CN115257816A - Obstacle identification method and system for automatic driving - Google Patents
Obstacle identification method and system for automatic driving Download PDFInfo
- Publication number
- CN115257816A CN115257816A CN202211028522.3A CN202211028522A CN115257816A CN 115257816 A CN115257816 A CN 115257816A CN 202211028522 A CN202211028522 A CN 202211028522A CN 115257816 A CN115257816 A CN 115257816A
- Authority
- CN
- China
- Prior art keywords
- obstacle
- point clouds
- road
- whole
- obstacles
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0015—Planning or execution of driving tasks specially adapted for safety
- B60W60/0018—Planning or execution of driving tasks specially adapted for safety by employing degraded modes, e.g. reducing speed, in response to suboptimal conditions
- B60W60/00184—Planning or execution of driving tasks specially adapted for safety by employing degraded modes, e.g. reducing speed, in response to suboptimal conditions related to infrastructure
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/001—Planning or execution of driving tasks
- B60W60/0027—Planning or execution of driving tasks using trajectory prediction for other traffic participants
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an automatic driving obstacle identification method and system, wherein the method comprises the steps of detecting obstacles, generating point clouds and fusing the point clouds into an integral obstacle; forming a predicted track according to the information of the whole obstacle; and planning a road in advance and avoiding a mode to avoid the obstacle according to the information and the predicted track of the whole obstacle. The problem that when the automatic driving vehicle flows into the auxiliary road from the main road, flows into the main road from the auxiliary road and runs through the ramp junction on the main road, due to the influence of obstacles such as roadside greening or guardrails, the running condition of the vehicle about to flow into the road cannot be detected is solved.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to an obstacle identification method and system for automatic driving.
Background
With the continuous increase of the intelligent degree of automobiles, a large number of automobile models with automatic driving auxiliary functions mainly used for urban expressways and expressways appear in the market. Autopilot is becoming a major trend in the development of the automotive industry. However, mass production cost is considered, the vehicle sensor is relatively cheap, the detection effect is poor, obstacles cannot be detected under various conditions, and potential safety hazards are formed. In the scenes of urban expressways and expressways, it is the most common scene that an automatic driving vehicle enters a secondary road from a main road and enters the main road from the secondary road or runs through a turn road junction on the main road, but the problem that the running condition of the vehicle about to enter the road cannot be detected due to the influence of obstacles such as roadside greening or guardrails is one of the potential safety hazards to be solved urgently. The vehicle with the automatic driving assistance function is limited by the sensing capability of the sensor, most of the vehicles with the automatic driving assistance function in mass production at present adopt a strategy that the vehicles can bypass and leave a lane space to bypass when meeting the conditions that the main road is converged into the auxiliary road, the auxiliary road is converged into the main road and the side lanes are seriously shielded when the vehicles run on the main road and pass through the ramp, and the vehicles cannot bypass and prompt a driver to take over.
However, the strategy also has the disadvantages that the number of urban vehicles is increased, road extension cannot catch up with the increased speed of the vehicles, the traffic flow on urban expressways is larger, the number of ramp openings is more, no lane can avoid in advance in driving, and frequent take-over is caused to influence the driving experience. Too frequent takeover applications can also cause the driver to ignore the takeover request, creating a significant safety hazard. Therefore, it is necessary to solve the problem of road vehicles gathering into the road caused by road greening or guardrail obstacles.
And in recent years, 4D millimeter wave radars introduced by some manufacturers have made it possible to solve the above problems. Compared with the traditional millimeter wave radar on mass-produced vehicles at the present stage, the 4D millimeter wave radar has a qualitative leap in the detection capability of the obstacle under the condition of low cost improvement. The traditional millimeter wave radar has a small obstacle detection range, and the output result is only on an XY plane without high-speed discrimination. This makes it difficult to select the vehicles actually traveling on the road from the plurality of obstacle points.
Disclosure of Invention
The invention aims to provide an automatic driving obstacle identification method and system, which solve the problem that the driving condition of a vehicle about to merge into a road cannot be detected due to the influence of obstacles such as roadside greening or guardrails when an automatic driving vehicle merges into a secondary road from a main road, merges into the main road from the secondary road and runs through a ramp junction on the main road.
