Fusion positioning method, system and device and computer readable storage medium
Through real-time differential positioning data, inertial measurement unit data, odometer data and camera perception data, air safety calculations are carried out, and local target maps are generated and high-precision maps are updated, which solves the problem of poor positioning accuracy caused by map freshness and achieves more accurate Vehicle positioning.
Patent Information
- Application Number
- CN202411544670.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-31
- Publication Date
- 2025-02-11
AI Technical Summary
In the prior art, the poor positioning accuracy caused by map freshness problems affects the vehicle's driving function.
By obtaining real-time differential positioning data, inertial measurement unit data, odometer data and camera perception data, air safety calculations are carried out, target local maps are generated, and high-precision maps are updated based on the map to form a fusion map to achieve vehicle positioning.
Improve positioning accuracy, accurately reflect the latest conditions of the road, solve the problem of freshness of high-precision maps, and ensure that vehicles will not lose their direction or go astray due to road changes.
Smart Images
Figure CN119413159A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle positioning technology, and in particular to a fusion positioning method, system, device and computer-readable storage medium. Background Art
[0002] Map freshness refers to the update speed and real-time nature of map information. In vehicle positioning, maps with high freshness can reflect the latest road conditions (such as newly built roads, closed intersections, temporary traffic controls, etc.) in real time. These real-time updated road information is crucial for vehicle positioning because it can improve positioning accuracy to ensure that the vehicle does not lose its way or go astray due to road changes.
[0003] In related technologies, road repairs and other events often occur on highways and urban expressways. However, due to the high production cost and large volume of high-precision maps HDMAP, updates are often released quarterly, resulting in map freshness problems in high-precision maps, which makes it impossible to ensure positioning accuracy, thus affecting driving functions. It can be seen that how to solve the map freshness problem to improve positioning accuracy is an urgent problem to be solved. Summary of the invention
[0004] The present application provides a fusion positioning method, system, device and computer-readable storage medium, which can solve the technical problem of poor positioning accuracy caused by the map freshness problem in the prior art.
[0005] In a first aspect, an embodiment of the present application provides a fusion positioning method, the fusion positioning method comprising:
[0006] Acquire real-time differential positioning data, inertial measurement unit data, odometer data, and camera perception data; perform satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data, and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude;
[0007] Converting the camera sensing data into a target position sensing result in an earth coordinate system according to the satellite dead reckoning position and satellite dead reckoning attitude;
[0008] Generate a local map of the target through the target position perception result;
[0009] The high-precision map is updated based on the target local map to obtain a fused map, so as to realize vehicle positioning through the fused map.
[0010] In combination with the first aspect, in one implementation, performing satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data, and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude includes:
[0011] When the solution state corresponding to the real-time differential positioning data is a fixed solution, the position and attitude angle in the real-time differential positioning data are used as the satellite dead reckoning position and the satellite dead reckoning attitude respectively;
[0012] When the solution state corresponding to the real-time differential positioning data is a non-fixed solution, satellite dead reckoning is performed based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude.
[0013] In combination with the first aspect, in one implementation, performing satellite dead reckoning based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data, and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude includes:
[0014] Calculating the dead reckoning position based on the attitude angle and position in the real-time differential positioning data, the vehicle speed in the odometer data, and the inter-frame time difference in the inertial measurement unit data;
[0015] The satellite dead reckoning attitude is calculated based on the attitude angle in the real-time differential positioning data, the angular velocity in the inertial measurement unit data and the time difference between frames.
[0016] In combination with the first aspect, in one implementation, converting the camera perception data into a target position perception result in an earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude includes:
[0017] Convert the satellite dead reckoning attitude into an attitude conversion matrix from the vehicle coordinate system to the earth coordinate system;
[0018] The camera perception data is transformed based on the attitude transformation matrix and the satellite estimated position to obtain a target position perception result in an earth coordinate system.
[0019] In combination with the first aspect, in one implementation, updating the high-precision map based on the target local map to obtain a fused map includes:
[0020] When it is determined according to the target local map that the lane line has been linearly modified, the lane line in the high-precision map is updated based on the target local map to obtain a fused map;
[0021] When a roadblock is detected according to the target local map, the high-precision map is updated using the target local map and the roadblock information to obtain a fused map.
[0022] In combination with the first aspect, in one implementation, the updating of the high-precision map based on the target local map to obtain a fused map further includes:
[0023] When the current road section is determined to be a road section where road repair is completed according to the target local map, determining whether the current road section information memorized in the target local map is consistent with the current road section information memorized in the high-precision map;
[0024] If they are consistent, the high-precision map will be used as the fused map;
[0025] If there is inconsistency, the inconsistency information is sent to the user, so that the user can determine whether a map update is required based on the inconsistency information;
[0026] If necessary, the high-precision map is updated based on the target local map to obtain a fused map;
[0027] If not necessary, the current road section is marked as a blacklist road section in the high-precision map to generate a fused map.
