Vehicle mapping method, map updating method, and mapping system
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-06-24
- Publication Date
- 2026-08-13
Smart Images
Figure CN2025103070_13082026_PF_FP_ABST
Abstract
Description
A vehicle mapping method, a map updating method, and a mapping system Technical Field
[0001] This application relates to the field of mapping technology, and in particular to a vehicle mapping method, a map updating method, and a mapping system. Background Technology
[0002] With the rapid development of autonomous driving technology and intelligent driving assistance systems, the perception capabilities of vehicles in low-speed driving scenarios are becoming increasingly important. Typical low-speed scenarios include parking, entering parking spaces, and following other vehicles at low speeds. In these situations, vehicles need to accurately perceive the position and shape of obstacles in the surrounding environment to improve driving safety and optimize the driving experience. To achieve this goal, vehicle-mounted sensor technology has been widely used. Among them, vehicle-mounted ultrasonic radar, with its low cost, simple structure, and strong environmental adaptability, has demonstrated significant application value in short-range obstacle detection.
[0003] When ultrasonic radar detects multiple obstacles, the echo signals may overlap, making it difficult to accurately distinguish and track multiple targets. Traditional algorithms often only detect the strongest echo, ignoring other possible obstacles and reducing the overall environmental awareness. Summary of the Invention
[0004] To address the aforementioned technical problems, this application provides a vehicle mapping method, a map updating method, and a mapping system that can improve environmental perception capabilities.
[0005] Specifically, this application provides a vehicle mapping method, which pre-sets a map matrix. The vehicle mapping method includes: acquiring radar echo signals through multiple sensors and obtaining a confidence distribution based on the radar echo signals; monitoring vehicle motion parameters in real time and obtaining obstacle coordinates based on the vehicle motion parameters and the confidence distribution, so as to update the map matrix based on the obstacle coordinates; and establishing a preliminary map based on the map matrix and performing fusion processing based on the preliminary maps corresponding to each sensor to obtain a target map.
[0006] In the above technical solution, a confidence-based algorithm is used instead of the traditional triangulation method, which can process radar echo signals more meticulously. By using the confidence of different echo signals, the probability and location of obstacles can be determined, improving the overall environmental perception capability. Especially in multi-target detection, it can effectively distinguish overlapping echoes, accurately identify multiple obstacles, and provide more accurate obstacle information in complex environments.
[0007] Furthermore, the process of acquiring radar echo signals also includes: acquiring the radar position coordinates corresponding to each sensor, and converting the radar position coordinates from the vehicle coordinate system to the world coordinate system.
[0008] In the above technical solution, by converting the data from all sensors to the same world coordinate system, it can be ensured that the data from different sensors are processed and fused under the same reference frame. Since the radar is fixedly mounted on the vehicle body, its position remains unchanged in the vehicle coordinate system, so it is necessary to convert the coordinates from the vehicle coordinate system to the world coordinate system. Processing data in a unified coordinate system can simplify the implementation of the algorithm and avoid the complexity caused by frequent conversions between multiple coordinate systems. This not only reduces the development difficulty but may also improve the real-time performance of the system.
[0009] Furthermore, the acquisition of the confidence distribution includes: filtering the radar echo signal to obtain obstacle echoes, and obtaining reflection point information based on multiple obstacle echoes; obtaining reflection point coordinates based on the converted radar position coordinates and the reflection point information; and obtaining the corresponding confidence based on the reflection point coordinates, and acquiring the confidence distribution based on the confidence obtained within a preset time period.
[0010] In the above technical solutions, filtering can remove background noise and interference signals, improve the quality of radar echo signals, thereby reducing false detections and missed detections. Through filtering, obstacles can be identified and located more accurately, improving the accuracy and reliability of detection. Reflection point information is obtained based on multiple obstacle echoes, making the acquisition of reflection points more accurate. Confidence distribution provides global environmental perception information, which helps to fully understand the distribution of obstacles in the surrounding environment and is more conducive to map building.
[0011] Furthermore, obtaining obstacle coordinates includes: determining a confidence threshold based on the vehicle motion parameters, comparing each confidence level in the confidence distribution with the confidence threshold, and if there is a confidence level higher than the confidence threshold, then using the reflection point coordinates corresponding to the confidence level as obstacle coordinates; otherwise, determining that there is no obstacle.
