CN110908379A - Vehicle track prediction method and device based on historical information and storage medium - Google Patents
Vehicle track prediction method and device based on historical information and storage medium Download PDFInfo
- Publication number
- CN110908379A CN110908379A CN201911202038.6A CN201911202038A CN110908379A CN 110908379 A CN110908379 A CN 110908379A CN 201911202038 A CN201911202038 A CN 201911202038A CN 110908379 A CN110908379 A CN 110908379A
- Authority
- CN
- China
- Prior art keywords
- information
- vehicles
- historical
- operation information
- current
- 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
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
Abstract
The invention relates to a vehicle track prediction method, a device and a storage medium based on historical information.A perception module acquires the running information of other vehicles on a road, and a prediction module acquires the current orientation and the acceleration of the other vehicles after correction according to the current running information and the historical running information of the other vehicles; and predicting the future tracks of other vehicles according to the corrected current orientation and acceleration of other vehicles. The method can tolerate the sensed environmental information error, effectively improve the prediction accuracy of other vehicle tracks and further ensure the traffic safety.
Description
Technical Field
The invention relates to a vehicle track prediction method and device based on historical information and a storage medium.
Background
With the development of science and technology, especially the rapid development of intelligent computing, the research of the automatic driving automobile technology becomes a focus of all industries. The '12 leading edge technologies for determining future economy' report issued by McKensin discusses the influence degree of the 12 leading edge technologies on the future economy and society, and analyzes and estimates the respective economic and social influence of the 12 technologies in 2025, wherein the automatic driving automobile technology is ranked at the 6 th position, and the influence of the automatic driving automobile technology in 2025 is estimated as follows: economic benefits are about $ 0.2-1.9 trillion per year, and social benefits can recover 3-15 million lives per year.
According to the technical field division, the automatic driving can be divided into a sensing module, a positioning module, a prediction module, a control module and an execution module. The sensing module is equivalent to eyes of people, the peripheral environment state is collected in real time through sensors such as a camera, a millimeter wave radar and a laser radar, the positioning module is used for obtaining position information of the vehicle, the prediction module is used for predicting the self running track of the vehicle or predicting the track of other vehicles, pedestrians and motor vehicles, the control is decision and planning of vehicle motion, and the execution is a decision planning command of executing the vehicle.
The prediction module is an integral part of the automated driving development. Autonomous vehicles are an individual in a traffic scene, and interaction with other individuals in the traffic scene is essential. In such a case, if the autonomous vehicle is unable to know the likely future behavior of other individuals in the traffic scene, it may be difficult for the autonomous vehicle to calculate the appropriate next action to perform. For example, an autonomous vehicle may choose to turn left without knowing the tendency of the vehicle to brake in front of the left lane, such that the autonomous vehicle is likely to rear-end the vehicle in front of the left lane, causing traffic accidents and hazards.
The input to the prediction module of the autonomous vehicle is information of the surroundings perceived by the perception module. Under the existing technical conditions, the ambient environment information sensed by the sensing module may not be very accurate, and some false detection or deviation may occur. In this case, if the prediction module predicts the travel tracks of other vehicles according to the deviated environmental information, it may be impossible for the autonomous vehicle to appropriately evade the threat in the traffic scene, thereby bringing about a traffic safety hazard.
Disclosure of Invention
The invention aims to provide a vehicle track prediction method and system based on historical information, which can effectively tolerate perception errors and improve the accuracy of other vehicle track predictions.
Based on the same inventive concept, the invention has three independent technical schemes:
1. a vehicle track prediction method based on historical information is characterized in that: the sensing module acquires the running information of other vehicles on the road, and the prediction module acquires the corrected current orientation and acceleration of the other vehicles according to the current running information and the historical running information of the other vehicles; and predicting the future tracks of other vehicles according to the corrected current orientation and acceleration of other vehicles.
Further, the operation information of the other vehicles includes position and speed.
Further, the prediction module obtains the corrected current orientation and acceleration of the other vehicle according to the current operation information and the historical operation information of the other vehicle, and the method comprises the following steps:
and fitting the historical operation information and the current operation information by using a univariate polynomial with the time as an independent variable and the position or the speed as a dependent variable to obtain a polynomial of the relation between the time and the position or the relation between the time and the speed.
Further, the prediction module calculates the corrected current position and speed of the other vehicle according to the polynomial obtained by fitting, and obtains the corrected current orientation and acceleration of the other vehicle according to the corrected position and speed information.