[0028] In combination with the first aspect, in one implementation, the vehicle positioning is realized by using the fused map, including:
[0029] Acquire the target vehicle navigation data at the current moment, and fuse the target vehicle navigation data with the fused map to obtain the target driving route;
[0030] Performing road matching on the target driving route according to target data corresponding to the current moment to determine the target road, wherein the target data includes target real-time differential positioning data, target inertial measurement unit data, target odometer data, and target camera perception data;
[0031] generating a new target local map based on the target data;
[0032] Performing lane matching on the target road based on the new target local map and the fused map to output first position information;
[0033] The target positioning information is outputted through the first position information, the target satellite dead reckoning attitude and the target camera perception data, and the target satellite dead reckoning attitude is determined by the target data.
[0034] In a second aspect, an embodiment of the present application provides a fusion positioning system, the fusion positioning system comprising:
[0035] A data acquisition module, which is used to acquire real-time differential positioning data, inertial measurement unit data, odometer data, and camera perception data;
[0036] A data solving module, which is used to perform satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude; and convert the camera perception data into a target position perception result in an earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude;
[0037] A map generation module, which is used to generate a local map of the target based on the target position perception result; update the high-precision map based on the local map of the target to obtain a fused map;
[0038] A positioning and solving module is used to realize vehicle positioning through the fused map.
[0039] In conjunction with the second aspect, in one implementation, the data solving module is specifically used to:
[0040] When the solution state corresponding to the real-time differential positioning data is a fixed solution, the position and attitude angle in the real-time differential positioning data are used as the satellite dead reckoning position and the satellite dead reckoning attitude respectively;
[0041] When the solution state corresponding to the real-time differential positioning data is a non-fixed solution, satellite dead reckoning is performed based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude.
[0042] In conjunction with the second aspect, in one implementation, the data solving module is further used to:
[0043] Calculating the dead reckoning position based on the attitude angle and position in the real-time differential positioning data, the vehicle speed in the odometer data, and the inter-frame time difference in the inertial measurement unit data;
[0044] The satellite dead reckoning attitude is calculated based on the attitude angle in the real-time differential positioning data, the angular velocity in the inertial measurement unit data and the time difference between frames.
[0045] In conjunction with the second aspect, in one implementation, the data solving module is further used to:
[0046] Convert the satellite dead reckoning attitude into an attitude conversion matrix from the vehicle coordinate system to the earth coordinate system;
[0047] The camera perception data is transformed based on the attitude transformation matrix and the satellite estimated position to obtain a target position perception result in an earth coordinate system.
[0048] In conjunction with the second aspect, in one implementation, the map generation module is specifically used to:
[0049] When it is determined according to the target local map that the lane line has been linearly modified, the lane line in the high-precision map is updated based on the target local map to obtain a fused map;
[0050] When a roadblock is detected according to the target local map, the high-precision map is updated using the target local map and the roadblock information to obtain a fused map.
[0051] In conjunction with the second aspect, in one implementation, the map generation module is further configured to:
[0052] When the current road section is determined to be a road section where road repair is completed according to the target local map, determining whether the current road section information memorized in the target local map is consistent with the current road section information memorized in the high-precision map;
[0053] If they are consistent, the high-precision map will be used as the fused map;
[0054] If there is inconsistency, the inconsistency information is sent to the user, so that the user can determine whether a map update is required based on the inconsistency information;
[0055] If necessary, the high-precision map is updated based on the target local map to obtain a fused map;
[0056] If not necessary, the current road section is marked as a blacklist road section in the high-precision map to generate a fused map.
[0057] In conjunction with the second aspect, in one implementation, the positioning solution module is specifically used to:
[0058] Acquire the target vehicle navigation data at the current moment, and fuse the target vehicle navigation data with the fused map to obtain the target driving route;
[0059] Performing road matching on the target driving route according to target data corresponding to the current moment to determine the target road, wherein the target data includes target real-time differential positioning data, target inertial measurement unit data, target odometer data, and target camera perception data;
[0060] generating a new target local map based on the target data;
[0061] Performing lane matching on the target road based on the new target local map and the fused map to output first position information;
[0062] The target positioning information is outputted through the first position information, the target satellite dead reckoning attitude and the target camera perception data, and the target satellite dead reckoning attitude is determined by the target data.
[0063] In a third aspect, an embodiment of the present application provides a fusion positioning device, which includes a processor, a memory, and a fusion positioning program stored in the memory and executable by the processor, wherein when the fusion positioning program is executed by the processor, the steps of the aforementioned fusion positioning method are implemented.
[0064] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a fusion positioning program is stored, wherein when the fusion positioning program is executed by a processor, the steps of the aforementioned fusion positioning method are implemented.