[0012] In the above technical solution, the confidence threshold is dynamically adjusted according to the vehicle motion parameters (such as speed), so that the sensitivity and reliability of obstacle detection can be optimized under different driving conditions. For example, the faster the vehicle speed, the sparser the reflection points, so the corresponding confidence threshold needs to be lower. By setting an appropriate confidence threshold, low-confidence reflection points can be effectively filtered out, thereby reducing false detections, improving the accuracy of obstacle detection, and helping the vehicle to more accurately identify real obstacles, avoiding unnecessary operation or safety risks caused by false alarms.
[0013] Furthermore, the map matrix includes several cells, each cell corresponding to a map location, and the value of each cell represents the probability that there is an obstacle at the current map location; wherein, if the current map location does not return a radar echo signal, the corresponding cell is set to the first value.
[0014] Furthermore, updating the map matrix based on the obstacle coordinates includes: when the reflection point coordinates are used as obstacle coordinates, setting the cell corresponding to the obstacle coordinates to a second value; otherwise, setting the cell corresponding to the reflection point coordinates to a third value.
[0015] In the above technical solution, the map matrix can reflect the obstacle situation around the target vehicle, and the detected echo information will be continuously superimposed on the map. As the vehicle moves and is detected, multiple obstacle outlines can be drawn based on the obtained map matrix, which helps to construct a local map and makes the obstacle information in the local map clear and complete.
[0016] Furthermore, the establishment of the preliminary map includes: obtaining the edge contours of obstacles based on several second values, and marking the obstacles based on the edge contours of the obstacles, so as to establish a preliminary map based on the marking results and the current position of the target vehicle.
[0017] In the above technical solution, by acquiring the edge contours of obstacles, the shape and position of obstacles can be depicted more accurately. Compared with maps that only rely on the coordinates of the center point of obstacles, this edge contour information can provide a more detailed environmental description, thereby improving the accuracy of the map. Based on the marking results and the current position of the target vehicle, a preliminary map is established, making map updates more flexible and real-time. During the vehicle's movement, the map can be continuously updated and corrected to ensure that the map always reflects the current environmental state.
[0018] Furthermore, based on the same concept, this application also provides a vehicle map updating method, including: determining in real time whether the target vehicle has traveled to the boundary of the target map based on the vehicle motion parameters; if so, re-establishing the target map based on the current position of the target vehicle using the vehicle mapping method; otherwise, maintaining the original target map.
[0019] In the above technical solution, by detecting in real time whether the vehicle has traveled to the boundary of the target map, it is possible to dynamically determine whether the map needs to be rebuilt. This method ensures that the map can be updated in real time as the vehicle moves, reflecting the latest environmental conditions, thereby improving the timeliness and accuracy of the map.
[0020] Furthermore, the vehicle map update method further includes: obtaining the overlapping areas between the re-established target map and the original target map, and copying the marked obstacles in the overlapping areas to the re-established target map.
[0021] In the above technical solution, by copying obstacle markers in overlapping areas, the consistency between the new map and the original map in overlapping areas is ensured, avoiding discontinuities between maps and making map updates smoother and more natural; it also avoids re-detecting and marking obstacles in overlapping areas, saving computing resources and time, and improving the efficiency of map updates.
[0022] Furthermore, determining whether the target vehicle has traveled to the target map boundary includes: obtaining the distance between the target vehicle and the target map boundary based on the current position of the target vehicle; if the distance is less than a preset distance threshold, then it is determined that the target vehicle has traveled to the target map boundary; otherwise, it is determined that the target vehicle has not traveled to the target map boundary.
[0023] In the above technical solution, by calculating the distance between the target vehicle and the map boundary, it is possible to accurately determine whether the vehicle is approaching or has reached the boundary, avoiding misjudgment caused by fuzzy judgment; the preset distance threshold can be adjusted according to actual needs, such as setting different thresholds in different driving scenarios, thereby improving the flexibility and adaptability of the judgment.
[0024] Based on the same concept, this application also provides a vehicle mapping system, which includes at least an intelligent driving domain controller; the intelligent driving domain controller is used to build a target map based on acquired radar echo signals and update the target map based on vehicle motion parameters.