In order to achieve the above object, the present invention provides an obstacle recognition method for automatic driving, comprising:
detecting the obstacles to generate point clouds and fusing the point clouds into an integral obstacle;
forming a predicted track according to the information of the whole obstacle;
and planning a road in advance and avoiding a mode to avoid the obstacle according to the information and the predicted track of the whole obstacle.
Further, the detecting the obstacle to generate the point cloud and fuse the point cloud into an integral obstacle, including,
detecting peripheral obstacles, processing coordinates of the obstacles to generate point clouds, screening and marking the point clouds;
and fusing the marked point clouds to form an integral barrier.
Further, detecting peripheral obstacles, processing the coordinates of the obstacles to generate point clouds, screening and marking the point clouds, including,
detecting peripheral obstacles through a 4D millimeter wave radar, processing coordinates of the obstacles to generate point clouds, and sending the point clouds to a fusion algorithm program;
screening out point clouds with the height higher than the preset height of the horizon by a fusion algorithm program;
comparing the screened point cloud with the horizontal range of the adjacent road in the high-precision map, and secondarily screening the point cloud of which the horizontal coordinate is positioned in the adjacent road;
and marking point clouds of which the relative speed with the vehicle is lower than the actual speed of the vehicle in the point clouds subjected to secondary screening.
Further, the marked point clouds are fused to form an integral barrier, which comprises,
and fusing the marked point clouds, and fusing the point clouds of which the coordinates are in a horizontal and vertical preset range and the speed is the same into an integral barrier.
Further, said forming a predicted trajectory based on said information of said overall obstacle comprises,
calculating the center and relative speed of the whole barrier;
and calculating the acceleration of the whole obstacle in continuous preset frames according to the center of the whole obstacle and the relative speed of the whole obstacle and the vehicle, and forming a predicted track along the center line of the road of the map.
Furthermore, according to the information and the predicted track of the whole obstacle, a road is planned in advance and an avoidance mode is carried out to avoid the obstacle, comprising,
and when the obstacle and the vehicle have collision risks, selecting a planned road and an avoidance mode in advance to avoid the obstacle according to the information and the predicted track of the whole obstacle based on the actual condition of the road.
The invention also provides an automatic driving obstacle identification system, which comprises a fusion unit, a track prediction unit and a planning control unit,
the fusion unit is used for detecting the obstacles to generate point clouds and fusing the point clouds into an integral obstacle;
the predicted track unit is used for forming a predicted track according to the information of the whole obstacle;
and the planning control unit is used for planning a road in advance and avoiding a mode to avoid the obstacle according to the information and the predicted track of the whole obstacle.
Further, the fusion unit is used for detecting the obstacle to generate a point cloud and fusing the point cloud into a whole obstacle, and comprises,
the fusion unit is used for detecting peripheral obstacles, processing the coordinates of the obstacles to generate point clouds, and screening and marking the point clouds;
the fusion unit is also used for fusing the marked point clouds to form an integral barrier.
Further, the predicted trajectory unit, configured to form a predicted trajectory according to the information of the whole obstacle, includes,
the predicted track unit is used for calculating the center of the whole obstacle and the relative speed of the whole obstacle with the vehicle;
the predicted track unit is used for measuring and calculating the acceleration of the whole obstacle in continuous preset frames according to the center and the relative speed of the vehicle, and forming a predicted track along the center line of a road of a map;
and the predicted track unit is used for sending the predicted track and the whole obstacle information to the planning control unit.
Further, the planning control unit is used for planning a road in advance and avoiding the obstacle in an avoidance mode according to the information and the predicted track of the whole obstacle, and comprises,
and the planning control unit is used for selecting a planned road and an avoidance mode in advance to avoid the obstacle based on the actual condition of the road when the obstacle and the vehicle have collision risk according to the information and the predicted track of the whole obstacle.