Further, when the historical operation information and the current operation information are matched by utilizing a univariate polynomial, a first-order polynomial or a second-order polynomial is adopted, and the polynomial with the minimum root-mean-square error is selected as a final polynomial result.
Further, according to the corrected current position and speed, the current orientation and acceleration of other vehicles are obtained, and the method is realized by the following steps:
the connecting direction of the previous historical moment position and the current position is the current orientation;
the acceleration a is (v 0-v 1)/t, v0 is the current speed, and v1 is the last historical time speed.
Further, the historical operation information is a set from the operation information at the latest moment to the operation information at the nth moment, and n is more than or equal to 3.
Further, the time interval selected by the historical operation information is 5-15 s.
2. The vehicle track prediction device based on historical information is characterized by comprising a sensing module, a judging module and a control module, wherein the sensing module is used for acquiring the running information of other vehicles on a road; comprises a prediction module for implementing the method.
3. A computer-readable storage medium having a computer program stored thereon, characterized in that: which when executed by a processor implements the method described above.
The invention has the following beneficial effects:
the prediction module obtains the corrected current orientation and acceleration of other vehicles according to the current operation information and the historical operation information of other vehicles; according to the corrected current orientation and acceleration of other vehicles, the future tracks of the other vehicles are predicted, so that the perceived environmental information error can be tolerated, the prediction accuracy of the tracks of the other vehicles is effectively improved, and the traffic safety is further ensured.
The method takes time as an independent variable, positions or speeds as a dependent variable, utilizes a univariate polynomial to fit historical operation information and current operation information to obtain a polynomial of the relation between the time and the positions or the relation between the time and the speeds, and utilizes the polynomial to realize the correction of the operation information of other vehicles to obtain more accurate current orientation and acceleration of other vehicles, thereby having simple and convenient operation and further ensuring the accuracy of the prediction of the tracks of other vehicles.
The invention uses unary polynomial to match the historical operation information and the current operation information, and uses the first-order polynomial or the second-order polynomial to select the polynomial with the minimum root-mean-square error as the final polynomial result. Because the behavior of the vehicle in a short time cannot accord with a complex polynomial, the method adopts the first-order polynomial or the second-order polynomial for fitting, and effectively improves the operation speed.
The historical operation information is a set from the operation information at the latest moment to the operation information at the nth moment, wherein n is more than or equal to 3; the time interval of the historical operation information is 5-15 s. According to the invention, the historical operation information effectively ensures the accuracy of fitting between the historical operation information and the current operation information through the selection of the parameters.
Drawings
FIG. 1 is a general flow diagram of a vehicle trajectory prediction method based on historical information in accordance with the present invention;
fig. 2 is a flow chart of the information optimization of the present invention.
Detailed Description
The present invention is described in detail with reference to the embodiments shown in the drawings, but it should be understood that these embodiments are not intended to limit the present invention, and those skilled in the art should understand that functional, methodological, or structural equivalents or substitutions made by these embodiments are within the scope of the present invention.
The first embodiment is as follows:
vehicle track prediction method based on historical information
The sensing module acquires the running information of other vehicles on the road, and as shown in fig. 1, the prediction module performs information optimization according to the current running information and historical running information of other vehicles to acquire the corrected current orientation and acceleration (transverse intention) of other vehicles; and predicting the future tracks of other vehicles according to the corrected current orientation and acceleration of other vehicles. The operation information of the other vehicles includes position and speed.
(I) historical operational information selection
The problem of sensing deviation or false detection of the sensing module is solved, and the sensing deviation or false detection is achieved through fitting with historical running information of the vehicle. Because the vehicle is a physical individual with a mass and volume, its attitude is unlikely to have a large jump in a short period of time. For example, the locomotive faces 30 degrees to the right before 0.3 seconds, and the locomotive faces 60 degrees to the left at the current moment is basically impossible. Because of this, the history information of use cannot be too long nor too short. If too short, the historical information is insufficient to assist in correcting the current error; if too long, much of the older historical information is not correlated well with the current information, and the resulting correction may be inaccurate.
In implementation, the historical operation information is a set of operation information from the most recent time to the current nth time, where n is 3 and the time interval is 10s in this embodiment.