[0025] Furthermore, the intelligent driving domain controller includes at least a memory and a processor; the memory is used to store computer instructions for multiple functional layers, the functional layers including at least a mapping functional layer and a vehicle position determination functional layer, each functional layer including one or more functional modules; the processor communicates with the memory via a bus and is used to execute the computer instructions for each of the multiple functional layers stored in the memory.
[0026] Furthermore, the mapping function layer includes at least an acquisition function module, a map matrix update function module, and a mapping function module; the acquisition function module includes computer instructions for acquiring radar echo signals and obtaining a confidence distribution based on the radar echo signals; the map matrix update function module includes computer instructions for real-time monitoring of vehicle motion parameters and obtaining obstacle coordinates based on the vehicle motion parameters and the confidence distribution, so as to update the map matrix based on the obstacle coordinates; the mapping function module includes computer instructions for building a preliminary map based on the map matrix and performing fusion processing based on the preliminary maps corresponding to each sensor to obtain a target map.
[0027] Furthermore, the vehicle position determination function layer includes at least a vehicle position determination function module; the vehicle position determination function module includes a computer instruction for determining in real time whether the target vehicle has traveled to the boundary of the target map based on the vehicle motion parameters, so as to issue an update instruction to the mapping function module when the determination result is that the target vehicle has traveled to the boundary of the target map.
[0028] Furthermore, the vehicle mapping system also includes multiple sensors; the sensors are used to acquire radar echo signals and send the radar echo signals as input data to the acquisition function module.
[0029] In the above technical solution, the confidence algorithm is used to avoid the problem of inconsistent reflection points in multiple detections. This reduces the amount of computation, simplifies the strategy, enhances the understandability, and improves the overall environmental perception capability.
[0030] Compared with the prior art, the beneficial effects of this application are as follows:
[0031] This application acquires radar echo signals through multiple sensors and obtains a confidence distribution based on the radar echo signals. Then, it monitors vehicle motion parameters in real time and obtains obstacle coordinates based on the vehicle motion parameters and the confidence distribution, updating the map matrix based on these obstacle coordinates. Furthermore, it establishes a preliminary map based on the map matrix and performs fusion processing on the preliminary maps corresponding to each sensor to obtain a target map. This application employs a confidence algorithm that enables more detailed processing of radar echo signals, improving the overall environmental perception capability. Especially in multi-target detection, it can effectively distinguish overlapping echoes, accurately identify multiple obstacles, and provide more accurate obstacle information in complex environments. Attached Figure Description
[0032] Figure 1 is a flowchart of the vehicle mapping method described in this application.
[0033] Figure 2 is a schematic diagram of the confidence distribution record described in this application.
[0034] Figure 3 is a schematic diagram of the preliminary map described in this application.
[0035] Figure 4 is a flowchart of the vehicle map update method described in this application.
[0036] Figure 5 is a schematic diagram of the target vehicle described in this application traveling to the boundary of the target map.
[0037] Figure 6 is a schematic diagram of the updated target map described in this application.
[0038] Figure 7 is a framework diagram of the vehicle mapping system described in this application.
[0039] Figure 8 is a framework diagram of the intelligent driving domain controller described in this application.
[0040] Figure 9 is a framework diagram of the memory described in this application.
[0041] Figure 10 is a schematic diagram of the connection between the sensor and the memory described in this application. Detailed Implementation
[0042] The following describes in further detail a vehicle mapping method, map updating method, and mapping system of this application with reference to specific embodiments and accompanying drawings.
[0043] Please refer to Figure 1. This application provides a vehicle mapping method, which pre-sets a map matrix and includes the following steps S100-S300.
[0044] In some embodiments, radar echo signals are acquired through APA (Automatic Parking Assistant) and UPA (Ultrasonic Parking Assistant). Then, through continuous echo recording as the target vehicle moves, the confidence level of real reflection points on the map accumulates to a relatively high level, while the confidence level of false reflection points decreases due to other radar echo signals. At this point, the confidence distribution of obstacles related to the target vehicle's environment can be obtained. Further, an appropriate confidence threshold is selected based on vehicle motion parameters (such as vehicle speed), and obstacles are determined according to the confidence threshold and the confidence distribution to confirm obstacle coordinates and a pre-set map matrix. Finally, a preliminary map can be built based on the updated map matrix. Since multiple sensors acquire radar echo signals, multiple preliminary maps are generated. At this point, the preliminary maps corresponding to each sensor need to be fused to obtain the final target map.
