used to develop vehicle navigation. However, there are still some factors that affect the positioning accuracy of GPS devices. For example, GPS signals may be blocked by buildings or affected by the atmosphere from the troposphere and ionosphere. Generally speaking, conventional GPS can provide an accuracy of 5-30 meters in dense urban environments. Obviously, the accuracy level of GPS positioning is not enough to support safe vehicle driving. Summary of the invention

[0004] One of the purposes of the present invention is to provide a method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration, which can improve GPS positioning accuracy.

[0005] The present invention provides a method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration, comprising the following contents:

[0006] Get the GPS location of the target vehicle and adjacent vehicles;

[0007] Use DR algorithm to correct the GPS position of the target vehicle and adjacent vehicles;

[0008] The IPC algorithm is used to correct the initial position deviation in the DR algorithm.

[0009] The beneficial effects of the present invention are as follows: The technology of the present invention solves the problem that the traditional technology of GPS positioning has low precision positioning capability in vehicle safety applications such as vehicle-to-roadside and vehicle-to-vehicle methods. The initial position deviation in the DR algorithm is corrected by the IPC algorithm. The IPC algorithm and the DR algorithm are combined to effectively reduce the positioning error, and a good effect is achieved even if there are only two vehicle nodes in the VANET. In addition, the IPC algorithm does not require any additional information, such as a digital map of the city. Experimental results show that in an actual urban environment, the average positioning accuracy of this method is improved by 15%.

[0010] Furthermore, in order to overcome some shortcomings of the DR algorithm, such as error accumulation and deviation in the position of the first reference point, the present invention designs an IPC algorithm, which uses the maximum allowable error PG of the GPS position and the relative distance dv between the two vehicles to correct the position range of the target vehicle and the adjacent vehicle respectively.

[0011] Further, the IPC algorithm steps are as follows:

[0012] Step 1: Receive the GPS positions of the target vehicle and adjacent vehicles, and measure the relative distance d between the target vehicle and adjacent vehicles v and angle, obtain the ID of the nearby vehicle;

[0013] Step 2: Perform parallel correction and vertical correction on the target vehicle and the adjacent vehicles respectively to determine whether they are within the correction vector constraint. If so, use the maximum allowable error P of the GPS position. G and the relative distance d between the two vehicles v Estimate the position range of the two vehicles and calculate the true error radius;

[0014] Step 3: Determine whether the true error radius is less than P G , if so, the GPS positions of the target vehicle and adjacent vehicles are corrected.

[0015] The IPC algorithm not only improves GPS positioning, but also provides accurate initial position input for the DR algorithm.

[0016] Furthermore, the method for correcting the GPS positions of the target vehicle and adjacent vehicles by using the IPC algorithm is as follows:

[0017] The actual positions of the target vehicle and a neighboring vehicle are defined as: C1: and C2: Due to the GPS position error effect, the position of the target vehicle and a neighboring vehicle drifts to G1: and G2: The relative distance between the two cars is d v, the angle between the two cars is θ v , respectively, parallel correction and vertical correction are performed. In parallel correction, and is the projection of G1 and G2 on the longitude axis, Yes v The projection on the longitude axis is expressed as follows:

[0018]

[0019] The distance between the GPS positions of the two vehicles is d G , d G The projection on the longitude axis is P G is the maximum allowable error of GPS. Due to the effect of GPS error, the true position must be within P G In, it is expressed as follows:

[0020]

[0021] Correction vector 1 and correction vector 2 are used for correction respectively. Correction vector 1 is as follows:

[0022]

[0023] Correction vector two is as follows:

[0024]

[0025] The vertical correction is performed in the same manner, and the vertical correction and parallel correction results are combined to obtain the corrected GPS positions of the target vehicle and an adjacent vehicle.

