Vehicle collision early warning method and system based on high-precision map

Abstract

The invention discloses a vehicle collision early warning method and system based on a high-precision map. The system comprises a map module, a position module, a track module, a risk module and an early warning module. And the map module calls a high-precision map of the current area of the target vehicle according to the preliminary positioning information. And the position module carries out matching positioning on the environmental perception information around the target vehicle and the high-precision map by adopting an SLAM positioning algorithm to determine the pose information of the target vehicle. And the track module is combined with the pose information and the sensing information of the target vehicle to accurately position the position information of the moving targets around the target vehicle and determine the moving tracks of the target vehicle and the surrounding moving targets. And the risk module evaluates a collision risk level between the target vehicle and the moving target through the moving tracks of the target vehicle and the moving target. And the early warning module compares the collision risk level with an early warning level, and when the collision risk level exceeds the early warning level, early warning information is generated according to the type of the moving target to give an alarm.

Description

Technical Field

[0001] The present invention relates to the technical field of predicting or avoiding possible or imminent collisions, and in particular to a vehicle collision warning method and system based on high-precision maps. Background Art

[0002] At present, most technologies used to assess vehicle collision risks require locating the position of target vehicles and other moving targets, and using the positioning information to assess the collision risk between the target vehicle and the moving target, so as to control the target vehicle to perform evasive maneuvers or issue warning information before a collision occurs, reminding the moving target to avoid the collision, etc. It can be seen that accurately locating the position of target vehicles and other moving targets is the main factor affecting the accuracy of vehicle collision risk assessment.

[0003] However, the current conventional satellite positioning systems, such as the GPS positioning system and the Beidou navigation positioning system, are easily interfered by other signals because they need to receive corresponding satellite signals, and are unable to perform positioning in areas with weak satellite signals, resulting in insufficient accuracy in the positioning results of the target vehicle, which in turn affects the accuracy of the vehicle's collision risk. Summary of the invention

[0004] In view of the shortcomings of the existing technology, the present invention proposes a vehicle collision warning method and system based on high-precision maps, which can improve the accuracy and adaptability of vehicle collision risks. The specific technical solution is as follows:

[0005] In a first aspect, a vehicle collision warning method based on a high-precision map is provided. In a first implementable manner of the first aspect, the method includes:

[0006] Obtain a high-precision map of the area where the target vehicle is located based on the current preliminary positioning information of the target vehicle;

[0007] Based on the environmental perception information around the target vehicle and the high-precision map, a SLAM positioning algorithm is used to locate the position information of the target vehicle in real time;

[0008] Determine the moving trajectory of the target vehicle and each moving target according to the real-time position information and perception information of the target vehicle;

[0009] According to the moving trajectories of the target vehicle and the moving targets, the collision risk level between the target vehicle and each moving target is evaluated.

[0010] In combination with the first implementable manner of the first aspect, in a second implementable manner of the first aspect, using a SLAM positioning algorithm to locate the position information of the target vehicle includes:

[0011] De-noising is performed on the environmental perception information perceived by the target vehicle.

[0012] In combination with the first implementable manner of the first aspect, in a third implementable manner of the first aspect, using a SLAM positioning algorithm to locate the position information of the target vehicle includes:

[0013] Feature extraction is performed on the environmental perception information perceived by the target vehicle.

[0014] In combination with the first implementable manner of the first aspect, in a fourth implementable manner of the first aspect, evaluating the collision risk between the target vehicle and the moving target includes:

[0015] Fitting the movement curves of the target vehicle and the moving target according to the movement trajectories of the target vehicle and the moving target respectively;

[0016] Determining whether there is an intersection between the moving curves of the target vehicle and the moving target;

[0017] In response to the existence of an intersection point between the moving curves of the target vehicle and the moving target, determining the coordinates of the intersection point, and calculating the collision time according to the coordinates of the intersection point, the current position information of the target vehicle, and the vehicle speed;

[0018] A collision risk level between the target vehicle and the moving object is evaluated according to the collision time.

[0019] In combination with the first implementable manner of the first aspect, in a fifth implementable manner of the first aspect, further comprising:

[0020] The collision risk level corresponding to each mobile target is compared with the warning level, and corresponding warning information is generated according to the perception information of the mobile target whose collision risk level exceeds the warning level for warning.