The invention has the technical effects and advantages that: 1. the invention solves the problem that the driving condition of the vehicle about to merge into the road can not be detected due to the influence of obstacles such as roadside greening or guardrails when the automatic driving vehicle merges into the auxiliary road from the main road, merges into the main road from the auxiliary road and runs through the ramp junction on the main road. The safety and the driving experience of the automatic driving auxiliary function are improved;
2. the 4D millimeter wave radar is used for replacing a conventional millimeter wave radar or a panoramic camera which is common at the present stage, so that the detection range is wider, the obstacles can be distinguished in height, and a vehicle can distinguish greening or guardrails conveniently;
3. the method and the device can greatly reduce the number of times of taking over requests of the automatic driving vehicle in the congestion environment, and improve the driving experience. The invention has the advantages of small change of the existing mass production automatic driving vehicle body system scheme, small cost increase and convenient realization and mass production.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart of an obstacle identification method for autonomous driving in accordance with an embodiment of the present invention;
FIG. 2 is a top plan view of the vehicle of the present embodiment of the invention;
FIG. 3 is a flowchart illustrating an embodiment of an obstacle recognition method for automatic driving;
fig. 4 is a schematic structural diagram of an automatic driving obstacle recognition system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to solve the defects of the prior art, the invention discloses an obstacle identification method for automatic driving, which comprises the following steps as shown in figure 1:
detecting the obstacles to generate point clouds and fusing the point clouds into an integral obstacle;
forming a predicted track according to the information of the whole obstacle;
and planning a road and an avoidance mode in advance to avoid the obstacle according to the information and the predicted track of the whole obstacle.
Specifically, as shown in fig. 2, two 4D millimeter wave radars are installed in front of and behind the vehicle.
As shown in fig. 3, after the 4D millimeter wave radar is used to detect and scan the obstacles around the vehicle, the coordinates of the obstacles are processed to generate a point cloud, and the point cloud of the obstacles is sent to the fusion algorithm program in the domain controller for intelligent driving. According to the technical point, the 4D millimeter wave radar is used for replacing a conventional millimeter wave radar or a panoramic camera which is common at the present stage, so that not only is the detection range wider, but also the obstacles can be distinguished in height, and the obstacles are further towards a low-resolution laser radar; and XYZ's three-dimensional obstacle position output mode, the vehicle of being convenient for can distinguish afforestation or guardrail. And the installation position of the 4D millimeter wave radar may be appropriately changed according to the vehicle model.
The fusion algorithm program firstly screens the obstacle point cloud with the height higher than the preset height (1 m to 2 m) of the horizon for the first time, compares the obstacle point cloud screened for the first time with the horizontal range of the adjacent ramp or the main and auxiliary roads in the high-precision map, and then screens the obstacle point cloud with the horizontal coordinate in the adjacent ramp or the main and auxiliary roads for the second time. The technical point is that the obstacles are primarily screened in height through primary screening and secondary screening, and the matching of the road range is formed by using high-precision map data.
And marking the obstacle point clouds of which the relative speed with the vehicle is lower than the actual speed of the vehicle in the obstacle point clouds subjected to secondary screening. The present technology excludes other driving vehicles by comparing the relative speed between the obstacle and the vehicle with the vehicle speed of the vehicle.
Fusing the marked obstacle point clouds: and fusing the obstacle point clouds with the coordinates in a horizontal and longitudinal preset range (4m × 2m) and the same speed to form an integral obstacle. According to the technology, the point clouds of the obstacles with the same speed in the range are fused, so that the trajectory planning of subsequent vehicles is facilitated.
The relative speed with the vehicle and the center thereof are estimated for the whole obstacle, and the acceleration of the whole obstacle is measured and calculated in a continuous predetermined number of frames (two frames). And then forming a predicted track along the center line of the road in the high-precision map for matching the driving route.
And forming a planning control algorithm for decision avoidance according to the predicted track and the information of the whole obstacle, and planning a road in advance according to the actual road condition or selecting ways of changing lanes in advance, decelerating avoidance or accelerating overtaking and the like to avoid the obstacle when the whole obstacle and the vehicle have collision risks.