[0026] The IPC algorithm works well with two vehicles working together, and the more cooperative nodes there are, the better the effect. The core of the IPC algorithm is to limit the range of its possible positions using only the maximum GPS error PG and the true distance dv between the two vehicles. Each vehicle Vi retrieves its GPS information, including its latitude, longitude, speed, and heading, and then begins measuring the relative distance between it and its neighbors. Finally, by sharing its GPS and measurement information with its neighbors through WAVE / DSRC communication, each vehicle Vi will continue to randomly select a candidate vehicle (such as the front vehicle, the right front vehicle, or the left front vehicle) and estimate the position using the IPC algorithm until all of its neighbors are correct.

[0027] Furthermore, in order to identify which adjacent vehicle the target vehicle is measuring, the present invention also includes identifying the ID of the adjacent vehicle, using a digital sample database to identify the adjacent vehicle ID. A digital sample database is set in advance in the server to facilitate comparison and identification of the adjacent vehicle ID.

[0028] Furthermore, the vehicle ID is shared among various vehicles to save image processing time.

[0029] Furthermore, it also includes identifying information of nearby vehicles and comparing the identification information with the list of vehicle IDs shared by other vehicles to identify the ID of the nearby vehicle. In this way, the image recognition software does not waste time identifying all the car numbers. It only needs to identify some information and then compare it with the list of car IDs broadcast by other vehicles.

[0030] A second object of the present invention is to provide a vehicle network positioning accuracy improvement system based on dead reckoning collaboration that can improve GPS positioning accuracy.

[0031] The present invention discloses a vehicle networking positioning accuracy improvement system based on dead reckoning collaboration, comprising a GPS receiving module for acquiring the GPS position of a target vehicle itself;

[0032] A communication module for obtaining the GPS position of nearby vehicles;

[0033] GPS position correction module, which uses DR algorithm to correct the GPS position of the target vehicle and adjacent vehicles;

[0034] The DR initial position correction module uses the IPC algorithm to correct the initial position deviation in the DR algorithm.

[0035] The system architecture of the present invention uses a DR algorithm to calibrate any jumping GPS signals by referring to old travel records. In this way, the CP method can filter out additional unreasonable positions.

[0036] To overcome the shortcomings of the DR algorithm, such as error accumulation and the fact that the first reference point must be correct, the system has designed an IPC algorithm to provide GPS with greater accuracy than the original GPS to overcome these shortcomings.

[0037] Furthermore, in order to save image processing time, the communication module is also used to share the vehicle ID between various vehicles.

[0038] Further, in order for the image recognition software not to waste time identifying all the car numbers, it only needs to identify some information and then compare it to the list of other vehicles broadcasting the car ID, and also includes a recognition module, the recognition module is used to identify the relative distance and angle between the adjacent vehicle and the target vehicle, and is also used to identify the ID of the adjacent vehicle;

[0039] The steps of correcting the initial position deviation in the DR algorithm using the IPC algorithm are as follows:

[0040] Step 1: Receive the GPS positions of the target vehicle and adjacent vehicles, and measure the relative distance d between the target vehicle and adjacent vehicles v and angle, obtain the ID of the nearby vehicle;

[0041] Step 2: Perform parallel correction and vertical correction on the target vehicle and the adjacent vehicles respectively to determine whether they are within the correction vector constraint. If so, use the maximum allowable error P of the GPS position. G and the relative distance d between the two vehicles v Estimate the position range of the two vehicles and calculate the true error radius;

[0042] Step 3: Determine whether the true error radius is less than P G , if so, the GPS positions of the target vehicle and adjacent vehicles are corrected.

[0043] The core of the IPC algorithm is to use only the maximum GPS error P G and the actual distance d between the two vehicles v The IPC algorithm not only improves the positioning of GPS, but also provides an accurate initial position for the DR algorithm, corrects the initial position deviation of the DR algorithm, and thus improves the positioning accuracy of GPS. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic principle diagram of a DR dead reckoning method in an embodiment of a method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration of the present invention;

[0045] Figure 2 A schematic principle diagram of a vehicle-to-roadside unit (CP) method in CP collaborative positioning of a vehicle network positioning accuracy improvement method based on dead reckoning collaboration and a system embodiment of the present invention;

[0046] Figure 3 A schematic diagram of a vehicle-to-vehicle CP method in CP collaborative positioning of a method for improving positioning accuracy of a vehicle network based on dead reckoning collaboration and a system embodiment of the present invention;