[0021] In a second aspect, a vehicle collision warning system based on a high-precision map is provided. In a first implementable manner of the second aspect, the system includes:

[0022] A map module, configured to obtain a high-precision map of the area where the target vehicle is located based on the current preliminary positioning information of the target vehicle;

[0023] A location module is configured to locate the position information of the target vehicle in real time using a SLAM positioning algorithm based on the environmental perception information around the target vehicle and the high-precision map;

[0024] A trajectory module, configured to determine the moving trajectory of the target vehicle and each moving target according to the real-time position information and perception information of the target vehicle;

[0025] The risk module is configured to evaluate the collision risk level between the target vehicle and each moving target according to the moving trajectories of the target vehicle and the moving targets.

[0026] In combination with the first implementable manner of the second aspect, in a second implementable manner of the second aspect, the position module includes: a noise processing unit configured to perform denoising processing on the environmental perception information perceived by the target vehicle.

[0027] In combination with the first implementable manner of the second aspect, in a third implementable manner of the second aspect, the position module includes: a feature extraction unit configured to extract features of environmental perception information perceived by the target vehicle.

[0028] In combination with the first implementable manner of the second aspect, in a fourth implementable manner of the second aspect, the risk module includes:

[0029] A curve fitting unit, configured to fit the movement curves of the target vehicle and the moving target according to the movement trajectories of the target vehicle and the moving target respectively;

[0030] a collision analysis unit configured to determine whether there is an intersection between the moving curves of the target vehicle and the moving target;

[0031] In response to the existence of an intersection point between the moving curves of the target vehicle and the moving target, determining the coordinates of the intersection point, and calculating the collision time according to the coordinates of the intersection point, the current position information of the target vehicle, and the vehicle speed;

[0032] The risk assessment unit is configured to assess a collision risk level between the target vehicle and the moving target according to the collision time.

[0033] In combination with the first implementable manner of the second aspect, in a fifth implementable manner of the second aspect, further comprising:

[0034] The warning module is configured to compare the collision risk level corresponding to each mobile target with the warning level, and generate corresponding warning information for warning based on the perception information of the mobile target whose collision risk level exceeds the warning level.

[0035] Beneficial effects: The vehicle collision warning method and system based on high-precision maps of the present invention can obtain a high-precision map of the area where the target vehicle is located according to the current approximate location of the target vehicle. The high-precision map can provide rich geographic information. The perception information of the target vehicle is matched with the high-precision map using the SLAM positioning algorithm, which can more accurately locate the position information of the target vehicle and is not easily interfered by other signals, thereby improving the positioning accuracy of the target vehicle and achieving real-time positioning of the target vehicle in areas with weak satellite signals. Combined with the precise position information and perception information of the target vehicle, the position of other mobile targets can be accurately located, thereby improving the positioning accuracy of the mobile target, and further improving the accuracy and adaptability of the entire vehicle collision risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the specific implementation of the present invention, the following will briefly introduce the drawings required for use in the specific implementation. In all the drawings, each element or part is not necessarily drawn according to the actual scale.

[0037] Figure 1 A flowchart of a vehicle collision warning method based on a high-precision map provided in one embodiment of the present invention;

[0038] Figure 2 A system block diagram of a high-precision map-based vehicle collision warning system provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0040] like Figure 1 The flowchart of the vehicle collision warning method based on high-precision map shown in FIG. 1 includes:

[0041] Obtain a high-precision map of the area where the target vehicle is located based on the current preliminary positioning information of the target vehicle;

[0042] Based on the environmental perception information around the target vehicle and the high-precision map, a SLAM positioning algorithm is used to locate the position information of the target vehicle in real time;

[0043] Determine the moving trajectory of the target vehicle and each moving target according to the real-time position information and perception information of the target vehicle;

[0044] According to the moving trajectories of the target vehicle and the moving targets, the collision risk level between the target vehicle and each moving target is evaluated.