[0065] The beneficial effects brought by the technical solution provided by the embodiments of the present application include:
[0066] The satellite dead reckoning position and dead reckoning attitude are obtained by performing satellite dead reckoning with real-time differential positioning data, inertial measurement unit data and odometer data, and the camera perception data is converted into the target position perception result in the earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude. The target local map is then generated based on the target position perception result, and the known high-precision map is updated based on the target local map to obtain a fused map, that is, the perceived road information is matched with the high-precision map, so that the fused map can accurately reflect the latest road conditions, and then solve the freshness problem of the high-precision map. Finally, the fused map is used for fusion positioning to output accurate positioning information, thereby effectively improving the positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0067] Figure 1 This is a flow chart of an embodiment of the fusion positioning method of the present application;
[0068] Figure 2 A schematic diagram of a fusion positioning framework involved in the embodiment of the present application;
[0069] Figure 3 For this application Figure 1 Detailed flow chart of step S30;
[0070] Figure 4 This is a schematic diagram of the functional modules of an embodiment of the fusion positioning system of the present application;
[0071] Figure 5 This is a schematic diagram of the hardware structure of the fusion positioning device involved in the embodiment of the present application. DETAILED DESCRIPTION
[0072] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0073] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below in conjunction with the accompanying drawings.
[0074] In a first aspect, an embodiment of the present application provides a fusion positioning method.
[0075] In one embodiment, referring to Figure 1 , Figure 1 This is a flow chart of an embodiment of the fusion positioning method of the present application.
[0076] like Figure 1 As shown, the fusion positioning method includes:
[0077] Step S10: Acquire real-time differential positioning data, inertial measurement unit data, odometer data, and camera perception data.
[0078] For example, it should be noted that real-time differential positioning data refers to RTK (Real-time Kinematic, real-time dynamic carrier phase differential technology) data, which includes but is not limited to timestamp, latitude and longitude, latitude and longitude covariance, three-axis speed, three-axis speed covariance, RTK status, number of satellite searches and signal-to-noise ratio, and can be stored as a variable rtk(t), rtk(t) represents the RTK data at time t; inertial measurement unit data refers to IMU (Inertial Measurement Unit) data. Unit, inertial measurement unit) data, which includes but is not limited to timestamp, three-axis angular velocity and three-axis acceleration, which can be stored as variable imu(t), imu(t) represents the data of IMU at the tth moment; odometer data refers to ODOM data, which includes but is not limited to timestamp, four-wheel speed, comprehensive vehicle speed, steering wheel angle and gear position, which can be stored as variable vcu(t), vcu(t) represents the data of VCU at the tth moment; camera perception data refers to CAMERA perception data (i.e., perception result) in the vehicle body coordinate system, which includes but is not limited to timestamp, boundary information, lane information and road signs, which can be stored as variable cam(t), cam(t) represents the perception data output at the tth moment.
[0079] In order to improve the positioning accuracy of subsequent calculations, the RTK data and CAMERA perception data will be checked for validity to determine whether the corresponding data can be used for subsequent positioning processing; therefore, the RTK data and CAMERA perception data obtained in this embodiment are all verified data, that is, see Figure 2 As shown, the IMU data, ODOM data, and verified RTK data and CAMERA perception data received from the outside are used as system inputs.
[0080] Among them, for the received external RTK data, first check whether the timestamp in the RTK data has a fallback. If so, discard the current RTK data. If not, continue to check whether the longitude and latitude of the RTK data exceed their corresponding longitude and latitude thresholds; if neither exceeds, retain the current RTK data, otherwise discard the current RTK data. For CAMERA perception data, first check whether the timestamp in the CAMERA perception data has a fallback. If so, discard the current CAMERA perception data. If not, continue to check whether the number of lane lines in the CAMERA perception data is not less than the preset number (for example, 2) or the total length of the lane lines is not less than the preset length (for example, 4 meters); if so, retain the current number of CAMERA perception data, and if the number of lane lines is less than 2 and the total length of the lane lines is less than 4 meters, then the current CAMERA perception data is considered invalid and the current CAMERA perception data is discarded.
[0081] Step S20: performing satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude.
[0082] Exemplarily, in this embodiment, satellite dead reckoning is performed using the attitude angle information and position information in the RTK data, the IMU inter-frame time difference information and angular velocity information in the IMU data, and the vehicle speed information in the ODOM data, and then the absolute pose information including the satellite-calculated position and the satellite-calculated pose is calculated to provide a data basis for the subsequent conversion of the CAMERA perception data from the vehicle coordinate system to the earth coordinate system; wherein the absolute pose information refers to the pose information in the earth coordinate system.
[0083] Further, in one embodiment, performing satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data, and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude includes:
[0084] When the solution state corresponding to the real-time differential positioning data is a fixed solution, the position and attitude angle in the real-time differential positioning data are used as the satellite dead reckoning position and the satellite dead reckoning attitude respectively;
[0085] When the solution state corresponding to the real-time differential positioning data is a non-fixed solution, satellite dead reckoning is performed based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude.
[0086] Exemplarily, in this embodiment, for dead reckoning, the data required can be determined according to the solution state corresponding to the RTK data, that is, if the solution state corresponding to the RTK data is a fixed solution, the attitude angle and position information in the RTK data are directly reused as the attitude angle and position of dead reckoning, that is, the position in the RTK data is directly used as the dead reckoning position, and the attitude angle in the RTK data is used as the dead reckoning attitude; however, if the solution state corresponding to the RTK data is a non-fixed solution, the attitude angle and position in the RTK data, the IMU inter-frame time difference information and angular velocity information in the IMU data, and the vehicle speed information in the ODOM data are used as calculation parameters of dead reckoning, so as to calculate the dead reckoning position and the dead reckoning attitude.