[0047] Figure 4 A flowchart of an IPC algorithm in a method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration and a system embodiment of the present invention;

[0048] Figure 5 In the embodiment of the vehicle networking positioning accuracy improvement method and system based on dead reckoning collaboration of the present invention, the GPS position ranges G1 and G2 of the two vehicles V1 and V2 are obtained respectively with the maximum allowable GPS error as PG;

[0049] Figure 6The real position range of the vehicle V2 is defined by the right movement of the vector dv in an embodiment of a method and system for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration of the present invention;

[0050] Figure 7 The real position range of the vehicle V1 is defined by the left movement of the vector dv in an embodiment of a method and system for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration of the present invention;

[0051] Figure 8 The modified position ranges of two vehicles in a method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration and a system embodiment of the present invention;

[0052] Fig. 9 The present invention is a schematic block diagram of a system for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration. DETAILED DESCRIPTION

[0053] The following is further described in detail through specific implementation methods:

[0054] This embodiment discloses a method for improving positioning accuracy of an Internet of Vehicles based on dead reckoning collaboration, including the following contents:

[0055] Get the GPS location of the target vehicle and adjacent vehicles;

[0056] Use DR algorithm to correct the GPS position of the target vehicle and adjacent vehicles;

[0057] The IPC algorithm is used to correct the initial position deviation in the DR algorithm.

[0058] Since the GPS system may be jumpy, with an error of about 5-30 meters, this method uses a DR algorithm to correct any jumpy GPS signal by referring to old travel records.

[0059] Before focusing on the method and system of the present embodiment, let's first introduce the Geographic Information System (GIS), which is a system that integrates all geographic data and the rapid development of geographic information systems in recent years. Information and communication technology is often applied to geographic information system technology to create digital maps of cities for vehicle navigation. GIS positioning is used to improve positioning accuracy. This technology is based on the use of digital maps to limit the estimated position of the vehicle relative to the road. Three different types of related map matching techniques are usually used: point curve, curve-to-curve distance, and curve-to-curve angle. The point-to-curve method matches the GPS point with the nearest road, and the curve-to-curve distance method matches the vehicle with the road based on the sum of the two end GPS points. The curve angle method matches the vehicle with the minimum angle between the two end GPS points and the road. However, map matching technology requires additional hardware costs and time to save GIS information in advance.

[0060] In the method of this embodiment, the DR algorithm, that is, DR dead reckoning, has the following principles:

[0061] Dead reckoning (DR) is a method that enables a vehicle to estimate its current position using a broken GPS signal. Using this technique, DR measurements use vehicle driving data (such as speed and direction) obtained from an inertial navigation system and a reference position to calculate its next position. Figure 1 The two-dimensional plane space in which the vehicle is traveling is shown. k Time(x k ,y k ) can be expressed as:

[0062]

[0063] Among them, (x0, y0) is the initial position of the vehicle at t0, s i and θ i are the distance traveled by the vehicle during time interval t and the displacement vector from the vehicle's position (x i ,y i ) to t i +1 (x i +1, y i This method lacks a continuous error feedback correction mechanism, that is, it cannot correct the initial position deviation.

[0064] In order to overcome some shortcomings of the DR algorithm, such as error accumulation and the deviation of the first reference point (i.e., initial position), this method further adopts the IPC algorithm to correct the initial position deviation in the DR algorithm, providing higher accuracy than the original GPS to overcome these shortcomings.

[0065] Before introducing the IPC algorithm in detail, the CP collaborative positioning technology is introduced first.

[0066] CP Co-location: In recent years, the rapid development of wireless communication technologies in vehicle networks, such as WAVE / DSRC systems,

[0067] Making many aspects of vehicle applications more powerful by sharing information, the communication protocol of the WAVE / DSRC system is the IEEE802.11p (IEEE1609) standard: using WAVE / DSRC technology to solve vehicle positioning problems, it is a method that can use wireless communication technology to improve vehicle positioning accuracy. Generally speaking, CP collaborative positioning is divided into two different methods: V2R and V2V, such as Figure 2 and Figure 3 The former uses referenced roadside units to inform vehicles of road conditions in advance. The latter uses multiple vehicles to share their driving information to improve awareness of their common locations.