[0045] Steps S100-S300 are described in detail below.
[0046] Step S100: Obtain radar echo signals through multiple sensors and obtain the confidence distribution based on the radar echo signals.
[0047] In some embodiments, radar echo signals are acquired, for example, through APA and UPA. The probes on the same side of the APA have similar angles, which can play a role in repeated detection during the journey. However, the probes of the UPA have dispersed angles, which can easily interfere with the APA under the reflection of special obstacles. Therefore, when building the initial map, it is necessary to build maps corresponding to APA and UPA separately before performing subsequent fusion processing.
[0048] It should be noted that, in other embodiments, those skilled in the art may also select other sensors or more sensors according to actual application requirements, and no limitation is imposed here.
[0049] Furthermore, the process of acquiring radar echo signals also includes: acquiring the radar position coordinates corresponding to each sensor, and converting the radar position coordinates from the vehicle coordinate system to the world coordinate system.
[0050] In some embodiments, when the target vehicle is in motion, the transformation of the radar position coordinates from the vehicle coordinate system to the world coordinate system is involved:
[0051] ;
[0052] The coordinates of the rear axle center of the car in the world coordinate system are ( ). , The direction of the car is The radar's coordinates in the vehicle coordinate system are ( , ).
[0053] In the above technical solution, by converting the data from all sensors to the same world coordinate system, it can be ensured that the data from different sensors are processed and fused under the same reference frame. Since the radar is fixedly mounted on the vehicle body, its position remains unchanged in the vehicle coordinate system, so it is necessary to convert the coordinates from the vehicle coordinate system to the world coordinate system. Processing data in a unified coordinate system can simplify the implementation of the algorithm and avoid the complexity caused by frequent conversions between multiple coordinate systems. This not only reduces the development difficulty but may also improve the real-time performance of the system.
[0054] Furthermore, the acquisition of the confidence distribution includes: filtering the radar echo signal to obtain obstacle echoes, and obtaining reflection point information based on multiple obstacle echoes; obtaining reflection point coordinates based on the converted radar position coordinates and the reflection point information; and obtaining the corresponding confidence based on the reflection point coordinates, and acquiring the confidence distribution based on the confidence obtained within a preset time period.
[0055] In some embodiments, each radar echo signal, after filtering, is considered a real obstacle echo (excluding ground echoes), and its reflection point lies on an arc centered on the probe, with a radius equal to half the flight distance, and defined by the probe's beam angle direction and angle; therefore, the reflection point satisfies the following equation:
[0056] ;
[0057] The reflection point is represented in the world coordinate system as follows: The radar position coordinates are represented in the world coordinate system as follows: The flight distance is half of r, and the beam angle range is... (i.e., the reflection point information).
[0058] Each echo-received arc is recorded on the map, with the arc thickness set to 2 grids. The confidence level for all grids on the arc is increased by 10, while the confidence level for all points from the arc to the center is decreased by 10. This is because if there are obstacles within the arc, the echo will be received earlier, thus reducing the probability of obstacles within the arc (see Figure 2, which shows the data recording of one echo).
[0059] The confidence level is determined based on the sensor error distribution characteristics to derive the confidence level distribution for each echo record. The test value exhibits a normal distribution near the true value. Therefore, when recording echoes, the confidence level is set to 10 for the arc along the radius of the measured value, while the confidence levels on either side are halved. Based on the fan-shaped detection characteristics of ultrasonic radar, the reflection intensity is strongest when the obstacle is directly in front, and lower or even nonexistent when the obstacle is to the side. Therefore, in a single echo, the probability of an obstacle being directly in front is greater than that of being to the side; when recording echoes, the confidence level on either side of the arc is lower than that for the front. Within the arc, the probability of an obstacle is considered lower because if there were an obstacle, an echo should have been generated earlier; therefore, the confidence level for within the arc is negative.
[0060] It should be noted that, since the faster the vehicle speed, the sparser the reflection points become, at least two different radar echoes are needed to determine the location of the same reflection point.
[0061] In the above technical solutions, filtering can remove background noise and interference signals, improve the quality of radar echo signals, thereby reducing false detections and missed detections. Through filtering, obstacles can be identified and located more accurately, improving the accuracy and reliability of detection. Reflection point information is obtained based on multiple obstacle echoes, making the acquisition of reflection points more accurate. Confidence distribution provides global environmental perception information, which helps to fully understand the distribution of obstacles in the surrounding environment and is more conducive to map building.