[0087] Further, in one embodiment, performing satellite dead reckoning based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data, and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude includes:
[0088] Calculating the dead reckoning position based on the attitude angle and position in the real-time differential positioning data, the vehicle speed in the odometer data, and the inter-frame time difference in the inertial measurement unit data;
[0089] The satellite dead reckoning attitude is calculated based on the attitude angle in the real-time differential positioning data, the angular velocity in the inertial measurement unit data and the time difference between frames.
[0090] Exemplarily, in this embodiment, when the solution state corresponding to the RTK data is a non-fixed solution, firstly based on the position P in the RTK data rtk and attitude angle attt rtk Initialize the satellite dead reckoning position and satellite dead reckoning attitude, that is, the position P rtk Assign the value to the satellite dead reckoning position to get the initialized satellite dead reckoning position P dr (0), and set the attitude angle attt rtk Assign the value to the satellite dead reckoning attitude to obtain the initialized satellite dead reckoning attitude att(0); then calculate the satellite dead reckoning position and satellite dead reckoning attitude through the IMU frame time difference information, angular velocity information, and vehicle speed information. The specific calculation process is shown in the following formula:
[0091]
[0092] Where Pdr (t) represents the dead-reckoning position at time t; att(t) represents the dead-reckoning attitude at time t; odom s (t) represents the vehicle speed information at time t; dalte_T represents the time difference between IMU frames; imu g (t) represents the angular velocity information at the tth moment in the IMU.
[0093] Step S30: converting the camera perception data into a target position perception result in the earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude.
[0094] For example, see Figure 2 As shown, this embodiment converts local perception into perception in absolute position based on the positioning result calculated by satellite, that is, the vehicle coordinate system posture output by CAMERA perception is converted to the earth coordinate system using the posture information in the earth coordinate system, that is, the CAMERA perception data in the vehicle coordinate system is converted into the target position perception result in the earth coordinate system based on the satellite calculated position and the satellite calculated posture.
[0095] Further, see Figure 3 As shown, the converting of the camera perception data into the target position perception result in the earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude includes:
[0096] Step S301: converting the satellite dead reckoning posture into a posture conversion matrix from the vehicle coordinate system to the earth coordinate system;
[0097] Step S302: transform the camera perception data based on the attitude transformation matrix and the satellite dead reckoning position to obtain a target position perception result in an earth coordinate system.
[0098] For example, in this embodiment, the satellite dead reckoning attitude att(t) in the earth coordinate system is first converted into an attitude conversion matrix from the vehicle coordinate system to the earth coordinate system: Right now Among them, tran() means converting the attitude angle into attitude moment; then based on the attitude transformation matrix CAMERA perception data in vehicle coordinate system and P dr (t) can get the target position perception result in the earth coordinate system Right now
[0099] Step S40: Generate a target local map based on the target position perception result.
[0100] For example, see Figure 2As shown, this embodiment will generate a local map of the target by sensing and splicing the target position sensing results. Specifically, it can be understood that the CAMERA sensing in this embodiment uses a distance method to track, so the target position sensing results can be used. Perform a sliding window cache of N frames of perception data, where N is a positive integer; if the number of perception data stored in the sliding window exceeds N, remove the outliers in the perception data to obtain the perception data after removal; then use mean filtering to splice the perception data in the sliding window to generate a target local map (i.e., a perception splicing map).
[0101] Step S50: updating the high-precision map based on the target local map to obtain a fused map, so as to realize vehicle positioning through the fused map.
[0102] For example, it can be understood that this embodiment mainly solves the map freshness problem existing in the high-precision map. Therefore, this embodiment will solve the exit problem of the positioning function due to the freshness problem by memorizing the latest local map data and the fused map data, that is, combining the prior map (i.e., high-precision map) and perception information to achieve fusion positioning to improve positioning accuracy; therefore, see Figure 2 As shown, this embodiment will update and memorize the known high-precision map based on the target local map, forming a fused map memorized by the vehicle side (i.e., the vehicle-side memory map) to complete the map verification, that is, use the perceived road information to match the high-precision map, so that the fused map can accurately reflect the latest road conditions, and then solve the freshness problem of the high-precision map. Therefore, by performing fusion positioning through the fused map, the accurate positioning information at the current moment can be output to complete the position output, thereby effectively improving the positioning accuracy.
[0103] Furthermore, in one embodiment, the updating of the high-precision map based on the target local map to obtain a fused map includes:
[0104] When it is determined according to the target local map that the lane line has been linearly modified, the lane line in the high-precision map is updated based on the target local map to obtain a fused map;
[0105] When a roadblock is detected according to the target local map, the high-precision map is updated using the target local map and the roadblock information to obtain a fused map.