[0045] Specifically, first, according to the current preliminary positioning information of the target vehicle, a high-precision map of the current area of ​​the target vehicle can be retrieved. It should be understood that the preliminary positioning information corresponding to the target vehicle does not need to be too accurate, and it can represent the approximate area where the target vehicle is located. The positioning coordinates of the target vehicle when the satellite positioning system can operate normally can be used as preliminary positioning information. Then, the environmental perception information around the target vehicle is perceived by the perception device carried by the target vehicle, such as laser radar, camera, etc., and the environmental perception information is matched and positioned with the high-precision map using the SLAM positioning algorithm to determine the position information of the target vehicle. After that, combined with the position information and perception information of the target vehicle, the position information of the mobile targets around the target vehicle can be accurately located, thereby locating the moving trajectory of the target vehicle and the surrounding mobile targets. Finally, the collision risk level between the target vehicle and the mobile target is evaluated by the moving trajectory of the target vehicle and the mobile target.

[0046] Since the SLAM positioning algorithm is not easily interfered by other signals, it can improve the positioning accuracy of the target vehicle, thereby improving the accuracy of the entire vehicle collision risk. It can also be used in areas with weak satellite signals, improving the adaptability of the entire evaluation method. In addition, high-precision maps can provide rich geographic information, making the matching results more accurate, further improving the accuracy of the entire vehicle collision risk.

[0047] In this embodiment, optionally, a SLAM positioning algorithm is used to locate the position information of the target vehicle, including: performing denoising processing on the environmental perception information perceived by the target vehicle.

[0048] Specifically, the environmental perception information collected by the vehicle-mounted perception device may contain some noise. Therefore, after obtaining the environmental perception information of the vehicle-mounted perception device, conventional filtering algorithms such as Kalman filtering and Bayesian filtering can be used to denoise the environmental perception information to eliminate the noise in the environmental perception information and improve the positioning accuracy of the target vehicle. The specific methods and principles of filtering algorithms such as Kalman filtering and Bayesian filtering belong to the common knowledge in the field and will not be repeated here.

[0049] In this embodiment, optionally, a SLAM positioning algorithm is used to locate the position information of the target vehicle, including: extracting features of the environmental perception information perceived by the target vehicle.

[0050] Specifically, after removing the noise in the environmental perception information, the SURF algorithm, SIFT algorithm, etc. can be used to extract the environmental features around the target vehicle from the environmental perception information, such as feature points or feature lines with unique properties. The extracted environmental features are then associated and matched with the high-precision map, so that the current position information of the target vehicle can be accurately located through the complex high-precision map, improving the positioning accuracy and adaptability, and reducing the amount of data involved in the association matching, thereby improving the efficiency of risk assessment.

[0051] After accurately locating the position and posture information of the target vehicle, the relative position relationship between the moving targets around the target vehicle and the target vehicle can be determined based on the environmental perception information of the target vehicle. For example, the image captured by the camera can accurately locate the relative position relationship between the moving targets in the image, such as pedestrians, bicycles, etc., and the target vehicle. Combined with the position and posture information of the target vehicle, the position of the moving target can be accurately located, and then the moving trajectory of the moving target can be determined.

[0052] In this embodiment, optionally, evaluating the collision risk between the target vehicle and the moving target includes:

[0053] Fitting the movement curves of the target vehicle and the moving target according to the movement trajectories of the target vehicle and the moving target respectively;

[0054] Determining whether there is an intersection between the moving curves of the target vehicle and the moving target;

[0055] In response to the existence of an intersection point between the moving curves of the target vehicle and the moving target, determining the coordinates of the intersection point, and calculating the collision time according to the coordinates of the intersection point, the current position information of the target vehicle, and the vehicle speed;

[0056] A collision risk level between the target vehicle and the moving object is evaluated according to the collision time.

[0057] Specifically, first, curve fitting can be performed based on the moving trajectories of the target vehicle and the mobile target in the past to determine the moving curve of the moving target of the target vehicle. After that, the intersection of the moving curves corresponding to the mobile target and the target vehicle is obtained by combining the fitting functions corresponding to the moving curves corresponding to the mobile target and the target vehicle. Then, the distance between the target vehicle and the intersection can be calculated based on the position coordinates corresponding to the intersection and the current position of the target vehicle, and the time for the target vehicle to reach the intersection can be calculated based on the current speed of the target vehicle, that is, the collision time at which a collision may occur. Finally, the calculated collision time is matched with the time range corresponding to different collision risk levels to determine the collision risk level between the target vehicle and the mobile target. The collision risk is negatively correlated with the collision time. The smaller the collision time, the greater the collision risk level.