[0062] After obtaining the confidence distribution, step S200 can be executed.
[0063] Step S200: Monitor vehicle motion parameters in real time, and obtain obstacle coordinates based on the vehicle motion parameters and confidence distribution, so as to update the map matrix based on the obstacle coordinates.
[0064] Furthermore, obtaining obstacle coordinates includes: determining a confidence threshold based on the vehicle motion parameters, comparing each confidence level in the confidence distribution with the confidence threshold, and if there is a confidence level higher than the confidence threshold, then using the reflection point coordinates corresponding to the confidence level as obstacle coordinates; otherwise, determining that there is no obstacle.
[0065] In some embodiments, by recording continuous echoes as the target vehicle moves, the real reflection points on the map will accumulate a relatively high confidence level, while the false reflection points on the arc will have their confidence level reduced due to other echoes; an appropriate confidence threshold is selected for error performance and vehicle speed (i.e., the vehicle motion parameters).
[0066] Among these requirements, the map resolution directly affects the limit on the maximum vehicle speed, requiring a maximum distance of L between reflection points and a radar duty cycle of [missing information]. Then the vehicle speed must meet the following requirements:
[0067] ;
[0068] Since the faster the vehicle speed, the sparser the reflection points become, the corresponding confidence threshold needs to be lower.
[0069] Furthermore, based on the determined confidence threshold, the corresponding obstacle coordinates can be obtained.
[0070] In the above technical solution, the confidence threshold is dynamically adjusted according to the vehicle motion parameters (such as speed), so that the sensitivity and reliability of obstacle detection can be optimized under different driving conditions. For example, the faster the vehicle speed, the sparser the reflection points, so the corresponding confidence threshold needs to be lower. By setting an appropriate confidence threshold, low-confidence reflection points can be effectively filtered out, thereby reducing false detections, improving the accuracy of obstacle detection, and helping the vehicle to more accurately identify real obstacles, avoiding unnecessary operation or safety risks caused by false alarms.
[0071] Specifically, the map matrix includes several cells, each corresponding to a map location. The value of each cell represents the probability that an obstacle exists at the current map location. If no radar echo signal is returned at the current map location, the corresponding cell is set to a first value. Updating the map matrix based on the obstacle coordinates includes: when the reflection point coordinates are used as obstacle coordinates, setting the cell corresponding to the obstacle coordinates to a second value; otherwise, setting the cell corresponding to the reflection point coordinates to a third value.
[0072] In some embodiments, the map matrix is, for example, a two-dimensional map matrix MAP, which is a matrix composed of 2000×2000 cells, and each cell MAP[x][y] corresponds to a position on the map. The value of each cell can be represented as follows:
[0073] MAP[x][y]=a1 (i.e., the first numerical value), indicating that the position is in an undetected state and it is unknown whether there is an obstacle (i.e., no radar echo signal is returned); MAP[x][y]=a2 (i.e., the second numerical value), indicating that the possibility of an obstacle existing at this position is increased and there are corresponding obstacle coordinates; MAP[x][y]=a3 (i.e., the third numerical value), indicating that the possibility of an obstacle existing at this position is decreased and the corresponding reflection point coordinates cannot be used as obstacle coordinates.
[0074] The above a3 < a1 < a2, and the magnitude of the numbers is used to represent the possibility of an obstacle existing in the cell.
[0075] After completing the update of the map matrix, step S300 can be executed.
[0076] Step S300: Establish a preliminary map based on the map matrix and perform fusion processing on the preliminary maps corresponding to each sensor to obtain a target map.
[0077] Furthermore, establishing the preliminary map includes: obtaining the edge contour of the obstacle according to a number of second numerical values, and marking the obstacle according to the edge contour of the obstacle, so as to establish a preliminary map based on the marking result and the current position of the target vehicle.
[0078] In some embodiments, with the center of the rear axle center of the target vehicle as the center (i.e., the current position), a square boundary with a side length of a preset length (such as 5 meters) is established as the map boundary; ultrasonic detection is continuously performed when the vehicle is moving forward, and the echo information is superimposed on the map, and as the position of the vehicle and the detection progress, the corresponding obstacle edge contour can be obtained based on the map matrix set by the obstacle coordinates (as shown in Figure 3), so as to construct a local map that can realize the surrounding environment of the target vehicle (i.e., the preliminary map).