[0106] Exemplarily, in this embodiment, map freshness issues are classified based on the positioning function abnormality scenario, and different strategies are adopted for map verification and update. Among them, if the positioning function abnormality scenario is the linear modification of the lane line, the positioning function will often cause the solid line and dotted line not changing lanes due to the map freshness problem (that is, the solid line is changed to a dotted line in practice but the map is not updated to a dotted line in time; or the dotted line is changed to a solid line in practice but the map is not updated to a solid line in time); based on this, this embodiment will determine whether the map needs to be updated and memorized based on the credibility of the lane line output by the target local map where the vehicle is located; specifically, assuming that the lane line output by the vehicle in the target local map is different from the lane line in the high-precision map where the vehicle is located, if the target The credibility of the lane line where the vehicle is located output by the target local map is very high (that is, the credibility is greater than the preset credibility threshold), which means that the lane line has been linearly modified, but the high-precision map has not been updated in time, that is, the high-precision map has a freshness problem; and because the update of the lane line is an update with a relatively small impact, in order not to add too much decision-making burden to the driver, the judgment on whether map memory is needed is not pushed to the customer interface, but the need for map data memory is directly determined, that is, the lane line in the high-precision map is updated through the target local map to obtain the fused map, and the fused map is memorized.
[0107] If the positioning function is abnormal when a road is under construction, in order not to add too much decision-making burden to the driver, the map will be updated directly by detecting the set roadblocks, that is, by matching the target local map and the roadblock information in the high-precision map to determine whether the map data has freshness problems. Specifically, when driving to a certain road section, if the target local map detects that there is a roadblock in the road section, a line segment fitting is performed according to the perceived coordinate points of the roadblock to obtain a temporary boundary; then the target lane in the road repair state is determined according to the location of the temporary boundary. For example, assuming that the current road section has three lanes (i.e., lanes 1 to 3 from left to right) and the vehicle is in the middle lane (i.e., lane 2), if the roadblock is located in lane 2 and on the left side of the vehicle, the temporary boundary is located in lane 2, then the lane where the temporary boundary is located (i.e., lane 2) and the lane on its left side (i.e., lane 1) are determined as the target lanes; it can be seen that, except for the lane where the vehicle is located, the positional relationship between the other target lanes and the vehicle is the same as the positional relationship between the roadblock and the vehicle, that is, if the roadblock is located on the left side of the vehicle, the target lane is also located on the left side of the vehicle, and if the roadblock is located on the right side of the vehicle, the target lane is also located on the right side of the vehicle; then the target lane is added to the lane blacklist to form new lane information for updating and memorizing the high-precision map data, and then a fused map is obtained. It should be noted that if there is a need to use the road due to road construction, only the route will be recorded and the map data will be memorized, but the high-precision map will not be updated or modified. The positioning function can pass this section based on tracking.
[0108] Furthermore, in one embodiment, the updating of the high-precision map based on the target local map to obtain a fused map further includes:
[0109] When the current road section is determined to be a road section where road repair is completed according to the target local map, determining whether the current road section information memorized in the target local map is consistent with the current road section information memorized in the high-precision map;
[0110] If they are consistent, the high-precision map will be used as the fused map;
[0111] If there is inconsistency, the inconsistency information is sent to the user, so that the user can determine whether a map update is required based on the inconsistency information;
[0112] If necessary, the high-precision map is updated based on the target local map to obtain a fused map;
[0113] If not necessary, the current road section is marked as a blacklist road section in the high-precision map to generate a fused map.
[0114] Exemplarily, in this embodiment, if the positioning function abnormal scenario is a completed road construction section, once the high-precision map has a freshness problem, it will cause the positioning function to miss the ramp or exit the function due to ramp diversion, road widening, and additional lanes; based on this, this embodiment will detect whether the road section where the vehicle is currently located has been set up with a roadblock based on the historical roadblock information stored in the target local map, or determine whether the current road section has been completed based on whether there is information on manual driving of the road section after the positioning function abnormally exits; if a roadblock has been set up before or the vehicle has been manually controlled to drive after the positioning function abnormally exits, the current road section is determined to be a road section where road construction has been completed, and the target local map and the high-precision map will be further compared and verified for the section information such as the number of lanes, lane width, and ramp coordinate points. If the two are the same, the map will not be updated or memorized, that is, the high-precision map will be directly used as the fused map.
[0115] However, if the two are different, since the update of the road section information such as the number of lanes, lane width, ramp coordinate points, etc. is an update with a greater impact, it is necessary to push the inconsistent information to the user so that the user can decide whether to update and memorize the map. If the user decides that the map needs to be updated and memorized, this embodiment will update and memorize the high-precision map according to the target local map to obtain a fused map to solve the freshness problem of the high-precision map; and if the user decides that the map does not need to be updated and memorized, this embodiment will add the current road section to the blacklist and memorize it in the high-precision map to generate a fused map, so that the road section will not become a navigation section later, so as to avoid affecting the driving function due to the freshness problem of the high-precision map, and remind the user to take over the vehicle. It can be seen that this embodiment pushes the road section with map freshness problem to the driver through the central control screen so that the driver can be informed of the road conditions in time.
[0116] Furthermore, in one embodiment, the performing fusion positioning through the fused map includes:
[0117] Acquire the target vehicle navigation data at the current moment, and fuse the target vehicle navigation data with the fused map to obtain the target driving route;
[0118] Performing road matching on the target driving route according to target data corresponding to the current moment to determine the target road, wherein the target data includes target real-time differential positioning data, target inertial measurement unit data, target odometer data, and target camera perception data;
[0119] generating a new target local map based on the target data;
[0120] Performing lane matching on the target road based on the new target local map and the fused map to output first position information;
[0121] The target positioning information is outputted through the first position information, the target satellite dead reckoning attitude and the target camera perception data, and the target satellite dead reckoning attitude is determined by the target data.