The invention also discloses an automatic driving obstacle recognition system, which comprises a fusion unit, a prediction track unit and a planning control unit as shown in figure 4,
and the fusion unit is used for detecting the obstacles, generating point clouds and fusing the point clouds into an integral obstacle.
Specifically, the fusion unit is used for detecting peripheral obstacles, processing detected obstacle coordinates to generate point clouds, and screening and marking the obstacle point clouds. And then the marked obstacle point clouds are fused by a fusion unit to form an integral obstacle.
And a predicted trajectory unit for forming a predicted trajectory from the information of the entire obstacle.
Specifically, the predicted trajectory unit is used for calculating the center of the whole obstacle and the relative speed of the vehicle according to the whole obstacle.
The predicted track unit is used for measuring and calculating the acceleration of the whole obstacle in continuous preset frames according to the center and the relative speed of the vehicle, and forming a predicted track along the center line of the road of the map; the predicted track unit is used for sending the predicted track and the whole obstacle information to the planning control unit.
And the planning control unit is used for planning a road in advance and avoiding the obstacle in an avoidance mode according to the information of the whole obstacle and the predicted track.
Specifically, the planning control unit generates a planning control algorithm according to the information of the whole obstacle and the predicted track, and selects a planned road and an avoidance mode in advance to avoid the obstacle based on the actual condition of the road when the obstacle and the vehicle have a collision risk.
With regard to the system in the above embodiment, the specific manner in which each unit module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated herein.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments or portions thereof without departing from the spirit and scope of the invention.
Claims (10)
1. An obstacle recognition method for automatic driving, comprising,
detecting the obstacles to generate point clouds and fusing the point clouds into an integral obstacle;
forming a predicted track according to the information of the whole obstacle;
and planning a road in advance and avoiding a mode to avoid the obstacle according to the information and the predicted track of the whole obstacle.
2. The method of claim 1, wherein detecting obstacles, generating point clouds and fusing them into a whole obstacle comprises,
detecting peripheral obstacles, processing coordinates of the obstacles to generate point clouds, screening and marking the point clouds;
and fusing the marked point clouds to form an integral barrier.
3. The method of claim 2, wherein detecting peripheral obstacles, processing coordinates of the obstacles to generate a point cloud, and screening and labeling the point cloud, comprises,
detecting peripheral obstacles through a 4D millimeter wave radar, processing coordinates of the obstacles to generate point clouds, and sending the point clouds to a fusion algorithm program;
screening out point clouds with the height higher than the preset height of the horizon by a fusion algorithm program;
comparing the screened point cloud with the horizontal range of the adjacent road in the high-precision map, and secondarily screening the point cloud of which the horizontal coordinate is positioned in the adjacent road;
and marking point clouds of which the relative speed with the vehicle is lower than the actual speed of the vehicle in the point clouds subjected to secondary screening.
4. The method according to claim 2 or 3, wherein the marked point clouds are fused to form an integral obstacle, comprising,
and fusing the marked point clouds, and fusing the point clouds of which the coordinates are in a horizontal and vertical preset range and the speed is the same into an integral barrier.
5. The method of claim 1, wherein said forming a predicted trajectory based on information of said overall obstacle comprises,
calculating the center and relative speed of the whole barrier;
and calculating the acceleration of the whole obstacle in continuous preset frames according to the center of the whole obstacle and the relative speed of the vehicle, and forming a predicted track along the center line of the road of the map.
6. The method and system for identifying the obstacle as claimed in claim 1 or 5, wherein a road and an avoidance mode are planned in advance to avoid the obstacle according to the information of the whole obstacle and the predicted track, including,
and when the obstacle has collision risk with the vehicle, selecting a planned road and an avoidance mode in advance to avoid the obstacle according to the information and the predicted track of the whole obstacle based on the actual condition of the road.
7. An automatic driving obstacle recognition system is characterized by comprising a fusion unit, a prediction track unit and a planning control unit,
the fusion unit is used for detecting the obstacles to generate point clouds and fusing the point clouds into an integral obstacle;
the predicted track unit is used for forming a predicted track according to the information of the whole obstacle;
and the planning control unit is used for planning a road in advance and avoiding a mode to avoid the obstacle according to the information and the predicted track of the whole obstacle.