[0079] Furthermore, after obtaining the preliminary maps corresponding to APA and UPA, perform fusion processing on them to obtain the final target map.
[0080] Among them, those skilled in the art can select a suitable fusion algorithm according to actual application requirements, which is not limited here.
[0081] In the above technical solution, by acquiring the edge contours of obstacles, the shape and position of obstacles can be depicted more accurately. Compared with maps that only rely on the coordinates of the center point of obstacles, this edge contour information can provide a more detailed environmental description, thereby improving the accuracy of the map. Based on the marking results and the current position of the target vehicle, a preliminary map is established, making map updates more flexible and real-time. During the vehicle's movement, the map can be continuously updated and corrected to ensure that the map always reflects the current environmental state.
[0082] Furthermore, based on the same concept, please refer to Figure 4. This application also provides a vehicle map updating method, including: determining whether the target vehicle has traveled to the boundary of the target map based on the vehicle motion parameters; if so, re-establishing the target map based on the current position of the target vehicle using the vehicle mapping method; otherwise, maintaining the original target map.
[0083] In some embodiments, the target vehicle is in motion. In order to provide a real-time map representation of obstacle distribution, it is necessary to ensure that the target vehicle is always within the map range of the world coordinate system. Whenever the target vehicle travels to the edge of the map, a new map is reconstructed with the rear axle center of the target vehicle as the center.
[0084] In the above technical solution, by detecting in real time whether the vehicle has traveled to the boundary of the target map, it is possible to dynamically determine whether the map needs to be rebuilt. This method ensures that the map can be updated in real time as the vehicle moves, reflecting the latest environmental conditions, thereby improving the timeliness and accuracy of the map.
[0085] Furthermore, the vehicle map update method further includes: obtaining the overlapping areas between the re-established target map and the original target map, and copying the marked obstacles in the overlapping areas to the re-established target map.
[0086] In some embodiments, the initial origin coordinates of the world coordinate system map are reset, and the overlapping parts of the original map and the new map are copied to the new map to preserve reasonable historical information. A square boundary with a side length of 5m is established with the center of the rear axle of the car as the centroid and the front of the car as the direction. Whenever the boundary exceeds the map boundary or coincides with the map boundary (as shown in Figure 5), the reconstruction of the new map begins (Figure 6 shows the updated target map).
[0087] In the above technical solution, by copying obstacle markers in overlapping areas, the consistency between the new map and the original map in overlapping areas is ensured, avoiding discontinuities between maps and making map updates smoother and more natural; it also avoids re-detecting and marking obstacles in overlapping areas, saving computing resources and time, and improving the efficiency of map updates.
[0088] Furthermore, determining whether the target vehicle has traveled to the target map boundary includes: obtaining the distance between the target vehicle and the target map boundary based on the current position of the target vehicle; if the distance is less than a preset distance threshold, then it is determined that the target vehicle has traveled to the target map boundary; otherwise, it is determined that the target vehicle has not traveled to the target map boundary.
[0089] In some embodiments, a preset distance threshold is used to determine whether a vehicle is approaching the map boundary. This threshold can be adjusted according to actual needs, such as based on factors like vehicle speed and map update frequency. If the calculated distance is less than the preset distance threshold, it is determined that the vehicle has reached the map boundary. Then, based on the vehicle's current position, a new map is reconstructed using a vehicle mapping method.
[0090] In the above technical solution, by calculating the distance between the target vehicle and the map boundary, it is possible to accurately determine whether the vehicle is approaching or has reached the boundary, avoiding misjudgment caused by fuzzy judgment; the preset distance threshold can be adjusted according to actual needs, such as setting different thresholds in different driving scenarios, thereby improving the flexibility and adaptability of the judgment.
[0091] Furthermore, based on the same concept, please refer to Figure 7. This application also provides a vehicle mapping system, which includes at least an intelligent driving domain controller 10. The intelligent driving domain controller 10 is used to build a target map based on the acquired radar echo signal and update the target map based on vehicle motion parameters.