(II) Current information optimization
For the vehicle orientation, in the case where the time of the last history information is very close to the current time, the orientation of the vehicle is substantially equal to the orientation from the last position to the current position. That is, knowing the more accurate location very close to the current last time and the current location, more accurate heading information for the vehicle can be obtained.
As for the acceleration of the vehicle, also in the case where the time of the last history information is very close to the present time, the acceleration of the vehicle can be basically represented by the change in the speed from the previous time to the present time. That is, knowing the speed very close to the current previous time and the current speed more accurately, more accurate acceleration information of the vehicle can be obtained. From the above, it is known that the current vehicle attitude and several historical vehicle attitudes closer to the current time conform to a certain trend, and the trend is obtained by fitting the historical operation information and the current operation information.
As shown in fig. 2, the prediction module performs current information optimization according to current operation information and historical operation information of other vehicles to obtain corrected current orientations and accelerations of other vehicles, and is implemented by the following method:
and fitting the historical operation information and the current operation information by using a univariate polynomial with the time as an independent variable and the position or the speed as a dependent variable to obtain a polynomial of the relation between the time and the position or the relation between the time and the speed.
And the prediction module calculates the corrected current position and speed of the other vehicles according to the polynomial obtained by fitting, and obtains the corrected current orientation and acceleration of the other vehicles according to the corrected position and speed information.
When the univariate polynomial is used for matching the historical operation information and the current operation information, the first-order polynomial or the second-order polynomial is adopted, and the polynomial with too high degree is not suitable to be selected, because the behavior of the vehicle in a short time cannot be matched with a complex polynomial. And selecting the polynomial with the minimum root mean square error as a final polynomial result.
Obtaining the current orientation and acceleration of other vehicles according to the corrected current position and speed and the position and speed of the previous historical moment, and realizing the method by the following steps:
the connecting direction of the previous historical moment position and the current position is the current orientation;
the acceleration a is (v 0-v 1)/t, v0 is the current speed, and v1 is the last historical time speed.
(III) future trajectory Generation
First, the end point of the vehicle on its target road is found. Simply, the location where the vehicle should be after reaching the target road. For example, the target road of the vehicle a is the left lane of the current lane of the vehicle. On this left lane, there is already a vehicle B in front, and it is predicted that the future trajectory of the vehicle B will be a constant speed travel on this left lane. If the vehicle B is relatively forward, the vehicle A is difficult to overtake the vehicle B in the process of changing to the left lane. The end point of the vehicle a on the target road (left lane) must be behind the position of the vehicle B at the time of the end point.
Then, some more general assumptions are made, such as that the vehicle a follows a constant longitudinal speed assumption when traveling on a future trajectory, and the vehicle a will eventually reach the center of the left lane. Then we know the lateral and longitudinal position of the end point of vehicle a, i.e. the end point of vehicle a on the target road.
By sampling the possible time for the vehicle to reach the destination road end, we can obtain a plurality of future trajectories for the vehicle to reach the destination road. From these future trajectories, the most suitable trajectory is found by some rules. In this embodiment, a driving consumption rule is adopted. In general, driving from point a to point B directly along line a-B is the shortest driving distance, and the driving distance from a to B along a fold line is extended. However, if the current head orientation is the direction of a certain broken line, not the direction of the line segment A-B, and the vehicle is required to be steered firstly to drive according to the line segment A-B, the vehicle is required to drive forwards in the steering process of the head, and the vehicle cannot drive along the line segment A-B strictly. And the steering of the locomotive is also consumed. Therefore, different weights need to be configured for different vehicle consumptions, so as to calculate the final driving consumption of the vehicle.
Example two:
vehicle track prediction method based on historical information
The historical operation information is a set from the operation information at the latest moment to the operation information at the nth moment, and in the second embodiment, n is 4 and the time interval is 8 s.
The rest is the same as the first embodiment.
Example three:
vehicle track prediction device based on historical information
The system comprises a sensing module, a control module and a display module, wherein the sensing module is used for acquiring and acquiring the running information of other vehicles on the road; the surrounding environment state is collected in real time through sensors such as a camera, a millimeter wave radar and a laser radar.
The prediction module is configured to implement the method of the first embodiment or the second embodiment, and the prediction module includes a control unit, a storage unit, and a buffer, where the buffer is configured to store historical operating information.
Example four:
computer readable storage medium
The computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the method of embodiment one or embodiment two.