[0068] The existing CP collaborative positioning usually includes three stages in the positioning process:

[0069] The first stage: the detection or measurement stage, usually designed for roadside units or on-board units (OBUs), estimates the state of the relative objects (i.e., relative distance and object ID).

[0070] Phase 2: Data sharing phase, during which vehicles communicate driving information or information on what is happening on the road to their neighbors.

[0071] The third stage: position update through data fusion, and finally use mathematical models to update the vehicle status.

[0072] V2R, or vehicle-to-roadside unit CP method, is usually implemented using differential GPS (i.e. DGPS): DGPS uses ground reference stations to determine GPS errors, compares known fixed positions with GPS positions, and broadcasts this error to help nearby vehicles determine their positions. Another method uses roadside units to provide accurate positions for vehicles. This can provide initial DR positions more accurately to help overcome DR error accumulation problems.

[0073] It can be seen that the vehicle-to-roadside unit CP method usually requires additional hardware cost to improve the position accuracy. In fact, the distribution of reference stations has a huge impact on the V2RCP method, including cost and ensuring that any vehicle must be within the coverage of a reference station in the VANET.

[0074] V2V, or vehicle-to-vehicle CP methods, can easily overcome the problem of additional hardware equipment costs, as many vehicles have the ability to measure the relative distance to each other through OBU devices such as radar and video sensors. There are currently some technologies that use the characteristics of radio to measure the relative distance of vehicles; however, due to environmental effects such as noise and multipath interference, the accuracy of radio-based ranging is insufficient. There are also technologies that use DSRC communication between two vehicles in opposite directions to measure relative speed based on Doppler shift. However, if the relative speed of the two vehicles is small, the Doppler effect may not be obvious. In addition, the analysis of Doppler shift offsets requires additional hardware for frequency analysis at the physical layer. In addition, there are also vehicle video recorders and lane-level digital maps that help overcome GPS positioning errors. When vehicles receive GPS positions, they match their positions with digital maps to identify a lane index, called a GPS lane. Due to GPS errors, their GPS lanes may also contain an error. Therefore, video images are used to determine the absolute lane index on the road, called a video lane. Each vehicle compares its GPS lane and video lane to verify the accuracy of the GPS position. Finally, each vehicle can update its own position using the verification values ​​of its neighbors and the positions predicted by other vehicles. However, in most cities, detailed lane-level digital maps are not available and the GPS error vector drifts in the vertical direction of the road.

[0075] In short, these CP collaborative positioning technologies all have some application difficulties in actual traffic environments.

[0076] In order to further overcome the problems existing in the above-mentioned CP collaborative positioning technology, an IPC algorithm is disclosed in an embodiment of the present invention. The IPC algorithm adopts the maximum allowable error P of GPS position. G and the relative distance d between the two vehicles v To correct the position range of the target vehicle and adjacent vehicles respectively.

[0077] Furthermore, as attached Figure 4 As shown, the IPC algorithm steps are as follows:

[0078] Step 1: Receive the GPS positions of the target vehicle and adjacent vehicles, and measure the relative distance d between the target vehicle and adjacent vehicles v and angle, obtain the ID of the nearby vehicle;

[0079] In this embodiment, Figure 4 As shown, if the image information is not obtained, the DR algorithm is directly used to update the GPS position.

[0080] Step 2: Perform parallel correction and vertical correction on the target vehicle and the adjacent vehicles respectively to determine whether they are within the correction vector constraint. If so, use the maximum allowable error P of the GPS position.G and the relative distance d between the two vehicles v Estimate the position range of the two vehicles and calculate the true error radius;

[0081] In this embodiment, if the correction vector is not within the restriction condition, the DR algorithm is directly used to update the GPS position.