[0092] Furthermore, referring to Figure 8, the intelligent driving domain controller 10 includes at least a memory 101 and a processor 102. The memory 101 (as shown in Figure 9) is used to store computer instructions for multiple functional layers, each functional layer including at least a mapping functional layer 104 and a vehicle position determination functional layer 105, with each functional layer including one or more functional modules. The processor 102 communicates with the memory 101 via a bus and is used to execute the computer instructions for each of the multiple functional layers stored in the memory 101.
[0093] The memory 101 includes, but is not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EPROM (Erasable Programmable Read-Only Memory), and CD-ROM (Compact Disc Read-Only Memory); the processor 102 includes a central processing unit 102 or a device or module with processing capabilities.
[0094] Furthermore, the mapping function layer 104 includes at least an acquisition function module 1041, a map matrix update function module 1042, and a mapping function module 1043; the acquisition function module 1041 includes computer instructions for acquiring radar echo signals and acquiring confidence distributions based on the radar echo signals; the map matrix update function module 1043 includes computer instructions for real-time monitoring of vehicle motion parameters and acquiring obstacle coordinates based on the vehicle motion parameters and confidence distributions, so as to update the map matrix based on the obstacle coordinates; the mapping function module 1043 includes computer instructions for building a preliminary map based on the map matrix and performing fusion processing based on the preliminary maps corresponding to each sensor to obtain a target map.
[0095] Furthermore, the vehicle position determination function layer 105 includes at least a vehicle position determination function module 1051; the vehicle position determination function module 1051 includes a computer instruction for determining in real time whether the target vehicle has traveled to the boundary of the target map based on the vehicle motion parameters, so as to issue an update instruction to the mapping function module 1043 when the determination result is that the target vehicle has traveled to the boundary of the target map.
[0096] Furthermore, referring to Figure 10, the vehicle mapping system also includes multiple sensors 20; the sensors 20 are used to acquire radar echo signals and send the radar echo signals as input data to the acquisition function module 1041; wherein, the sensors 20 include APA and UPA, etc.
[0097] In summary, this application provides a vehicle mapping method, a map updating method, and a mapping system. First, radar echo signals are acquired through multiple sensors, and a confidence distribution is obtained based on the radar echo signals. Then, vehicle motion parameters are monitored in real time, and obstacle coordinates are obtained based on the vehicle motion parameters and the confidence distribution, thereby updating the map matrix based on the obstacle coordinates. Furthermore, a preliminary map is built based on the map matrix, and the preliminary maps corresponding to each sensor are fused to obtain the target map. This application employs a confidence algorithm that can process radar echo signals more meticulously, improving the overall environmental perception capability. Especially in multi-target detection, it can effectively distinguish overlapping echoes, accurately identify multiple obstacles, and provide more accurate obstacle information in complex environments.
[0098] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of this application. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of this application. All such changes and modifications are intended to be included within the scope of this application as claimed in the appended claims.
[0099] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0100] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.
[0101] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules according to the embodiments of this application. This application can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
[0102] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0103] Although the description of this application has been made in conjunction with the specific embodiments described above, it will be apparent to those skilled in the art that many substitutions, modifications, and variations can be made based on the foregoing. Therefore, all such substitutions, modifications, and variations are included within the spirit and scope of the appended claims.
Claims
1. A vehicle mapping method, wherein, The vehicle mapping method comprises the following steps of: obtaining radar echo signals through a plurality of sensors, and obtaining a confidence distribution according to the radar echo signals (S100); monitoring a vehicle body motion parameter in real time, and obtaining an obstacle coordinate according to the vehicle body motion parameter and the confidence distribution, so as to update the map matrix based on the obstacle coordinate (S200); and establishing a preliminary map according to the map matrix, and performing fusion processing on the preliminary map corresponding to each sensor, so as to obtain a target map (S300).
2. The vehicle mapping method of claim 1, wherein, In the process of obtaining the radar echo signals, the following steps are further included: obtaining radar position coordinates corresponding to each sensor, and converting the radar position coordinates from an automobile coordinate system to a world coordinate system.
3. The vehicle mapping method of claim 2, wherein, The confidence distribution is obtained by: filtering the radar echo signals to obtain obstacle echoes, and obtaining reflection point information based on a plurality of obstacle echoes; obtaining reflection point coordinates according to the converted radar position coordinates and the reflection point information; obtaining corresponding confidence according to the reflection point coordinates, and obtaining a confidence distribution based on the confidence obtained within a preset time period.