[0122] Exemplarily, it should be noted that the target data refers to the newly generated data corresponding to the current moment of the vehicle, which includes the target real-time differential positioning data, the target inertial measurement unit data, the target odometer data and the target camera perception data; in this embodiment, the vehicle navigation data will also be obtained, which includes but is not limited to the timestamp, the lane id set of the navigation route within a few kilometers, and can be stored as the variable nav(t), nav(t) represents the navigation data output by the vehicle navigation at the tth moment; wherein, in order to improve the positioning accuracy, the vehicle navigation data will be verified to determine whether the vehicle navigation data can be used for subsequent positioning processing, that is, the vehicle navigation data that passes the verification will also be used as the original input of the system; specifically, for the received external vehicle navigation data, first check whether there is a fallback in the navigation frame time stamp, if so, discard the current vehicle navigation data; and if not, continue to check whether there is a 0 value in the lane id set, if so, discard the current data, otherwise retain the current vehicle navigation data.
[0123] Then, the navigation data of the target vehicle at the current moment output by the navigation is fused with the fused map to solve the route communication between the two. This route is the route of the intelligent driving vehicle (i.e., the target driving route), which can be recorded as path(t), and this target driving route path(t) is used as the input information for road matching; when performing road matching, if the solution state corresponding to the received target RTK data at the current moment is a fixed solution, the longitude and latitude positioning of the RTK is directly projected to the target driving route path(t) to obtain the road road(t) where the vehicle is currently located.
[0124] If the solution state corresponding to the received target RTK data is a non-fixed solution, a key point is selected at 1 meter or 5° for sliding window caching. At this time, road matching is divided into two cases: (1) before the key point does not exceed N points (the specific value is determined according to actual needs and is not limited here); (2) after the cache exceeds N points.
[0125] For the first case, the accuracy of the satellite dead reckoning (high short-term accuracy) result can usually meet the requirements. Therefore, the latest satellite dead reckoning positioning point projection can be used for map projection to solve the road (t) where the vehicle is currently located. Specifically, the position P in the target RTK data is used to calculate the road (t). rtk and attitude angle att rtk, the IMU frame time difference information and angular velocity information in the target IMU data, and the vehicle speed information in the target ODOM data are used to calculate the target satellite dead reckoning position and the target satellite dead reckoning attitude. It should be noted that the specific solution process and principle of the target satellite dead reckoning position and the target satellite dead reckoning attitude can refer to the solution process and principle of the aforementioned satellite dead reckoning position and the satellite dead reckoning attitude, which will not be repeated here. Then, road matching is performed according to the positioning result of the dead reckoning to obtain the matching road road(t).
[0126] For the second case, the accuracy of the satellite dead reckoning (long-term position divergence) result cannot meet the requirements. Therefore, the N points in the solution cache can be rotated to minimize the lateral distance to the target driving route path(t), so as to solve the projection of the last position point of the sliding window after rotation to the fused map, and then solve the current road road(t) of the vehicle.
[0127] Then, a new target local map is generated through the target RTK data, target IMU data, target ODOM data and target CAMERA. It should be noted that the generation process and principle of the new target local map can refer to the generation process and principle of the aforementioned target local map, which will not be repeated here; then the new target local map and the solved road road(t) are matched with each other for lane matching to determine the target lane where the vehicle is currently located; and the lane line type and color of the target lane memorized in the new target local map and the fused map are matched to perform position compensation based on the matching result, and then the final position information (i.e., the first position information) is output, and the lane matching is completed.
[0128] Finally, the target satellite dead reckoning posture is converted into the target posture conversion matrix from the vehicle coordinate system to the earth coordinate system, and the target satellite dead reckoning position is replaced with the first position information. That is, according to the first position information, the target posture conversion matrix and the target camera perception data, the final position perception result in the earth coordinate system corresponding to the current moment can be accurately calculated to complete the vehicle positioning.
[0129] Visible, see Figure 2As shown, when implementing map update and fusion positioning, this embodiment needs to first perform primary processing of road matching, absolute perception and navigation and satellite estimation on the original data. It can be understood that the purpose of the primary processing is to process the original data into data that can be used for secondary processing; then the primary processing result is subjected to secondary processing of lane matching, perception splicing and map verification, and finally positioning output is performed based on the secondary processing result to compensate for the influence of map freshness on fusion positioning; wherein, this embodiment will output fused map data and location information to the downstream in positioning mode. It should be noted that if there is no vehicle-side memory map, the high-precision map is directly output, and if there is a vehicle-side memory map, the vehicle-side memory map is output; in addition, for the vehicle controller, whether navigation is based on lane memory, that is, whether navigation is performed through the vehicle-side memory map, will be determined according to the flag used to characterize whether the memory map output is adopted; in addition, if it is determined based on the verification that the original data is abnormal, the corresponding fault information will also be output.