[0082] Step 3: Determine whether the true error radius is less than P G If so, the GPS positions of the target vehicle and the adjacent vehicles are corrected, and the corrected GPS positions are used as the initial positions of the DR algorithm, that is, the initial positions of the DR algorithm are corrected. And the GPS positioning is updated based on the corrected GPS positions.

[0083] Similarly, if the true error radius is greater than or equal to P G , the DR algorithm is also used directly to update the GPS position.

[0084] The IPC algorithm can work well in the case of two vehicles cooperating, and the effect is better when more vehicle cooperation nodes are added. The core of the IPC algorithm is to use only the maximum GPS error P G and the actual distance d between the two vehicles v To limit the position range of the two vehicles. Each vehicle V i The target vehicle will retrieve its GPS information every 100ms, including its latitude, longitude, speed and heading, and then start measuring the relative distance between it and the neighboring vehicles. Finally, the target vehicle will continue to randomly select a neighboring vehicle (such as the front vehicle, the right front vehicle or the left front vehicle) through WAVE / DSRC communication, and use the vehicle network positioning accuracy improvement method based on dead reckoning collaboration disclosed in this embodiment to correct the GPS position of the neighboring vehicle until the GPS positions of the target vehicle and all neighboring vehicles meet the requirements. The meeting of the requirements in this embodiment means that the convergence speed is reduced to a specified degree after each vector correction.

[0085] Each vehicle calculates its GPS information and prepares to broadcast it to its neighboring vehicles. When the target vehicle receives the information of its neighboring vehicles through DSRC communication, it uses the webcam sensor to capture the view in front of itself and uses this image to identify the ID of the neighboring vehicles and the relative distance between them.

[0086] If it successfully recognizes this information, that is, when the target vehicle has both GPS information and relative distance, the IPC algorithm performs vertical correction and parallel correction according to latitude and longitude, respectively. Then the positioning error is corrected according to the correction vector in Table 1.

[0087] Finally, the IPC algorithm uses the position after vertical correction and parallel correction as the new positioning position. The IPC algorithm not only improves the positioning accuracy of GPS, but also provides the most accurate initial position input for the DR algorithm.

[0088] More specifically, the correction principle of the IPC algorithm is introduced as follows Figure 5 , Figure 6 , Figure 7 , Figure 8 As shown:

[0089] First, if Figure 5 As shown, assuming that the maximum GPS error in this area is P G , the two vehicles V1 and V2 obtain their respective GPS positions G1 and G2. It can be considered that V i The true position must be within the circle whose center is its GPS position and whose radius is equal to P G Therefore, if Figure 6 As shown, the measured distance d is used v Reduce the actual position range and use vector d v The right movement can determine the maximum amount of movement affected by V1 and P G Therefore, the real position of vehicle V2 is limited to Figure 6 Similarly, in Figure 7 We shift left by d v to constrain the true position of V1. Then, Figure 8 As shown, the modified positions of the two vehicles are found within the circles.

[0090] The method for correcting the GPS positions of the target vehicle and adjacent vehicles by using the IPC algorithm is as follows:

[0091] The actual positions of the target vehicle and a neighboring vehicle are defined as: C1: and C2: Due to the GPS position error effect, the position of the target vehicle and a neighboring vehicle drifts to G1: and G2: The relative distance between the two cars is d v , the angle between the two cars is θ v , respectively, parallel correction and vertical correction are performed. In parallel correction, and is the projection of G1 and G2 on the longitude axis, Yes v The projection on the longitude axis is expressed as follows:

[0092]

[0093] The distance between the GPS positions of the two vehicles is d G , can be calculated by comparing the difference between its latitude and longitude, d G The projection on the longitude axis is Directly by and The distance between the two vehicles is calculated as shown in Table 1. The IPC algorithm calculates the correction value, and it must also determine the correction direction. Table 1 details the four situations with different relationships between the real position and the GPS position. The classification criteria of these four situations are based on the GPS positions of the two vehicles and P G The parameters are then used to determine the correction direction.

[0094]

[0095] Table 1

[0096] P G is the maximum allowable error of GPS, P G The value is determined according to the vehicle's positioning capability. For example, when the positioning technology used is GPS, assuming P G The value is 20-30m. If the GPS positioning technology is converted to wide area augmentation system or differential GPS, P G The value will change to 7 or 10m.