4. The vehicle mapping method of claim 3, wherein, The obstacle coordinate is obtained by: determining a confidence threshold based on the vehicle body motion parameter, comparing each confidence in the confidence distribution with the confidence threshold, if there is a confidence higher than the confidence threshold, then the reflection point coordinate corresponding to the confidence is taken as the obstacle coordinate; otherwise, it is determined that there is no obstacle.
5. The vehicle mapping method of claim 4, wherein, The map matrix comprises a plurality of cells, each cell corresponding to a map position, and the value corresponding to each cell represents the probability of the existence of an obstacle at the current map position; wherein if no radar echo signal is returned at the current map position, the corresponding cell is set to a first value.
6. The vehicle mapping method of claim 5, wherein, The map matrix is updated based on the obstacle coordinate, comprising: when the reflection point coordinate is taken as the obstacle coordinate, the cell corresponding to the obstacle coordinate is set to a second value; otherwise, the cell corresponding to the reflection point coordinate is set to a third value.
7. The vehicle mapping method of claim 6, wherein, The preliminary map is established by: obtaining an obstacle edge contour according to a plurality of second values, and marking the obstacle according to the obstacle edge contour, so as to establish a preliminary map based on the marking result and the current position of the target vehicle.
8. A vehicle map updating method, comprising: determining whether the target vehicle travels to the target map boundary based on the vehicle body motion parameter in real time, if yes, re-establishing the target map based on the current position of the target vehicle by using the vehicle mapping method according to claim 1; otherwise, keeping the original target map.
9. The vehicle map updating method according to claim 8, further comprising: obtaining a repeated area of the re-established target map and the original target map, and copying the marked obstacle in the repeated area to the re-established target map.
10. The vehicle map update method according to claim 8, wherein The determination of whether the target vehicle travels to the target map boundary comprises: According to a current position of the target vehicle, a distance between the target vehicle and a target map boundary is obtained, and if the distance is less than a preset distance threshold, it is determined that the target vehicle has traveled to the target map boundary; otherwise, it is determined that the target vehicle has not traveled to the target map boundary.
11. A vehicle mapping system, comprising at least an intelligent driving domain controller (10); wherein, The intelligent driving domain controller (10) is configured to establish a target map based on the obtained radar echo signal and update the target map based on a vehicle body motion parameter.
12. The vehicle mapping system of claim 11, wherein, The intelligent driving domain controller (10) comprises at least a memory (101) and a processor (102); The memory (101) is configured to store computer instructions of a plurality of functional layers, the functional layers comprising at least a mapping functional layer (104) and a vehicle position judgment functional layer (105), each functional layer comprising one or more functional modules; The processor (102) is configured to execute the computer instructions of each of the plurality of functional layers stored in the memory (101) through a bus (103) in communication with the memory (101).
13. The vehicle mapping system of claim 12, wherein, The mapping functional layer (104) comprises at least an obtaining functional module (1041), a map matrix updating functional module (1042), and a mapping functional module (1043); The obtaining functional module (1041) comprises computer instructions for obtaining a radar echo signal and obtaining a confidence distribution based on the radar echo signal; The map matrix updating functional module (1042) comprises computer instructions for monitoring a vehicle body motion parameter in real time and obtaining an obstacle coordinate based on the vehicle body motion parameter and the confidence distribution, so as to update the map matrix based on the obstacle coordinate; The mapping functional module (1043) comprises computer instructions for establishing a preliminary map based on the map matrix and performing fusion processing on the preliminary maps corresponding to each sensor, so as to obtain a target map.
14. The vehicle mapping system of claim 13, wherein, The vehicle position judgment functional layer (105) comprises at least a vehicle position judgment functional module (1051); The vehicle position judgment functional module (1051) comprises computer instructions for judging in real time whether a target vehicle has traveled to a boundary of the target map based on the vehicle body motion parameter, and issuing an updating instruction to the mapping functional module (1043) when the judgment result is that the target vehicle has traveled to the boundary of the target map.
15. The vehicle mapping system of claim 13, further comprising a plurality of sensors (20); wherein, The sensors (20) are configured to obtain a radar echo signal and issue the radar echo signal as input data to the obtaining functional module (1041).