[0130] In a second aspect, an embodiment of the present application also provides a fusion positioning system.
[0131] In one embodiment, referring to Figure 4 , Figure 4 This is a functional module diagram of an embodiment of the fusion positioning system of the present application. Figure 4 As shown, the fusion positioning system includes:
[0132] A data acquisition module, which is used to acquire real-time differential positioning data, inertial measurement unit data, odometer data, and camera perception data;
[0133] A data solving module, which is used to perform satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude; and convert the camera perception data into a target position perception result in an earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude;
[0134] A map generation module, which is used to generate a local map of the target based on the target position perception result; update the high-precision map based on the local map of the target to obtain a fused map;
[0135] A positioning and solving module is used to realize vehicle positioning through the fused map.
[0136] Furthermore, in one embodiment, the data solving module is specifically used for:
[0137] When the solution state corresponding to the real-time differential positioning data is a fixed solution, the position and attitude angle in the real-time differential positioning data are used as the satellite dead reckoning position and the satellite dead reckoning attitude respectively;
[0138] When the solution state corresponding to the real-time differential positioning data is a non-fixed solution, satellite dead reckoning is performed based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude.
[0139] Furthermore, in one embodiment, the data solving module is further used for:
[0140] Calculating the dead reckoning position based on the attitude angle and position in the real-time differential positioning data, the vehicle speed in the odometer data, and the inter-frame time difference in the inertial measurement unit data;
[0141] The satellite dead reckoning attitude is calculated based on the attitude angle in the real-time differential positioning data, the angular velocity in the inertial measurement unit data and the time difference between frames.
[0142] Furthermore, in one embodiment, the data solving module is further used for:
[0143] Convert the satellite dead reckoning attitude into an attitude conversion matrix from the vehicle coordinate system to the earth coordinate system;
[0144] The camera perception data is transformed based on the attitude transformation matrix and the satellite estimated position to obtain a target position perception result in an earth coordinate system.
[0145] Furthermore, in one embodiment, the map generation module is specifically used for:
[0146] When it is determined according to the target local map that the lane line has been linearly modified, the lane line in the high-precision map is updated based on the target local map to obtain a fused map;
[0147] When a roadblock is detected according to the target local map, the high-precision map is updated using the target local map and the roadblock information to obtain a fused map.
[0148] Furthermore, in one embodiment, the map generation module is further configured to:
[0149] When the current road section is determined to be a road section where road repair is completed according to the target local map, determining whether the current road section information memorized in the target local map is consistent with the current road section information memorized in the high-precision map;
[0150] If they are consistent, the high-precision map will be used as the fused map;
[0151] If there is inconsistency, the inconsistency information is sent to the user, so that the user can determine whether a map update is required based on the inconsistency information;
[0152] If necessary, the high-precision map is updated based on the target local map to obtain a fused map;
[0153] If not necessary, the current road section is marked as a blacklist road section in the high-precision map to generate a fused map.
[0154] Furthermore, in one embodiment, the positioning solution module is specifically used for:
[0155] Acquire the target vehicle navigation data at the current moment, and fuse the target vehicle navigation data with the fused map to obtain the target driving route;
[0156] Performing road matching on the target driving route according to target data corresponding to the current moment to determine the target road, wherein the target data includes target real-time differential positioning data, target inertial measurement unit data, target odometer data, and target camera perception data;
[0157] generating a new target local map based on the target data;
[0158] Performing lane matching on the target road based on the new target local map and the fused map to output first position information;
[0159] The target positioning information is outputted through the first position information, the target satellite dead reckoning attitude and the target camera perception data, and the target satellite dead reckoning attitude is determined by the target data.
[0160] Among them, the functional implementation of each module in the above-mentioned fusion positioning system corresponds to the various steps in the above-mentioned fusion positioning method embodiment, and its functions and implementation processes will not be repeated here one by one.
[0161] In a third aspect, an embodiment of the present application provides a fusion positioning device, which may be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities.
[0162] Reference Figure 5 , Figure 5 Schematic diagram of the hardware structure of the fusion positioning device involved in the embodiment of the present application. In the embodiment of the present application, the fusion positioning device may include a processor, a memory, a communication interface and a communication bus.
[0163] The communication bus may be of any type and is used to interconnect the processor, the memory, and the communication interface.
[0164] The communication interface includes input / output (I / O) interface, physical interface and logical interface, etc., which are used to realize the interconnection of devices inside the fusion positioning device, and the interface used to realize the interconnection between the fusion positioning device and other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, a fiber optic interface, an ATM interface, etc.; the user device can be a display, a keyboard, etc.
[0165] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.
[0166] The processor may be a general-purpose processor, and the general-purpose processor may call the fusion positioning program stored in the memory and execute the fusion positioning method provided in the embodiment of the present application. For example, the general-purpose processor may be a central processing unit (CPU). The method executed when the fusion positioning program is called may refer to the various embodiments of the fusion positioning method of the present application, and will not be repeated here.
[0167] Those skilled in the art will understand that Figure 5 The hardware structure shown in the figure does not constitute a limitation on the present application, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.