[0097] Due to GPS error effects, the true position must be within P G In, it is expressed as follows:

[0098]

[0099] Correction vector 1 and correction vector 2 are used for correction respectively. For correction vector 1, and exist and In the range of; for correction vector 2, and Beyond and range.

[0100] When the correction vector is one, the public notice to reduce the GPS error range is as follows:

[0101]

[0102] When the correction vector is 2, the public notice to reduce the GPS error range is as follows:

[0103]

[0104] The vertical correction is performed in the same manner, and the vertical correction and parallel correction results are combined to obtain the corrected GPS positions of the target vehicle and an adjacent vehicle.

[0105] In this embodiment, the ID of the adjacent vehicle is also identified, and the digital sample database is used to identify the ID of the adjacent vehicle. In this embodiment, the vehicle ID is a license plate, and the vehicle ID is identified by image recognition technology. In order to reduce the image recognition processing time, the image can be divided into different data segments. For example, if the original image pixels are 640×480, it can be divided into small segments of 300×480.

[0106] In other embodiments, DSRC technology is also used to share vehicle IDs between vehicles, so when identifying the ID of an adjacent vehicle, only the information of the adjacent vehicle needs to be identified, and the identification information is compared with the vehicle ID list shared by other vehicles (such as adjacent vehicles) to identify the adjacent vehicle ID. For example, if a vehicle knows the license plate of a neighboring vehicle, such as 4567AB, image recognition only needs to identify some numbers, such as 4567, and it can be directly identified, making the image processing time faster.

[0107] In this embodiment, corresponding to the method, a vehicle network positioning accuracy improvement system based on dead reckoning collaboration is also provided. Fig. 9 As shown, it includes a GPS receiving module for obtaining the GPS position of the target vehicle itself, and the GPS receiving module adopts a GPS receiver;

[0108] A communication module is used to obtain the GPS position of the nearby vehicle. The communication module adopts a WAVE / DSRC communication module. The communication protocols of WAVE / DSRC are IEEE802.119 (IEEE1609) standards respectively;

[0109] GPS position correction module, which uses DR algorithm to correct the GPS position of the target vehicle and adjacent vehicles;

[0110] The DR initial position correction module uses the IPC algorithm to correct the initial position deviation in the DR algorithm.

[0111] The communication module described in this embodiment is also used to share vehicle ID and position information between various vehicles, and also includes an identification module, which is used to identify the relative distance and angle between the adjacent vehicle and the target vehicle, as well as the ID of the adjacent vehicle; the GPS position of the target vehicle, the relative distance and angle between the adjacent vehicle and the target vehicle, as well as the ID of the adjacent vehicle are data fused, and after fusion, the initial position deviation in the DR algorithm is corrected through the IPC algorithm.

[0112] In the system described in this embodiment, the steps of using the IPC algorithm to correct the initial position deviation in the DR algorithm are as follows:

[0113] Step 1: Receive the GPS positions of the target vehicle and adjacent vehicles, and measure the relative distance d between the target vehicle and adjacent vehicles v and angle, obtain the ID of the nearby vehicle;

[0114] Step 2: Perform parallel correction and vertical correction on the target vehicle and the adjacent vehicles respectively to determine whether they are within the correction vector constraint. If so, use the maximum allowable error P of the GPS position. G and the relative distance d between the two vehicles v Estimate the position range of the two vehicles and calculate the true error radius;

[0115] Step 3: Determine whether the true error radius is less than P G , if so, the GPS positions of the target vehicle and adjacent vehicles are corrected.