[0058] In this embodiment, optionally, it further includes:

[0059] The collision risk level of each mobile target is compared with the warning level, and warning information is generated according to the category of the mobile target whose collision risk level exceeds the warning level.

[0060] Specifically, after determining the collision risk level between the target vehicle and the moving target, the collision risk level can be compared with the pre-set warning level. If the collision risk level exceeds the warning level, a corresponding warning message is generated according to the type of the moving target to give an alarm. For example, if the moving target is a pedestrian, a voice prompt is generated to remind the pedestrian to avoid the moving target. If the moving target is a vehicle, a signal light message can be generated to remind the vehicle to pay attention.

[0061] like Figure 2 The system block diagram of the vehicle collision warning system based on high-precision map shown in the figure includes:

[0062] A map module, configured to obtain a high-precision map of the area where the target vehicle is located based on the current preliminary positioning information of the target vehicle;

[0063] A location module is configured to locate the position information of the target vehicle in real time using a SLAM positioning algorithm based on the environmental perception information around the target vehicle and the high-precision map;

[0064] A trajectory module, configured to determine the moving trajectory of the target vehicle and each moving target according to the real-time position information and perception information of the target vehicle;

[0065] The risk module is configured to evaluate the collision risk level between the target vehicle and each moving target according to the moving trajectories of the target vehicle and the moving targets.

[0066] Specifically, the early warning system includes a map module, a position module, a trajectory module and a risk module. Among them, the map module can retrieve a high-precision map of the area where the target vehicle is currently located based on the current preliminary positioning information of the target vehicle. The position module can perceive the environmental perception information around the target vehicle through the perception device carried by the target vehicle, and match the environmental perception information with the high-precision map using the SLAM positioning algorithm to determine the position information of the target vehicle. The trajectory module can combine the position information and perception information of the target vehicle to accurately locate the position information of the mobile targets around the target vehicle, thereby determining the moving trajectory of the target vehicle and the surrounding mobile targets. The risk module can evaluate the collision risk level between the target vehicle and the mobile target through the moving trajectory of the target vehicle and the mobile target.

[0067] In this embodiment, optionally, the location module includes: a noise processing unit configured to perform denoising processing on the environmental perception information perceived by the target vehicle.

[0068] Specifically, the location module includes a noise processing unit and a feature extraction unit. Among them, the noise processing unit can use conventional filtering algorithms such as Kalman filtering and Bayesian filtering to denoise the environmental perception information to eliminate the noise in the environmental perception information and improve the positioning accuracy of the target vehicle. The feature extraction unit can use SURF algorithm, SIFT algorithm, etc. to extract environmental features around the target vehicle from the environmental perception information, such as feature points or feature lines with unique properties, so that the location module can associate and match the extracted environmental features with the high-precision map, thereby accurately locating the current posture information of the target vehicle through complex high-precision maps, thereby improving positioning accuracy and adaptability. Moreover, the feature extraction unit can reduce the amount of data involved in association matching and improve risk assessment efficiency.

[0069] In this embodiment, optionally, the risk module includes:

[0070] A curve fitting unit, configured to fit the movement curves of the target vehicle and the moving target according to the movement trajectories of the target vehicle and the moving target respectively;

[0071] a collision analysis unit configured to determine whether there is an intersection between the moving curves of the target vehicle and the moving target;

[0072] In response to the existence of an intersection point between the moving curves of the target vehicle and the moving target, determining the coordinates of the intersection point, and calculating the collision time according to the coordinates of the intersection point, the current position information of the target vehicle, and the vehicle speed;

[0073] The risk assessment unit is configured to assess a collision risk level between the target vehicle and the moving target according to the collision time.

[0074] Specifically, the risk module includes a curve fitting unit, a collision risk unit and a risk assessment unit. Among them, the curve fitting unit can perform curve fitting according to the moving trajectories of the target vehicle and the mobile target in the past to determine the moving curve of the mobile target of the target vehicle. The collision analysis unit can combine the fitting functions corresponding to the moving curves corresponding to the mobile target and the target vehicle to obtain the intersection of the moving curves corresponding to the mobile target and the target vehicle. The collision analysis unit can calculate the distance between the target vehicle and the intersection according to the position coordinates corresponding to the intersection and the current position of the target vehicle, and calculate the collision time at which the target vehicle may collide according to the current speed of the target vehicle. The risk assessment unit can match the calculated collision time with the time range corresponding to different collision risk levels, thereby determining the collision risk level between the target vehicle and the mobile target.