[0168] In a fourth aspect, an embodiment of the present application also provides a computer-readable storage medium.
[0169] The readable storage medium of the present application stores a fusion positioning program, wherein when the fusion positioning program is executed by a processor, the steps of the fusion positioning method as described above are implemented.
[0170] Among them, the method implemented when the fusion positioning program is executed can refer to the various embodiments of the fusion positioning method of the present application, and will not be repeated here.
[0171] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.
[0172] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.
[0173] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.
[0174] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; the “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.
[0175] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.
[0176] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.
[0177] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A fusion positioning method, characterized in that: The fusion positioning method comprises: Obtain real-time differential positioning data, inertial measurement unit data, odometer data, and camera perception data; Performing satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data, and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude; Converting the camera sensing data into a target position sensing result in an earth coordinate system according to the satellite dead reckoning position and satellite dead reckoning attitude; Generate a local map of the target through the target position perception result; The high-precision map is updated based on the target local map to obtain a fused map, so as to realize vehicle positioning through the fused map.
2. The fusion positioning method according to claim 1, characterized in that: The performing satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude includes: When the solution state corresponding to the real-time differential positioning data is a fixed solution, the position and attitude angle in the real-time differential positioning data are used as the satellite dead reckoning position and the satellite dead reckoning attitude respectively; When the solution state corresponding to the real-time differential positioning data is a non-fixed solution, satellite dead reckoning is performed based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude.
3. The fusion positioning method according to claim 2, characterized in that: The performing satellite dead reckoning based on the position and attitude angle in the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain the satellite dead reckoning position and the satellite dead reckoning attitude includes: Calculating the dead reckoning position based on the attitude angle and position in the real-time differential positioning data, the vehicle speed in the odometer data, and the inter-frame time difference in the inertial measurement unit data; The satellite dead reckoning attitude is calculated based on the attitude angle in the real-time differential positioning data, the angular velocity in the inertial measurement unit data and the time difference between frames.
4. The fusion positioning method according to claim 1, characterized in that: The converting the camera perception data into a target position perception result in an earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude includes: Convert the satellite dead reckoning attitude into an attitude conversion matrix from the vehicle coordinate system to the earth coordinate system; The camera perception data is transformed based on the attitude transformation matrix and the satellite estimated position to obtain a target position perception result in an earth coordinate system.
5. The fusion positioning method according to claim 1, characterized in that: The updating of the high-precision map based on the target local map to obtain a fused map includes: When it is determined according to the target local map that the lane line has been linearly modified, the lane line in the high-precision map is updated based on the target local map to obtain a fused map; When a roadblock is detected according to the target local map, the high-precision map is updated using the target local map and the roadblock information to obtain a fused map.
6. The fusion positioning method according to claim 1, characterized in that: The updating of the high-precision map based on the target local map to obtain a fused map also includes: When the current road section is determined to be a road section where road repair is completed according to the target local map, determining whether the current road section information memorized in the target local map is consistent with the current road section information memorized in the high-precision map; If they are consistent, the high-precision map will be used as the fused map; If there is inconsistency, the inconsistency information is sent to the user, so that the user can determine whether a map update is required based on the inconsistency information; If necessary, the high-precision map is updated based on the target local map to obtain a fused map; If not necessary, the current road section is marked as a blacklist road section in the high-precision map to generate a fused map.
7. The fusion positioning method according to claim 1, characterized in that: The method of realizing vehicle positioning through the fused map includes: Acquire the target vehicle navigation data at the current moment, and fuse the target vehicle navigation data with the fused map to obtain the target driving route; Performing road matching on the target driving route according to target data corresponding to the current moment to determine the target road, wherein the target data includes target real-time differential positioning data, target inertial measurement unit data, target odometer data, and target camera perception data; generating a new target local map based on the target data; Performing lane matching on the target road based on the new target local map and the fused map to output first position information; The target positioning information is outputted through the first position information, the target satellite dead reckoning attitude and the target camera perception data, and the target satellite dead reckoning attitude is determined by the target data.
8. A fusion positioning system, characterized in that: The fusion positioning system comprises: A data acquisition module, which is used to acquire real-time differential positioning data, inertial measurement unit data, odometer data, and camera perception data; A data solving module, which is used to perform satellite dead reckoning based on the real-time differential positioning data, the inertial measurement unit data and the odometer data to obtain a satellite dead reckoning position and a satellite dead reckoning attitude; and convert the camera perception data into a target position perception result in an earth coordinate system according to the satellite dead reckoning position and the satellite dead reckoning attitude; A map generation module, which is used to generate a local map of the target based on the target position perception result; update the high-precision map based on the local map of the target to obtain a fused map; A positioning and solving module is used to realize vehicle positioning through the fused map.
9. A fusion positioning device, characterized in that: The fusion positioning device comprises a processor, a memory, and a fusion positioning program stored in the memory and executable by the processor, wherein when the fusion positioning program is executed by the processor, the steps of the fusion positioning method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a fusion positioning program, wherein when the fusion positioning program is executed by a processor, the steps of the fusion positioning method according to any one of claims 1 to 7 are implemented.