The above-listed detailed description is only a specific description of a possible embodiment of the present invention, and they are not intended to limit the scope of the present invention, and equivalent embodiments or modifications made without departing from the technical spirit of the present invention should be included in the scope of the present invention.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims (10)
1. A vehicle track prediction method based on historical information is characterized in that: the sensing module acquires the running information of other vehicles on the road, and the prediction module acquires the corrected current orientation and acceleration of the other vehicles according to the current running information and the historical running information of the other vehicles; and predicting the future tracks of other vehicles according to the corrected current orientation and acceleration of other vehicles.
2. The history information-based vehicle track prediction method according to claim 1, characterized in that: the operation information of the other vehicles includes position and speed.
3. The history information-based vehicle track prediction method according to claim 2, characterized in that: the prediction module obtains the corrected current orientation and acceleration of other vehicles according to the current operation information and the historical operation information of other vehicles, and the method is realized by the following steps:
and fitting the historical operation information and the current operation information by using a univariate polynomial with the time as an independent variable and the position or the speed as a dependent variable to obtain a polynomial of the relation between the time and the position or the relation between the time and the speed.
4. The history information-based vehicle track prediction method according to claim 3, characterized in that: and the prediction module calculates the corrected current position and speed of the other vehicles according to the polynomial obtained by fitting, and obtains the corrected current orientation and acceleration of the other vehicles according to the corrected position and speed information.
5. The history information-based vehicle track prediction method according to claim 4, characterized in that: and when the historical operation information and the current operation information are matched by utilizing a univariate polynomial, a first-order polynomial or a second-order polynomial is adopted, and the polynomial with the minimum root-mean-square error is selected as a final polynomial result.
6. The history information-based vehicle track prediction method according to claim 5, characterized in that: obtaining the current orientation and acceleration of other vehicles according to the corrected current position and speed, and realizing the following method:
the connecting direction of the previous historical moment position and the current position is the current orientation;
the acceleration a is (v 0-v 1)/t, v0 is the current speed, and v1 is the last historical time speed.
7. The history information-based vehicle track prediction method according to any one of claims 1 to 6, characterized in that: the historical operation information is a set from the operation information at the latest moment to the operation information at the nth moment, and n is more than or equal to 3.
8. The history information-based vehicle track prediction method according to claim 7, characterized in that: the time interval of the historical operation information is 0.5 s-1.5 s.
9. The vehicle track prediction device based on historical information is characterized by comprising a sensing module, a judging module and a control module, wherein the sensing module is used for acquiring the running information of other vehicles on a road; comprising a prediction module for implementing the method of any one of claims 1 to 8.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: the computer program, when executed by a processor, implements the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911202038.6A CN110908379A (en) | 2019-11-29 | 2019-11-29 | Vehicle track prediction method and device based on historical information and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911202038.6A CN110908379A (en) | 2019-11-29 | 2019-11-29 | Vehicle track prediction method and device based on historical information and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110908379A true CN110908379A (en) | 2020-03-24 |
Family
ID=69820739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911202038.6A Pending CN110908379A (en) | 2019-11-29 | 2019-11-29 | Vehicle track prediction method and device based on historical information and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110908379A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111595303A (en) * | 2020-07-03 | 2020-08-28 | 成都微宇科技有限责任公司 | Method for screening aerial photos |
| CN111707258A (en) * | 2020-06-15 | 2020-09-25 | 中国第一汽车股份有限公司 | External vehicle monitoring method, device, equipment and storage medium |
| CN112926930A (en) * | 2021-04-14 | 2021-06-08 | 广州市钛菱云科技有限公司 | Takeaway fresh-keeping method, system, equipment and computer readable storage medium |
| CN113419258A (en) * | 2021-07-07 | 2021-09-21 | 东软集团股份有限公司 | Positioning abnormity detection method and related equipment thereof |
| CN113635900A (en) * | 2021-08-10 | 2021-11-12 | 吉林大学 | Energy management-based lane change decision control method in prediction cruise process |
| CN113672845A (en) * | 2020-05-14 | 2021-11-19 | 阿波罗智联(北京)科技有限公司 | Vehicle track prediction method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101324440A (en) * | 2008-07-29 | 2008-12-17 | 光庭导航数据（武汉）有限公司 | Map-matching method based on forecast ideology |
| CN106525033A (en) * | 2016-09-22 | 2017-03-22 | 深圳市元征科技股份有限公司 | Running track determination method and device thereof |
| CN110146100A (en) * | 2018-02-13 | 2019-08-20 | 华为技术有限公司 | Trajectory predictions method, apparatus and storage medium |
-
2019
- 2019-11-29 CN CN201911202038.6A patent/CN110908379A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101324440A (en) * | 2008-07-29 | 2008-12-17 | 光庭导航数据（武汉）有限公司 | Map-matching method based on forecast ideology |
| CN106525033A (en) * | 2016-09-22 | 2017-03-22 | 深圳市元征科技股份有限公司 | Running track determination method and device thereof |
| CN110146100A (en) * | 2018-02-13 | 2019-08-20 | 华为技术有限公司 | Trajectory predictions method, apparatus and storage medium |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113672845A (en) * | 2020-05-14 | 2021-11-19 | 阿波罗智联(北京)科技有限公司 | Vehicle track prediction method, device, equipment and storage medium |
| CN111707258A (en) * | 2020-06-15 | 2020-09-25 | 中国第一汽车股份有限公司 | External vehicle monitoring method, device, equipment and storage medium |
| CN111707258B (en) * | 2020-06-15 | 2022-05-31 | 中国第一汽车股份有限公司 | External vehicle monitoring method, device, equipment and storage medium |
| CN111595303A (en) * | 2020-07-03 | 2020-08-28 | 成都微宇科技有限责任公司 | Method for screening aerial photos |
| CN112926930A (en) * | 2021-04-14 | 2021-06-08 | 广州市钛菱云科技有限公司 | Takeaway fresh-keeping method, system, equipment and computer readable storage medium |
| CN113419258A (en) * | 2021-07-07 | 2021-09-21 | 东软集团股份有限公司 | Positioning abnormity detection method and related equipment thereof |
| CN113419258B (en) * | 2021-07-07 | 2024-03-01 | 东软集团股份有限公司 | Positioning abnormality detection method and related equipment thereof |
| CN113635900A (en) * | 2021-08-10 | 2021-11-12 | 吉林大学 | Energy management-based lane change decision control method in prediction cruise process |
| CN113635900B (en) * | 2021-08-10 | 2024-03-19 | 吉林大学 | Channel switching decision control method based on energy management in predicted cruising process |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110908379A (en) | Vehicle track prediction method and device based on historical information and storage medium | |
| EP3822142A1 (en) | Confidence levels along the same predicted trajectory of an obstacle | |
| JP6637400B2 (en) | Vehicle control device | |
| US11267474B2 (en) | Vehicle control device, vehicle control method, and storage medium | |
| Cosgun et al. | Towards full automated drive in urban environments: A demonstration in gomentum station, california | |
| US11462099B2 (en) | Control system and control method for interaction-based long-term determination of trajectories for motor vehicles | |
| CN110389580B (en) | Method for drift correction for planning a path of an autonomous vehicle | |
| JP6432679B2 (en) | Stop position setting apparatus and method | |
| CN109421742A (en) | Method and apparatus for monitoring autonomous vehicle | |
| CN109050533B (en) | Automatic lane changing control method and device for vehicle | |
| WO2021103510A1 (en) | Method, device and storage medium for controlling autonomous vehicle | |
| KR20200014931A (en) | Vehicle information storage method, vehicle driving control method, and vehicle information storage device | |
| CN110678372B (en) | Vehicle control device | |
| US20200247415A1 (en) | Vehicle, and control apparatus and control method thereof | |
| JP6558214B2 (en) | Automatic driving device | |
| US10909377B2 (en) | Tracking objects with multiple cues | |
| CN116209611A (en) | Method and system for using other road user's responses to self-vehicle behavior in autopilot | |
| US11618473B2 (en) | Vehicle control system | |
| US20220121213A1 (en) | Hybrid planning method in autonomous vehicle and system thereof | |
| US20220392276A1 (en) | Vehicle behavior evaluation device, vehicle behavior evaluation method, and vehicle behavior evaluation program product | |
| US20240043034A1 (en) | Method and system for conditional operation of an autonomous agent | |
| JPWO2019065564A1 (en) | Automatic operation control device and method | |
| WO2020164090A1 (en) | Trajectory prediction for driving strategy | |
| Weisswange et al. | Intelligent traffic flow assist: Optimized highway driving using conditional behavior prediction | |
| JPWO2018211645A1 (en) | Driving support method and driving support device |
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 |