[0075] In this embodiment, optionally, it further includes:

[0076] The warning module is configured to compare the collision risk level of each mobile target with the warning level, and generate warning information according to the category of the mobile target whose collision risk level exceeds the warning level.

[0077] Specifically, the warning system also includes a warning module, which can compare the collision risk level with a pre-set warning level. If the collision risk level exceeds the warning level, corresponding warning information is generated according to the category of the mobile target for alarm.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A vehicle collision warning method based on high-precision maps, characterized in that: include: Obtain a high-precision map of the area where the target vehicle is located based on the current preliminary positioning information of the target vehicle; Based on the environmental perception information around the target vehicle and the high-precision map, a SLAM positioning algorithm is used to locate the position information of the target vehicle in real time; Determine the moving trajectory of the target vehicle and each moving target according to the real-time position information and perception information of the target vehicle; According to the moving trajectories of the target vehicle and the moving targets, the collision risk level between the target vehicle and each moving target is evaluated.

2. The vehicle collision warning method according to claim 1, characterized in that: The SLAM positioning algorithm is used to locate the position information of the target vehicle, including: De-noising is performed on the environmental perception information perceived by the target vehicle.

3. The vehicle collision warning method according to claim 1, characterized in that: The SLAM positioning algorithm is used to locate the position information of the target vehicle, including: Feature extraction is performed on the environmental perception information perceived by the target vehicle.

4. The vehicle collision warning method according to claim 1, characterized in that: Assessing the collision risk between the target vehicle and the moving target, including: Fitting the movement curves of the target vehicle and the moving target according to the movement trajectories of the target vehicle and the moving target respectively; Determining whether there is an intersection between the moving curves of the target vehicle and the moving target; In response to the existence of an intersection point between the moving curves of the target vehicle and the moving target, determining the coordinates of the intersection point, and calculating the collision time according to the coordinates of the intersection point, the current position information of the target vehicle, and the vehicle speed; A collision risk level between the target vehicle and the moving object is evaluated according to the collision time.

5. The vehicle collision warning method according to claim 1, characterized in that: Also includes: The collision risk level of each mobile target is compared with the warning level, and warning information is generated according to the category of the mobile target whose collision risk level exceeds the warning level.

6. A vehicle collision warning system based on high-precision maps, characterized in that: include: A map module, configured to obtain a high-precision map of the area where the target vehicle is located based on the current preliminary positioning information of the target vehicle; A location module is configured to locate the position information of the target vehicle in real time using a SLAM positioning algorithm based on the environmental perception information around the target vehicle and the high-precision map; A trajectory module, configured to determine the moving trajectory of the target vehicle and each moving target according to the real-time position information and perception information of the target vehicle; The risk module is configured to evaluate the collision risk level between the target vehicle and each moving target according to the moving trajectories of the target vehicle and the moving targets.

7. The vehicle collision warning system according to claim 6, characterized in that: The position module includes: a noise processing unit configured to perform denoising on the environmental perception information perceived by the target vehicle.

8. The vehicle collision warning system according to claim 6, characterized in that: The location module includes: a feature extraction unit configured to extract features from the environmental perception information perceived by the target vehicle.

9. The vehicle collision warning system according to claim 6, characterized in that: The risk module includes: A curve fitting unit, configured to fit the movement curves of the target vehicle and the moving target according to the movement trajectories of the target vehicle and the moving target respectively; a collision analysis unit configured to determine whether there is an intersection between the moving curves of the target vehicle and the moving target; In response to the existence of an intersection point between the moving curves of the target vehicle and the moving target, determining the coordinates of the intersection point, and calculating the collision time according to the coordinates of the intersection point, the current position information of the target vehicle, and the vehicle speed; The risk assessment unit is configured to assess a collision risk level between the target vehicle and the moving target according to the collision time.

10. The vehicle collision warning system according to claim 6, characterized in that: Also includes: The warning module is configured to compare the collision risk level of each mobile target with the warning level, and generate warning information according to the category of the mobile target whose collision risk level exceeds the warning level.

Images