[0116] It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the technical concept of the present invention, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for improving positioning accuracy of Internet of Vehicles based on dead reckoning collaboration, characterized in that: It includes the following: Get the GPS location of the target vehicle and adjacent vehicles; Use DR algorithm to correct the GPS position of the target vehicle and adjacent vehicles; The IPC algorithm is used to correct the initial position deviation in the DR algorithm; The IPC algorithm uses the maximum allowable GPS position error P G and the relative distance d between the two vehicles v To correct the position range of the target vehicle and adjacent vehicles respectively; The IPC algorithm steps are as follows: Step 1: Receive the GPS positions of the target vehicle and adjacent vehicles, and measure the relative distance d between the target vehicle and adjacent vehicles v and angle, obtain the ID of the nearby vehicle; Step 2: Perform parallel correction and vertical correction on the target vehicle and the adjacent vehicles respectively to determine whether they are within the correction vector constraint. If so, use the maximum allowable error P of the GPS position. G and the relative distance d between the two vehicles v Estimate the position range of the two vehicles and calculate the true error radius; Step 3: Determine whether the true error radius is less than P G , if so, the GPS positions of the target vehicle and adjacent vehicles are corrected.

2. The method for improving positioning accuracy of Internet of Vehicles based on dead reckoning collaboration according to claim 1, characterized in that: The method for correcting the GPS positions of the target vehicle and adjacent vehicles by using the IPC algorithm is as follows: The actual positions of the target vehicle and a neighboring vehicle are defined as: C1: and C2: Due to the GPS position error effect, the position of the target vehicle and a neighboring vehicle drifts to G1: and G2: The relative distance between the two cars is d v , the angle between the two cars is θ v , respectively, parallel correction and vertical correction are performed. In parallel correction, and is the projection of G1 and G2 on the longitude axis, Yes v The projection on the longitude axis is expressed as follows: The distance between the GPS positions of the two vehicles is d G , d G The projection on the longitude axis is P G is the maximum allowable error of GPS. Due to the effect of GPS error, the true position must be within P G In, it is expressed as follows: Correction vector 1 and correction vector 2 are used for correction respectively. Correction vector 1 is as follows: Correction vector two is as follows: The vertical correction is performed in the same manner, and the vertical correction and parallel correction results are combined to obtain the corrected GPS positions of the target vehicle and an adjacent vehicle.

3. The method for improving positioning accuracy of Internet of Vehicles based on dead reckoning collaboration according to claim 1, characterized in that: It also includes identifying the ID of adjacent vehicles, using a digital sample database to identify the ID of adjacent vehicles.

4. The method for improving positioning accuracy of Internet of Vehicles based on dead reckoning collaboration according to claim 1 or 3, characterized in that: It also includes sharing the vehicle ID between vehicles.

5. The method for improving positioning accuracy of Internet of Vehicles based on dead reckoning collaboration according to claim 4, characterized in that: It also includes identifying information of adjacent vehicles and comparing the identification information with a vehicle ID list shared by other vehicles to identify the adjacent vehicle ID.

6. A vehicle network positioning accuracy improvement system based on dead reckoning collaboration, characterized in that: It includes a GPS receiving module for obtaining the GPS position of the target vehicle itself; A communication module for obtaining the GPS position of nearby vehicles; GPS position correction module, which uses DR algorithm to correct the GPS position of the target vehicle and adjacent vehicles; DR initial position correction module, which uses IPC algorithm to correct the initial position deviation in DR algorithm; An identification module, the identification module is used to identify the relative distance and angle between the adjacent vehicle and the target vehicle, and is also used to identify the ID of the adjacent vehicle; The steps of correcting the initial position deviation in the DR algorithm using the IPC algorithm are as follows: Step 1: Receive the GPS positions of the target vehicle and adjacent vehicles, and measure the relative distance d between the target vehicle and adjacent vehicles v and angle, obtain the ID of the nearby vehicle; Step 2: Perform parallel correction and vertical correction on the target vehicle and the adjacent vehicles respectively to determine whether they are within the correction vector constraint. If so, use the maximum allowable error P of the GPS position. G and the relative distance d between the two vehicles v Estimate the position range of the two vehicles and calculate the true error radius; Step 3: Determine whether the true error radius is less than P G , if so, the GPS positions of the target vehicle and adjacent vehicles are corrected.

7. The vehicle network positioning accuracy improvement system based on dead reckoning collaboration according to claim 6 is characterized in that: The communication module is also used to share the vehicle ID between various vehicles.

Images