8. The automatic driving obstacle recognition system according to claim 7, wherein the fusion unit is used for detecting obstacles to generate point clouds and fusing the point clouds into a whole obstacle, and comprises,
the fusion unit is used for detecting peripheral obstacles, processing the coordinates of the obstacles to generate point clouds, and screening and marking the point clouds;
the fusion unit is also used for fusing the marked point clouds to form an integral barrier.
9. The automatic driving obstacle recognition system according to claim 7 or 8, wherein the predicted trajectory unit for forming a predicted trajectory based on information of the entire obstacle includes,
the predicted track unit is used for calculating the center of the whole obstacle and the relative speed of the whole obstacle with the vehicle;
the predicted track unit is used for measuring and calculating the acceleration of the whole obstacle in continuous preset frames according to the center and the relative speed of the vehicle, and forming a predicted track along the center line of a road of a map;
the predicted track unit is used for sending the predicted track and the whole obstacle information to the planning control unit.
10. The automatic driving obstacle recognition system according to claim 7, wherein the planning control unit is configured to plan a road and avoid an obstacle in advance based on the information of the entire obstacle and the predicted trajectory, and comprises,
and the planning control unit is used for selecting a planned road and an avoidance mode in advance to avoid the obstacle based on the actual condition of the road when the obstacle and the vehicle have collision risk according to the information and the predicted track of the whole obstacle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211028522.3A CN115257816A (en) | 2022-08-25 | 2022-08-25 | Obstacle identification method and system for automatic driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211028522.3A CN115257816A (en) | 2022-08-25 | 2022-08-25 | Obstacle identification method and system for automatic driving |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115257816A true CN115257816A (en) | 2022-11-01 |
Family
ID=83753689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211028522.3A Pending CN115257816A (en) | 2022-08-25 | 2022-08-25 | Obstacle identification method and system for automatic driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115257816A (en) |
-
2022
- 2022-08-25 CN CN202211028522.3A patent/CN115257816A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102347036B1 (en) | Track prediction method and device for obstacle at junction | |
| CN110675656B (en) | Intelligent vehicle lane change early warning method based on instantaneous risk identification | |
| US7209051B2 (en) | Intersection assistance system and method | |
| Zhao et al. | On-road vehicle trajectory collection and scene-based lane change analysis: Part i | |
| CN111540237A (en) | Method for automatically generating vehicle safety driving guarantee scheme based on multi-data fusion | |
| CN105620489A (en) | Driving assistance system and real-time warning and prompting method for vehicle | |
| US20220234615A1 (en) | In-vehicle device and driving assist method | |
| TWI493513B (en) | Lane departure warning system and lane identification apparatus and related method thereof | |
| CN110807412B (en) | Vehicle laser positioning method, vehicle-mounted equipment and storage medium | |
| US11914041B2 (en) | Detection device and detection system | |
| CN111252066A (en) | Emergency braking control method and device, vehicle and storage medium | |
| JP7289760B2 (en) | electronic controller | |
| CN113112830B (en) | Signal control intersection emptying method and system based on laser radar and track prediction | |
| CN112606831A (en) | Anti-collision warning information external interaction method and system for passenger car | |
| CN114454878B (en) | Method and device for determining vehicle speed control model training sample | |
| CN114056324A (en) | Parking space identification method and system based on data fusion | |
| CN111932884A (en) | Traffic jam recognition method and device, vehicle and storage medium | |
| CN116052472A (en) | Vehicle-mounted V2X collision early warning method based on road perception information fusion | |
| CN113432615A (en) | Detection method and system based on multi-sensor fusion drivable area and vehicle | |
| CN115257816A (en) | Obstacle identification method and system for automatic driving | |
| WO2022050006A1 (en) | Image processing device | |
| US11667304B2 (en) | Enhanced vehicle operation | |
| CN113432616A (en) | Map making and/or processing method, map storage device and path planning method | |
| KR20220136679A (en) | Apparatus and method for controlling driving of vehicle | |
| CN113386771A (en) | Road model generation method and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |