CN115973197A - Lane planning method and device, electronic equipment and readable storage medium - Google Patents
Lane planning method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN115973197A CN115973197A CN202310273726.1A CN202310273726A CN115973197A CN 115973197 A CN115973197 A CN 115973197A CN 202310273726 A CN202310273726 A CN 202310273726A CN 115973197 A CN115973197 A CN 115973197A
- Authority
- CN
- China
- Prior art keywords
- lane
- global
- path
- road
- planning
- 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.)
- Granted
Links
Images
Landscapes
- Navigation (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a lane planning method, a lane planning device, electronic equipment and a readable storage medium, which are used for segmenting according to intersections on the basis of global road-level planning, performing local lane-level planning according to real-time road conditions and then controlling an intelligent agent to follow a planned path form. The technical scheme provided by the invention can select the optimal driving path in real time, so that the intelligent agent can flexibly deal with different conditions, and the intelligence of the whole system is improved.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a lane planning method, a lane planning device, electronic equipment and a readable storage medium.
Background
With the development of an automatic driving technology, a navigation path provided by a traditional road-level planning cannot serve an intelligent driving system well, for example, if the road-level planning path cannot guide the intelligent driving system to reach a required target lane at an intersection, for example, a left-turn lane is at the rightmost intersection, and other complex intersection lane distribution conditions, if the intelligent driving system cannot reach the target lane in time, a subsequent global navigation path will be invalid, recalculation is required, and driving cost is increased and riding experience is affected. At some busy intersections of some cities, the red light is as long as 180 seconds, and the traditional road-level planning cannot update the path planning result by utilizing real-time dynamic data information.
Disclosure of Invention
In order to solve the problems, the invention provides an automatic driving lane planning method based on a high-precision map, which comprises the following steps:
s10, global road level planning: calculating global path cost according to coordinate information of a starting point, a passing point and a terminal point and the topological relation among roads provided by the high-precision map, and taking a track with the lowest global path cost as a global road-level track;
s20, local lane level planning: segmenting the global road level track according to intersections of the global road level track path; calculating local costs from the temporary starting point to the N temporary end points by taking the current lane as a temporary starting point and the N lanes at the next intersection as temporary end points; taking the track with the lowest local cost as a local lane-level track;
s30, judging whether the intelligent agent can drive from the current lane to the target lane according to the local lane level track,
when the intelligent agent can drive to the target lane according to the judgment result, planning a driving path of the intelligent agent according to the action attribute to be generated in the target lane and the time attribute of the target lane;
and when the intelligent body cannot drive to the target lane, controlling the intelligent body to drive to the virtual lane connected with the current lane, and performing global road-level planning by taking the next road connected with the virtual lane as a starting point.
Because the navigation path planning process in the related art lacks intelligence, the technical scheme provided by the invention can effectively increase the intelligence of an intelligent system by introducing and applying a high-precision graph based on dynamic data provided by a V2X technology. Specifically, based on the result of global road-level planning, segmentation is performed according to the intersection information, and then local lane-level planning is performed on the local road from the current position to the next intersection, so that on one hand, unnecessary calculation time can be reduced, and on the other hand, the requirement of an intelligent driving system on real-time performance can be met.
Further, planning a driving path of the intelligent agent according to the action attribute about to occur in the target lane and the time attribute of the target lane; the method comprises the following steps:
when the action attribute is right turn, planning a driving path of the intelligent agent according to the straight-going waiting time of the next intersection;
when the action attribute is not right turn, planning a driving path of the intelligent agent according to the time attribute;
and the time attribute is the waiting time of the traffic lights at the intersection corresponding to the target lane.
Further, planning a running path of the intelligent agent according to the straight-going waiting time of the next intersection; the method comprises the following steps:
if the waiting time of straight going is 0, the next road reached by straight going is taken as the starting point to carry out global road-level planning, and a new global path cost C is obtained through calculation new1 ；
When C is present new1 +C 1 ＜C old +C extra Then the target lane is changed into a lane with straight-going attribute, and the new global path cost C is used new1 Updating the navigation path;
when C is present new1 +C 1 ≥C old +C extra Taking the current global road level track as a subsequent navigation path;
wherein, C new1 Carrying out global road-level planning for the next road which is reached in a straight line as a starting point to obtain a new global path cost; c 1 The cost is the straight line cost; c old Global path cost for a current global road-level trajectory; c extra Is a constant;
and if the straight-going waiting time is more than 0, taking the current global road level track as a subsequent navigation path.
Further, planning a driving path of the intelligent agent according to the time attribute; the method comprises the following steps:
if the time attribute is determined to exceed the time threshold,
when the action attribute is left turn, judging whether a local lane-level track of the target lane as a straight lane can be planned or not to obtain a first judgment result, and planning a driving path of the intelligent agent according to the first judgment result;
when the action attribute is straight, judging whether a local lane-level track with a right turn/left turn target lane can be planned or not to obtain a second judgment result, and planning a running path of the intelligent agent according to the second judgment result;
and when the action attribute is turning around, taking the current global road-level track as a subsequent navigation path.
Further, if the first judgment result is yes, performing global road-level planning by taking the next road reached in the straight line as a starting point, and calculating to obtain a new global path cost C new1 ；
The step of planning the driving path of the intelligent agent according to the first judgment result comprises the following steps:
when C is new1 +C 1 +C t1 ＜C old +C t3 Then the target lane is changed into a lane with straight-going attribute, and the new global path cost C is used new1 Updating the navigation path;
when C is new1 +C 1 +C t1 ≥C old +C t3 Then, taking the current global road level track as a subsequent navigation path;
wherein, C new1 Carrying out global road-level planning for the next road which is reached in a straight line as a starting point to obtain a new global path cost; c 1 The cost is the straight line cost; c t1 Waiting for a time penalty for straight-ahead; c old Global path cost for a current global road-level trajectory; c t3 Latency penalty for left turn;
and if the first judgment result is negative, taking the current global road level track as a subsequent navigation path.
Further, if the second judgment result is yes, the next road which is reached by left turning/right turning is taken as a starting point to perform global road-level planning, and a new global path cost C is obtained through calculation new2 /C new3 ；
Planning the driving path of the agent according to the second judgment result comprises the following steps:
when C is new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 ＜C new3 +C 3 +C t3 When the current is over; the target lane is changed to a lane containing a right turn attribute and based on the new global path cost C new2 Updating the navigation path;
when C is present new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 Greater than or equal to C new3 +C 3 +C t3 When the current is over; the target lane is changed to a lane containing a left turn attribute and based on the new global path cost C new3 Updating the navigation path;
when not satisfying C new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 Then, taking the current global road level track as a subsequent navigation path;
wherein, C new2 Performing global road-level planning for the next road reached by the right turn as a starting point to obtain a new global path cost; c new3 Carrying out global road-level planning for the next road reached by the left turn as a starting point to obtain a new global path cost; c 2 A right turn cost; c 3 A left turn penalty; c t1 Waiting for a time penalty for straight-ahead; c t2 Latency penalty for right turn; c t3 Latency penalty for left turn; c old Global path cost for a current global road-level trajectory;
and if the second judgment result is negative, taking the current global road level track as a subsequent navigation path.
Further, planning a driving path of the intelligent agent according to the time attribute; the method comprises the following steps: and if the time attribute is judged not to exceed the time threshold, taking the current global road level track as a subsequent navigation path.
This technical scheme provides an autopilot lane planning device based on high-precision map, includes:
a global road level planning module; the global path cost is calculated according to the coordinate information of the starting point, the passing point and the end point and the topological relation among the roads provided by the high-precision map, and the track with the lowest global path cost is used as a global road-level track;
a local lane level planning module; for following a global road-level track path the intersection segments the global road-level track; taking the current lane as a temporary starting point and the N lanes of the next intersection as temporary end points, and calculating the lane change cost from the temporary starting point to the N temporary end points; taking the track with the lowest lane change cost as a local lane-level track;
an automatic driving decision module; for controlling the formal path of the agent according to the local lane-level trajectory.
The technical scheme provides an electronic device which comprises a processor, a memory and a program or an instruction which is stored on the memory and can run on the processor, wherein the program or the instruction realizes the steps of the technical scheme method when being executed by the processor.
The technical solution provides a readable storage medium, on which a program or an instruction is stored, and when the program or the instruction is executed by a processor, the steps of the technical solution method are implemented.
Wherein, the processor is a processor in the electronic device in the above technical solution. Readable storage media include computer readable storage media such as Read-Only Memory (ROM), random Access Memory (RAM), magnetic or optical disk, and so on.
The local lane-level planning set by the invention can better serve automatic driving, the traditional global planning module is split, the traditional global road-level planning module is reserved, and the track and the intersection information on the track output by the traditional global planning module are used as prior information to guide the lower local lane-level planning module to carry out lane-level planning. And the local lane-level planning guides an upper-layer road-level planning module to carry out new global road-level planning according to real-time dynamic data information provided by the high-precision map, and judges whether a new global planning track is used or not. The technical scheme has at least one of the following advantages:
1. a local lane-level planning module is provided, and a lane-level planning track to the optimal target lane can be provided for the intelligent system in real time by taking the target lane at the intersection as a temporary terminal.
2. Real-time signal lamp data provided by a high-precision map are fully utilized and subjected to prediction and judgment, and a road-level planning system is reasonably guided to plan a new global track. The intelligent system can flexibly deal with different conditions, the red light waiting time of the intelligent system is reduced, and the intelligence of the intelligent system is improved.
3. The global road level planning module operates in a triggering mode, the local lane level planning module operates in real time, and the real-time requirement of the intelligent system is met by using fewer computing resources.
4. The global path cost is stored in a segmented mode, and the cost is updated in real time in the running process of the intelligent system, so that the calculation of an old global planning track is reduced, the calculation resources are saved, and the real-time performance of the system is improved.
Drawings
Fig. 1 is a schematic diagram of virtual lane planning in an embodiment of the present invention.
Fig. 2 is a schematic diagram of lane attributes according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a lane planning apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. 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.
Example 1
In the field of automatic driving technology, planning of a driving path of an intelligent agent often fails to take into account real-time road conditions, and therefore driving cost is likely to increase.
Based on the above reasons, the present embodiment provides an automatic driving lane planning method based on a high-precision map, including: s10, global road level planning: calculating global path cost according to coordinate information of a starting point, a passing point and a terminal point and the topological relation among roads provided by the high-precision map, and taking a track with the lowest global path cost as a global road-level track; s20, local lane level planning: segmenting the global road-level track according to the intersections of the global road-level track path; calculating local costs from the temporary starting point to the N temporary end points by taking the current lane as a temporary starting point and the N lanes at the next intersection as temporary end points; taking the track with the lowest local cost as a local lane level track; s30, judging whether the intelligent body can drive to a target lane from a current lane according to the local lane-level track, and planning a driving path of the intelligent body according to the action attribute to be generated in the target lane and the time attribute of the target lane when the intelligent body can drive to the target lane according to the judgment result; and when the intelligent agent cannot drive to the target lane, controlling the intelligent agent to drive to the virtual lane connected with the current lane, and performing global road-level planning by taking the next road connected with the virtual lane as a starting point.
In the embodiment, the global road-level planning is performed based on coordinate information of a starting point, a passing point and a terminal point and a topological relation among roads provided by a high-precision map; generally, a planning algorithm is adopted to calculate the global path cost, the planning algorithm includes an a-star algorithm, a Dijkstra algorithm and the like, and the global path cost includes road cost, intersection action cost and the like. Wherein the road cost is calculated by referring to the length and congestion degree of the road, and the movement of the intersectionThe operation cost includes, for example, the cost of straight movement, left turn, right turn, and turning around. The global path cost is stored in a segmented mode in the calculation process and is updated in real time according to the current state of the intelligent agent. For example: after the agent finishes driving the road A, the current global path cost is updated to C 0 -C A-road Wherein, C 0 For global path costs before driving the A road, C A-road A road cost of A; if the agent finishes the straight-going action at the intersection, the current global path cost is updated to C 0 -C 1 Wherein, C 0 For global path costs before straight-ahead action, C 1 Is a straight line cost.
And obtaining a global road-level track with the lowest global path cost according to the algorithm, and segmenting according to the global road-level track obtained by planning and each intersection of the track path. Generally, the next intersection to be reached by the intelligent agent comprises a plurality of lanes, the lane where the intelligent agent is located at present is used as a starting point, the N lanes of the next intersection are used as temporary end points, and a planning algorithm is adopted to calculate local cost according to the topological relation between the lanes; and selecting the track with the lowest local cost as the local lane-level track. The planning algorithm includes, for example, an a-algorithm, a Dijkstra algorithm, etc.; the local cost is obtained by calculating lane change cost, lane attribute cost, lane crowding degree cost and the like. The lane change cost is the cost required by the intelligent agent to travel to the target lane, and includes the number of lane changes. Lane attributes generally include single lane attributes and multi-lane attributes, see fig. 2, a single lane attribute can only perform one action, such as straight-going; the multi-lane attribute may perform at least two actions, such as straight and right turns. And the cost of the traffic lane congestion degree is calculated according to the congestion degree of the real-time road condition.
After the local lane-level track is obtained, whether the intelligent object can drive from the current lane to the target lane is further judged; and planning the driving path of the intelligent agent according to different judgment results. The specific judgment process is as follows: when the intelligent agent can drive to the target lane according to the judgment result, planning a driving path of the intelligent agent according to the action attribute to be generated in the target lane and the time attribute of the target lane; and when the intelligent body cannot drive to the target lane, controlling the intelligent body to drive to the virtual lane connected with the current lane, and performing global road-level planning by taking the next road connected with the virtual lane as a starting point.
For example, referring to fig. 1, when the current lane of the agent is a left-turn lane and the lane cannot be changed due to the solid line at the position of the current lane, the agent turns left according to the path of the current lane and enters a virtual lane, and the virtual lane is a transition lane connecting the current left-turn lane and a left-turn lane. Then, the next road connected by the virtual lane is taken as a starting point to carry out global road-level planning, the planning method is repeated,
further, planning a driving path of the intelligent agent according to the action attribute about to occur in the target lane and the time attribute of the target lane; the method comprises the following steps:
when the action attribute is right turn, planning a driving path of the intelligent agent according to the straight-going waiting time of the next intersection;
when the action attribute is not right turn, planning a driving path of the intelligent agent according to the time attribute;
and the time attribute is the waiting time of the traffic lights at the intersection corresponding to the target lane.
In this embodiment, the motion attributes are generally divided into right turn, left turn, straight going, and u-turn; for right turn, it is often unnecessary to wait for a direct turn, so in the process of determination, the present embodiment preferentially determines whether the action attribute is right turn. When the action attribute is right turn, predicting the waiting time of the traffic light of the next intersection, and planning the driving path of the intelligent agent according to the waiting time; when the action attribute is not turning right, the driving path of the intelligent agent needs to be planned according to the waiting time of the traffic light of the next intersection.
Further, planning a running path of the intelligent agent according to the straight-going waiting time of the next intersection; the method comprises the following steps:
if the straight-going waiting time is 0, the next road which is reached by straight-going is taken as a starting point to carry out global road-level planning, and a new global path cost C is obtained through calculation new1 ；
When C is new1 +C 1 ＜C old +C extra Then the target lane is changed into a lane with straight-going attribute, and the new global path cost C is used new1 Updating the navigation path;
when C is present new1 +C 1 ≤C old +C extra Then, taking the current global road level track as a subsequent navigation path;
wherein, C new1 Carrying out global road-level planning for the next road which is reached in a straight line as a starting point to obtain a new global path cost; c 1 The cost is the straight line cost; c old Global path cost for a current global road-level trajectory; c extra Is a constant;
and if the straight-ahead waiting time is more than 0, taking the current global road level track as a subsequent navigation path.
In this embodiment, when C new1 +C 1 ＜C old +C extra In time, it is shown that the global path cost of the new global road-level track is lower, so the navigation path is updated to the global path cost C new1 Corresponding global road level planning; otherwise, the original navigation path is used.
Further, planning a driving path of the intelligent agent according to the time attribute; the method comprises the following steps:
if the time attribute is determined to exceed the time threshold,
when the action attribute is left turn, judging whether a local lane-level track of which the target lane is a straight lane can be planned or not to obtain a first judgment result, and planning a driving path of the intelligent agent according to the first judgment result;
when the action attribute is straight, judging whether a local lane-level track with a right turn/left turn target lane can be planned or not to obtain a second judgment result, and planning a running path of the intelligent agent according to the second judgment result;
and when the action attribute is turning around, taking the current global road-level track as a subsequent navigation path.
In this embodiment, when the motion attribute is not right-handed, a more detailed operation is performed according to a different motion attribute; for example, when the action attribute of the target lane is left turn, whether the vehicle can run straight is judged; when the action attribute of the target lane is straight, judging whether a left-turn lane or a right-turn lane can be planned; if the action attribute is a U-turn, the path is possibly deviated, and the current global road-level track is taken as a subsequent navigation path.
Further, if the first judgment result is yes, performing global road-level planning by taking the next road which arrives straight as a starting point, and calculating to obtain a new global path cost C new1 ；
The step of planning the driving path of the intelligent agent according to the first judgment result comprises the following steps:
when C is new1 +C 1 +C t1 ＜C old +C t3 Then the target lane is changed into a lane with straight-going attribute, and the new global path cost C is used new1 Updating the navigation path;
when C is new1 +C 1 +C t1 ≥C old +C t3 Taking the current global road level track as a subsequent navigation path;
wherein, C new1 Carrying out global road-level planning for the next road which is reached in a straight line as a starting point to obtain a new global path cost; c 1 The cost is the straight line cost; c t1 Waiting for a time penalty for straight-ahead; c old Global path cost for a current global road-level trajectory; c t3 Latency penalty for left turn;
if the first judgment result is negative, the current global road level track is taken as a subsequent navigation path.
In this embodiment, when C new1 +C 1 +C t1 ＜C old +C t3 In time, the global path cost of the new global road-level track is lower, so that the navigation path is updated to the global path cost C new1 Corresponding global road level planning; otherwise, the original navigation path is used.
Further, if the second judgment result is yes, the next road which is reached by left turning/right turning is taken as a starting point to carry out global road-level planning, and a new global path generation is obtained through calculationValue C new2 /C new3 ；
Planning the driving path of the agent according to the second judgment result comprises the following steps:
when C is present new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 ＜C new3 +C 3 +C t3 When the current is over; the target lane is changed to a lane containing a right turn attribute and based on the new global path cost C new2 Updating the navigation path;
when C is present new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 Greater than or equal to C new3 +C 3 +C t3 When the current is in the normal state; the target lane is changed to a lane containing a left turn attribute and based on the new global path cost C new3 Updating the navigation path;
when not satisfying C new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 Then, taking the current global road level track as a subsequent navigation path;
wherein, C new2 Performing global road-level planning for the next road reached by the right turn as a starting point to obtain a new global path cost; c new3 Performing global road-level planning for the next road reached by the left turn as a starting point to obtain a new global path cost; c 2 A right turn cost; c 3 A left turn penalty; c t1 Wait time cost for straight going; c t2 Latency penalty for right turn; c t3 Latency penalty for left turn; c old Global path cost for a current global road-level trajectory;
if the second judgment result is negative, the current global road level track is taken as a subsequent navigation path.
In this embodiment, the travel of the agent is planned according to the second determination resultAnd the path, in brief, selects a route with the lowest global path cost from straight going, left turning and right turning as the navigation path of the agent. Specifically, when C new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 ＜C new3 +C 3 +C t3 In this case, it is indicated that the global path cost corresponding to the right turn is lower, and thus the navigation path is updated to the global path cost C new2 Corresponding global road level planning. When C is present new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 Greater than or equal to C new3 +C 3 +C t3 In time, it is shown that the global path cost corresponding to the left turn is lower, so the navigation path is updated to the global path cost C new3 Corresponding global road level planning. And when the situations are not met, taking the current global road-level track as a subsequent navigation path.
Further, planning a driving path of the intelligent agent according to the time attribute; the method comprises the following steps: and if the time attribute is judged not to exceed the time threshold, taking the current global road level track as a subsequent navigation path.
In this embodiment, the time attribute does not exceed the time threshold, meaning that the time required for the target lane to wait for the traffic light is short, within an acceptable range, and thus continues to be updated with the current navigation.
Example 2
Referring to fig. 3, the present embodiment provides an automatic driving lane planning apparatus based on a high-precision map, including:
a global road level planning module; the global route cost calculation method is used for calculating global route cost according to coordinate information of a starting point, a passing point and a finishing point and a topological relation among roads provided by the high-precision map, and taking a track with the lowest global route cost as a global road-level track;
a local lane level planning module; for following a global road-level track path the intersection segments the global road-level track; calculating local costs from the temporary starting point to the N temporary end points by taking the current lane as a temporary starting point and the N lanes at the next intersection as temporary end points; taking the track with the lowest local cost as a local lane level track;
an automatic driving decision module; for controlling the formal path of the agent according to the local lane-level trajectory.
Example 3
The present embodiment provides an electronic device, which includes a processor, a memory, and a program or an instruction stored in the memory and executable on the processor, wherein the program or the instruction implements the steps of the method of the above embodiment when executed by the processor.
Example 4
The present embodiment provides a readable storage medium on which a program or instructions are stored, which when executed by a processor implement the steps of the method of the above embodiment.
The processor is the processor in the electronic device in the above embodiment. Readable storage media, including computer-readable storage media, such as a computer Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic or optical disk, and so forth.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An automatic driving lane planning method based on a high-precision map, characterized by comprising the following steps:
s10, global road level planning: calculating global path cost according to coordinate information of a starting point, a passing point and a terminal point and a topological relation among roads provided by a high-precision map, and taking a track with the lowest global path cost as a global road-level track;
s20, local lane level planning: segmenting the global road-level track according to the intersections of the global road-level track path; taking a current lane as a temporary starting point and N lanes of a next intersection as temporary end points, and calculating local costs from the temporary starting point to the N temporary end points; taking the track with the lowest local cost as a local lane-level track;
s30, judging whether the intelligent agent can drive from the current lane to a target lane according to the local lane level track,
when the intelligent agent can drive to the target lane according to the judgment result, planning a driving path of the intelligent agent according to the action attribute to be generated in the target lane and the time attribute of the target lane;
and when the intelligent agent cannot drive to the target lane, controlling the intelligent agent to drive to a virtual lane connected with the current lane, and performing the global road-level planning by taking the next road connected with the virtual lane as a starting point.
2. The method of claim 1, wherein the route of travel of the agent is planned according to the action attribute about to occur in the target lane and the time attribute of the target lane; the method comprises the following steps:
when the action attribute is right turn, planning a driving path of the intelligent agent according to the straight-going waiting time of the next intersection;
when the action attribute is not right turn, planning a driving path of the intelligent agent according to the time attribute;
and the time attribute is the waiting time of the traffic light of the next intersection corresponding to the target lane.
3. The method of claim 2, wherein the routing of the agent is based on a straight-ahead wait time for a next intersection; the method comprises the following steps:
if the straight-going waiting time is 0, performing global road-level planning by taking the next road which is reached by straight-going as a starting point, and calculating to obtain a new global path cost C new1 ；
When C is present new1 +C 1 ＜C old +C extra Then, the target lane is changed into a lane containing straight-going attribute, and the new global path cost C is used for solving the problem that the target lane is changed into the lane containing straight-going attribute new1 Updating the navigation path;
when C is present new1 +C 1 ≥C old +C extra Taking the current global road level track as a subsequent navigation path;
wherein, C new1 Carrying out global road-level planning for the next road which is reached in a straight line as a starting point to obtain a new global path cost; c 1 The cost is the straight line cost; c old Global path cost for a current global road-level trajectory; c extra Is a constant;
and if the straight-going waiting time is more than 0, taking the current global road level track as a subsequent navigation path.
4. The method of claim 2, wherein the planning of the travel path of the agent is based on the time attribute; the method comprises the following steps:
if the time attribute is determined to exceed a time threshold,
when the action attribute is left turn, judging whether a straight local lane-level track can be planned or not to obtain a first judgment result, and planning a running path of the intelligent agent according to the first judgment result;
when the action attribute is straight, judging whether a local lane-level track of right turn/left turn can be planned or not to obtain a second judgment result, and planning a running path of the intelligent agent according to the second judgment result;
and when the action attribute is turning around, taking the current global road-level track as a subsequent navigation path.
5. The method of claim 4,
if the first judgment result is yes, performing global road-level planning by taking the next road which arrives straight as a starting point, and calculating to obtain a new global path cost C new1 ；
The planning of the driving path of the agent according to the first judgment result comprises:
when C is new1 +C 1 +C t1 ＜C old +C t3 Then, the target lane is changed into a lane containing straight-going attribute, and the new global path cost C is used for solving the problem that the target lane is changed into the lane containing straight-going attribute new1 Updating the navigation path;
when C is present new1 +C 1 +C t1 ≥C old +C t3 Taking the current global road level track as a subsequent navigation path;
wherein, C new1 Carrying out global road-level planning for the next road which is reached in a straight line as a starting point to obtain a new global path cost; c 1 The cost is the straight line cost; c t1 Wait time cost for straight going; c old Global path cost for a current global road-level trajectory; c t3 Latency penalty for left turn;
and if the first judgment result is negative, taking the current global road level track as a subsequent navigation path.
6. The method of claim 4,
if the second judgment result is yes, performing global road-level planning by taking the next road arrived by left turn/right turn as a starting point, and calculating to obtain a new global path cost C new2 /C new3 ；
The planning of the driving path of the agent according to the second judgment result comprises:
when C is present new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And are each and everyC new2 +C 2 +C t2 ＜C new3 +C 3 +C t3 When the current is over; the target lane is changed to a lane containing a right turn attribute and based on the new global path cost C new2 Updating the navigation path;
when C is present new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 And C is new2 +C 2 +C t2 Greater than or equal to C new3 +C 3 +C t3 When the current is in the normal state; the target lane is changed to a lane containing a left turn attribute and based on the new global path cost C new3 Updating the navigation path;
when not satisfying C new2 +C 2 +C t2 ＜C old +C t1 Or C new3 +C 3 +C t3 ＜C old +C t1 Taking the current global road level track as a subsequent navigation path;
wherein, C new2 Performing global road-level planning for the next road reached by the right turn as a starting point to obtain a new global path cost; c new3 Performing global road-level planning for the next road reached by the left turn as a starting point to obtain a new global path cost; c 2 A right turn cost; c 3 A left turn penalty; c t1 Wait time cost for straight going; c t2 Latency penalty for right turn; c t3 Latency penalty for left turn; c old Global path cost for a current global road-level trajectory;
and if the second judgment result is negative, taking the current global road level track as a subsequent navigation path.
7. The method of claim 2, wherein the planning of the travel path of the agent is based on the time attribute; the method comprises the following steps:
and if the time attribute is judged not to exceed the time threshold, taking the current global road level track as a subsequent navigation path.
8. An autonomous driving lane planning apparatus based on a high-precision map, comprising:
a global road level planning module; the global path cost is calculated according to the coordinate information of the starting point, the passing point and the end point and the topological relation among the roads provided by the high-precision map, and the track with the lowest global path cost is used as a global road-level track;
a local lane level planning module; the system comprises a global road level track path, a road surface path and a road surface path, wherein the global road level track path is used for carrying out segmentation on a global road level track according to intersections of the global road level track path; taking a current lane as a temporary starting point and N lanes of a next intersection as temporary end points, and calculating lane change costs from the temporary starting point to the N temporary end points; taking the track with the lowest lane change cost as a local lane level track;
an automatic driving decision module; and the intelligent agent driving path is used for controlling the driving path of the intelligent agent according to the local lane-level track.
9. An electronic device comprising a processor, a memory, and a program or instructions stored on the memory and executable on the processor, the program or instructions when executed by the processor implementing the autopilot lane planning method of any of claims 1-7.
10. A readable storage medium, characterized in that it stores thereon a program or instructions which, when executed by a processor, implements the autopilot lane planning method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310273726.1A CN115973197B (en) | 2023-03-21 | 2023-03-21 | Lane planning method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310273726.1A CN115973197B (en) | 2023-03-21 | 2023-03-21 | Lane planning method and device, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115973197A true CN115973197A (en) | 2023-04-18 |
| CN115973197B CN115973197B (en) | 2023-08-11 |
Family
ID=85965262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310273726.1A Active CN115973197B (en) | 2023-03-21 | 2023-03-21 | Lane planning method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115973197B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008176593A (en) * | 2007-01-19 | 2008-07-31 | Sumitomo Electric Ind Ltd | Information providing system, on-road device, on-vehicle device, and vehicle |
| JP2014035273A (en) * | 2012-08-09 | 2014-02-24 | Aisin Aw Co Ltd | Cost calculation device, cost calculation program, and navigation device |
| US20160129907A1 (en) * | 2014-11-12 | 2016-05-12 | Hyundai Motor Company | Driving path planning apparatus and method for autonomous vehicle |
| CN109557928A (en) * | 2019-01-17 | 2019-04-02 | 湖北亿咖通科技有限公司 | Automatic driving vehicle paths planning method based on map vector and grating map |
| CN109579858A (en) * | 2017-09-28 | 2019-04-05 | 腾讯科技（深圳）有限公司 | Navigation data processing method, device, equipment and storage medium |
| CN111397631A (en) * | 2020-04-10 | 2020-07-10 | 上海安吉星信息服务有限公司 | Navigation path planning method and device and navigation equipment |
| CN111664864A (en) * | 2020-05-31 | 2020-09-15 | 武汉中海庭数据技术有限公司 | Dynamic planning method and device based on automatic driving |
| CN112161636A (en) * | 2020-08-28 | 2021-01-01 | 深圳市跨越新科技有限公司 | Truck route planning method and system based on one-way simulation |
| CN113155145A (en) * | 2021-04-19 | 2021-07-23 | 吉林大学 | Lane-level path planning method for automatic driving lane-level navigation |
| CN113932823A (en) * | 2021-09-23 | 2022-01-14 | 同济大学 | Unmanned multi-target-point track parallel planning method based on semantic road map |
| WO2022193584A1 (en) * | 2021-03-15 | 2022-09-22 | 西安交通大学 | Multi-scenario-oriented autonomous driving planning method and system |
| CN115675520A (en) * | 2022-09-23 | 2023-02-03 | 深圳元戎启行科技有限公司 | Unmanned driving implementation method and device, computer equipment and storage medium |
| CN115752499A (en) * | 2022-11-25 | 2023-03-07 | 长城汽车股份有限公司 | Path planning method and device, terminal equipment and vehicle |
-
2023
- 2023-03-21 CN CN202310273726.1A patent/CN115973197B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008176593A (en) * | 2007-01-19 | 2008-07-31 | Sumitomo Electric Ind Ltd | Information providing system, on-road device, on-vehicle device, and vehicle |
| JP2014035273A (en) * | 2012-08-09 | 2014-02-24 | Aisin Aw Co Ltd | Cost calculation device, cost calculation program, and navigation device |
| US20160129907A1 (en) * | 2014-11-12 | 2016-05-12 | Hyundai Motor Company | Driving path planning apparatus and method for autonomous vehicle |
| CN109579858A (en) * | 2017-09-28 | 2019-04-05 | 腾讯科技（深圳）有限公司 | Navigation data processing method, device, equipment and storage medium |
| CN109557928A (en) * | 2019-01-17 | 2019-04-02 | 湖北亿咖通科技有限公司 | Automatic driving vehicle paths planning method based on map vector and grating map |
| CN111397631A (en) * | 2020-04-10 | 2020-07-10 | 上海安吉星信息服务有限公司 | Navigation path planning method and device and navigation equipment |
| CN111664864A (en) * | 2020-05-31 | 2020-09-15 | 武汉中海庭数据技术有限公司 | Dynamic planning method and device based on automatic driving |
| CN112161636A (en) * | 2020-08-28 | 2021-01-01 | 深圳市跨越新科技有限公司 | Truck route planning method and system based on one-way simulation |
| WO2022193584A1 (en) * | 2021-03-15 | 2022-09-22 | 西安交通大学 | Multi-scenario-oriented autonomous driving planning method and system |
| CN113155145A (en) * | 2021-04-19 | 2021-07-23 | 吉林大学 | Lane-level path planning method for automatic driving lane-level navigation |
| CN113932823A (en) * | 2021-09-23 | 2022-01-14 | 同济大学 | Unmanned multi-target-point track parallel planning method based on semantic road map |
| CN115675520A (en) * | 2022-09-23 | 2023-02-03 | 深圳元戎启行科技有限公司 | Unmanned driving implementation method and device, computer equipment and storage medium |
| CN115752499A (en) * | 2022-11-25 | 2023-03-07 | 长城汽车股份有限公司 | Path planning method and device, terminal equipment and vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115973197B (en) | 2023-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108151751B (en) | Path planning method and device based on combination of high-precision map and traditional map | |
| EP4086578A1 (en) | Vehicle behavior prediction method and apparatus, electronic device, and storage medium | |
| CN110426050B (en) | Map matching correction method, device, equipment and storage medium | |
| JP2022104639A (en) | Vehicle lane change control method, device, storage medium, and program | |
| CN109916414B (en) | Map matching method, apparatus, device and medium | |
| CN114212110B (en) | Obstacle trajectory prediction method and device, electronic equipment and storage medium | |
| CN113682318A (en) | Vehicle running control method and device | |
| CN113532448A (en) | Navigation method and system for automatically driving vehicle and driving control equipment | |
| CN113985871A (en) | Parking path planning method, parking path planning device, vehicle and storage medium | |
| CN113997954B (en) | Method, device and equipment for predicting vehicle driving intention and readable storage medium | |
| CN114061606B (en) | Path planning method, path planning device, electronic equipment and storage medium | |
| CN115062202A (en) | Method, device, equipment and storage medium for predicting driving behavior intention and track | |
| CN114519931B (en) | Method and device for predicting behavior of target vehicle in intersection environment | |
| CN114620071A (en) | Detour trajectory planning method, device, equipment and storage medium | |
| CN114485706A (en) | Route planning method and device, storage medium and electronic equipment | |
| RU2574040C2 (en) | Method and device for determining probability of exiting area identified on digital map as open terrain | |
| CN115973197B (en) | Lane planning method and device, electronic equipment and readable storage medium | |
| CN114858176B (en) | Path navigation method and device based on automatic driving | |
| CN113788028B (en) | Vehicle control method, device and computer program product | |
| CN115402323A (en) | Lane changing decision method and electronic equipment | |
| CN113050452A (en) | Vehicle lane change control method and device, computer equipment and storage medium | |
| Lakshna et al. | Smart Navigation for Vehicles to Avoid Road Traffic Congestion using Weighted Adaptive Navigation* Search Algorithm | |
| CN116331190B (en) | Correction method, device and equipment for memory route of memory parking and vehicle | |
| CN118209127A (en) | Path optimization method, path optimization device, electronic equipment and storage medium | |
| CN113052350A (en) | Path planning method and device, electronic equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |