CN112102616A - Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model - Google Patents
Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model Download PDFInfo
- Publication number
- CN112102616A CN112102616A CN202010939349.7A CN202010939349A CN112102616A CN 112102616 A CN112102616 A CN 112102616A CN 202010939349 A CN202010939349 A CN 202010939349A CN 112102616 A CN112102616 A CN 112102616A
- Authority
- CN
- China
- Prior art keywords
- error
- reference point
- vehicle
- evaluation
- tracking
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
Abstract
The invention discloses a method, a system and a device for tracking and evaluating a comfortable unmanned vehicle path based on an error model, which are composed of evaluation indexes of two latitudes: eerrorReflecting the safety level of the trace, EcomfortReflecting riding comfort, and improving the objectivity and continuity of the evaluation method on a path with large curvature change based on a tracking error model, a future reference point and a past reference point; and (5) providing an evaluation formula. The evaluation method provided by the invention can evaluate the unmanned vehicle tracking safety and riding comfort in different weights by adjusting the weight matrix.
Description
Technical Field
The invention relates to the field of path tracking of unmanned vehicles, in particular to an evaluation method for the actual path tracking accuracy and riding comfort of an intelligent vehicle.
Background
Under the background that the problems of traffic jam and traffic safety become more serious, the unmanned technology with more objective decision-making capability and more accurate execution capability is increasingly paid attention by people. In the whole vehicle-road interaction process of the unmanned vehicle, three steps of environment perception, path planning and path tracking are carried out hierarchically, wherein the path tracking layer is a concrete implementation of a vehicle bottom-layer execution mechanism on an upper-layer expected execution target, and the driving performance and the intelligent degree rating of the unmanned vehicle are directly reflected by the path tracking effect. At present, the quantitative evaluation method aiming at the path tracking effect is less, the rapid development of unmanned driving in foreign countries is countered, and the perfect vehicle intelligent behavior evaluation method and standard are greatly benefited. The good path tracking evaluation method indicates the direction for the subsequent optimization of the control program, and can also feedback and influence the further improvement of an upper path planning layer. A comprehensive path tracking evaluation method not only can reflect and evaluate the motion safety of a vehicle, but also can fully consider the riding comfort of passengers so as to really realize unmanned driving benefiting for people.
Disclosure of Invention
1. Objects of the invention
The invention provides an evaluation method which can quantitatively evaluate the path tracking effect of an unmanned vehicle and also can give consideration to the riding comfort of passengers.
2. The technical scheme adopted by the invention
The invention discloses a comfortable unmanned vehicle path tracking evaluation method based on an error model, which comprises two evaluation indexes of latitude degree: eerrorReflecting the safety level of the trace, EcomfortThe riding comfort is reflected, and the riding comfort are weighted through the weight matrix;
the unmanned vehicle takes an expected path as a control target, and an included angle between a central axis of the unmanned vehicle body and a tangent line on a current reference point is taken as a direction error and is recorded as theta; the distance between the mass center of the vehicle body and the current reference point is a position error and is marked as e; establishing a tracking error model based on a direction error and a position error, wherein the direction error is the deviation between the vehicle course and the expected course, and the position error is the difference between the vehicle position and the expected position;
based on the direction error and the position error of the tracking error model, a future reference point and a past reference point are further considered so as to improve the objectivity and the continuity of the evaluation method on a path with large curvature change; the evaluation formula is proposed as follows:
wherein ek+1,ek,ek-1Forming a direction error matrix, namely a position error of the vehicle relative to a next reference point, a position error relative to a current reference point and a position error relative to a previous reference point; a. thek+1,Ak,Ak-1Form a position error coefficient matrix, wherein Ak+1,Ak,Ak-1Respectively representing the evaluation weights of the k-th time, k +1, k-1 and the like, so as to perform weighted evaluation on the position errors of the vehicle relative to different reference points; in the actual evaluation, the weight coefficient A corresponding to the current position can be usedkSet to 1, the remainder to 0.2 or less; likewise, θk+1,θk,θk-1Forming a directional error matrix, also representing the directional error of the vehicle with respect to future, current and past reference points, respectively, Bk+1,Bk,Bk-1Then the corresponding weighting matrix.
Preferably, reference points are continuously increased, the requirement of the evaluation method on the smoothness degree of the whole tracking process is improved, and the method is expanded to n latitude:
calculating the actual yaw rate E of the vehicleyawYaw rate E generated by neutral steering of the vehicle along a reference pathrefThe difference value of (a) is used as a new factor for evaluating the riding attribute; where k is the right to comfortThe weight factor, namely:
Ecomfort=k(Eyaw-Eref)
the neutral steering yaw rate is a reference yaw rate and can be calculated by the following formula;
Eref=v/r
wherein r is the curvature radius 7 of the current reference point, v is the speed of the mass center of the rear axle of the vehicle, a is the distance between the future reference point X and the current reference point Y, b is the distance between the current reference point Y and the past reference point, and k is the weight coefficient of comfort level; v is the vehicle speed, c is the distance between the future reference point X and the past reference point; x, Y, Z are expressed by (x1, y1), (x2, y2), and (x3, y 3).
The invention provides an error model-based comfortable unmanned vehicle path tracking evaluation system, which comprises an evaluation index module, an evaluation index module and a control module, wherein the evaluation index module comprises two latitude evaluation indexes: eerrorReflecting the safety level of the trace, EcomfortThe riding comfort is reflected, and the riding comfort are weighted through the weight matrix;
the unmanned vehicle takes an expected path as a control target, and an included angle between a central axis of the unmanned vehicle body and a tangent line on a current reference point is taken as a direction error and is recorded as theta; the distance between the mass center of the vehicle body and the current reference point is a position error and is marked as e; establishing a tracking error model based on a direction error and a position error, wherein the direction error is the deviation between the vehicle course and the expected course, and the position error is the difference between the vehicle position and the expected position;
the error evaluation module is used for further considering a future reference point and a past reference point based on the direction error and the position error of the tracking error model so as to improve the objectivity and the continuity of the evaluation method on a path with larger curvature change; the evaluation formula is proposed as follows:
wherein ek+1,ek,ek-1Forming a direction error matrix, namely a position error of the vehicle relative to a next reference point, a position error relative to a current reference point and a position error relative to a previous reference point; a. thek+1, Ak,Ak-1Form a position error coefficient matrix, where Ak+1,Ak,Ak-1Respectively representing the evaluation weights of the k-th time, k +1, k-1 and the like, so as to perform weighted evaluation on the position errors of the vehicle relative to different reference points; in actual evaluation, the weighting coefficient A corresponding to the current position can be usedkSet to 1, the remainder to 0.2 or less; likewise, θk+1,θk,θk-1Forming a direction error matrix which also represents the direction error of the vehicle relative to the future, the current and the past reference points respectively, Bk+1,Bk,Bk-1Then the corresponding weighting matrix.
Preferably, the multi-dimensional evaluation module continues to increase the reference points, improves the requirement of the evaluation method on the smoothness degree of the whole tracking process, and extends to n latitude:
further comprising a comfort evaluation module:
calculating the actual yaw rate E of the vehicleyawWith yaw rate E resulting from neutral steering of the vehicle along a reference pathrefThe difference value of (a) is used as a new factor for evaluating the riding attribute; where k is the weight coefficient for comfort, i.e.:
Ecomfort=k(Eyaw-Eref)
the neutral steering yaw rate is a reference yaw rate, and can be calculated by the following formula:
Eref=v/r
wherein r is the curvature radius 7 of the current reference point, v is the speed of the mass center of the rear axle of the vehicle, a is the distance between the future reference point X and the current reference point Y, b is the distance between the current reference point Y and the past reference point, and k is the weight coefficient of comfort; v is the vehicle speed, c is the distance between the future reference point X and the past reference point; x, Y, Z are expressed by (x1, y1), (x2, y2), and (x3, y 3).
The invention discloses a comfortable unmanned vehicle path tracking evaluation system based on an error model, which comprises:
a memory;
one or more processors, and
one or more programs stored in the memory and configured to be executed by the one or more processors, the programs when executed by the processors implementing any of the unmanned vehicle path tracking quantitative rating methods.
3. Advantageous effects adopted by the present invention
(1) The evaluation method provided by the invention can evaluate the two dimensions of the unmanned vehicle tracking safety and the riding comfort by adjusting the weight matrix according to different weights.
(2) The invention can adjust the evaluation weight of the direction error and the position error in the aspect of tracking safety. Meanwhile, the number of reference points brought into difference calculation can be selected, the evaluation dimensionality of the error model is further expanded, and the flexibility of the evaluation method is improved, so that the method can play a role in special scenes with high requirements on tracking smoothness.
(3) In the aspect of tracking comfort, the neutral steering yaw rate is calculated by adopting three reference point coordinates to serve as a reference yaw rate of an evaluation method, and the riding experience of passengers is considered to a great extent.
Drawings
FIG. 1 is a schematic view of the unmanned path planning of the present invention;
FIG. 2 is a simulation diagram of the present invention.
1. The unmanned vehicle, 2, tangent line of the expected path on the current reference point, 3, direction error, 4, central axis of the vehicle body, 5, expected path, 6, curvature center of the expected path on the current reference point, 7, curvature radius of the expected path on the current reference point, 8 and the current reference point; 9. position error, 10, unmanned vehicle yaw rate.
Detailed Description
The technical solutions in the examples of the present invention are clearly and completely described below with reference to the drawings in the examples 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 of the present invention without inventive step, are within the scope of the present invention.
The present invention will be described in further detail with reference to the accompanying drawings.
Example 1
The evaluation method provided by the invention comprises two latitude evaluation indexes: eerrorReflecting the safety degree of the trace, EcomfortThe ride comfort is reflected and both are weighted by a weight matrix.
As shown in fig. 1, the unmanned vehicle takes a desired path 5 as a control target, and an included angle between a central axis 4 of the vehicle body and a tangent 2 at a current reference point is a direction error 3 and is denoted by θ. The distance between the center of mass 1 of the vehicle body and the current reference point 8 is a position error 9 and is marked as e. The invention provides a tracking error model based on the direction error 3 and the position error 9, the model is not limited to the existing model, and the direction error 3 and the position error 9 can be added on the basis of the existing model path planning model, wherein the direction error reflects the deviation between the vehicle course and the expected course, the position error reflects the difference between the vehicle position and the expected position, and the two functions reflect the tracking effect of the unmanned vehicle together.
Based on such a tracking error model, future reference points and past reference points are further considered to improve the objectivity and continuity of the evaluation method on paths where the curvature changes greatly. The evaluation formula is proposed as follows:
wherein ek+1,ek,ek-1Forming direction error matrixes, namely the position error of the vehicle relative to the next reference point, the position error relative to the current reference point and the position error relative to the previous reference pointAnd (4) poor. A. thek+1,Ak, Ak-1And forming a position error coefficient matrix so as to perform weighted evaluation on the position errors of the vehicle relative to different reference points. In the actual evaluation, the weight coefficient A corresponding to the current position can be usedkSet to 1 and the rest to 0.2 or less. Likewise, θk+1,θk,θk-1Forming a directional error matrix, also representing the directional error of the vehicle with respect to future, current and past reference points, respectively, Bk+1,Bk,Bk-1Then the corresponding weighting matrix.
If the reference points are increased continuously, the objective accuracy of the evaluation method is not necessarily improved, but if the requirement of the evaluation method on the smoothness degree of the whole tracking process needs to be further improved, the method can be further expanded to n picks.
The invention relates to the actual yaw rate E of the vehicleyawYaw rate E generated by neutral steering of the vehicle along a reference pathrefThe difference value of (a) is used as a new factor for evaluating the ride attribute. Namely:
Ecomfort=k(Eyaw-Eref)
the neutral steering yaw rate is a reference yaw rate and can be calculated by the following formula.
Eref=v/r
Where r is the curvature radius 7 of the current reference point, a is the distance between the future reference point X and the current reference point Y, b is the distance between the current reference point Y and the past reference point, and c is the distance between the future reference point X and the past reference point. X, Y, Z are expressed by (x1, y1), (x2, y2), and (x3, y 3). As shown in FIG. 2, the method for quantitatively evaluating the error model comfort unmanned vehicle path tracking can realize more smooth and comfortable path tracking
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are also within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (7)
1. A comfortable unmanned vehicle path tracking evaluation method based on an error model is characterized by comprising the following steps:
evaluation indexes including two latitudes: eerrorReflecting the safety level of the trace, EcomfortThe riding comfort is reflected, and the riding comfort are weighted through the weight matrix;
the unmanned vehicle takes an expected path as a control target, and an included angle between a central axis of the vehicle body and a tangent line on a current reference point is taken as a direction error and is recorded as theta; the distance between the mass center of the vehicle body and the current reference point is a position error and is marked as e; establishing a tracking error model based on a direction error and a position error, wherein the direction error is the deviation between the vehicle course and the expected course, and the position error is the difference between the vehicle position and the expected position;
based on the direction error and the position error of the tracking error model, a future reference point and a past reference point are further considered so as to improve the objectivity and the continuity of the evaluation method on a path with large curvature change; the evaluation formula is proposed as follows:
wherein ek+1,ek,ek-1Forming a direction error matrix, namely a position error of the vehicle relative to a next reference point, a position error relative to a current reference point and a position error relative to a previous reference point; a. thek+1,Ak,Ak-1Form a position error coefficient matrix, where Ak+1,Ak,Ak-1Respectively representing the evaluation weights of the k-th time, k +1, k-1 and the like, so as to perform weighted evaluation on the position errors of the vehicle relative to different reference points; in the actual evaluation, the weight coefficient A corresponding to the current position can be usedkSet to 1, the remainder to 0.2 or less; likewise, θk+1,θk,θk-1Forming a directional error matrix, also representing the directional error of the vehicle with respect to future, current and past reference points, respectively, Bk+1,Bk,Bk-1Then the corresponding weighting matrix.
2. The method for tracking and evaluating the path of the comfortable unmanned vehicle based on the error model according to claim 1, characterized in that reference points are continuously increased, the requirement of the evaluation method on the smoothness degree of the whole tracking process is improved, and the method is expanded to n latitude:
3. the error model-based comfortable unmanned vehicle path tracking evaluation method according to claim 1, characterized in that: make the vehicle in factYaw rate EyawYaw rate E generated by neutral steering of the vehicle along a reference pathrefThe difference value of (a) is used as a new factor for evaluating the riding attribute; where k is the weight coefficient for comfort, i.e.:
Ecomfort=k(Eyaw-Eref)
the neutral steering yaw rate is a reference yaw rate and can be calculated by the following formula;
Eref=v/r
wherein r is the curvature radius 7 of the current reference point, v is the speed of the mass center of the rear axle of the vehicle, a is the distance between the future reference point X and the current reference point Y, b is the distance between the current reference point Y and the past reference point, and k is the weight coefficient of comfort level; v is the vehicle speed, c is the distance between the future reference point X and the past reference point; x, Y, Z are expressed by (x1, y1), (x2, y2), and (x3, y 3).
4. The utility model provides a comfortable unmanned vehicles route tracking evaluation system based on error model which characterized in that:
the evaluation index module comprises two latitude evaluation indexes: eerrorReflecting the safety level of the trace, EcomfortThen reflect multiplicationThe sitting comfort degree is weighted by the weight matrix;
the unmanned vehicle takes an expected path as a control target, and an included angle between a central axis of the vehicle body and a tangent line on a current reference point is taken as a direction error and is recorded as theta; the distance between the mass center of the vehicle body and the current reference point is a position error and is marked as e; establishing a tracking error model based on a direction error and a position error, wherein the direction error is the deviation between the vehicle course and the expected course, and the position error is the difference between the vehicle position and the expected position;
the error evaluation module is used for further considering a future reference point and a past reference point based on the direction error and the position error of the tracking error model so as to improve the objectivity and the continuity of the evaluation method on a path with larger curvature change; the evaluation formula is proposed as follows:
wherein ek+1,ek,ek-1Forming a direction error matrix, namely a position error of the vehicle relative to a next reference point, a position error relative to a current reference point and a position error relative to a previous reference point; a. thek+1,Ak,Ak-1Form a position error coefficient matrix, where Ak+1,Ak,Ak-1Respectively representing the evaluation weights of the k-th time, k +1, k-1 and the like, so as to perform weighted evaluation on the position errors of the vehicle relative to different reference points; in the actual evaluation, the weight coefficient A corresponding to the current position can be usedkSet to 1, the remainder to 0.2 or less; likewise, θk+1,θk,θk-1Forming a directional error matrix, also representing the directional error of the vehicle with respect to future, current and past reference points, respectively, Bk+1,Bk,Bk-1Then isA corresponding weighting matrix.
5. The method for tracking and evaluating the path of the comfortable unmanned vehicle based on the error model according to claim 4, wherein the multi-dimensional evaluation module continuously increases the reference points, improves the requirement of the evaluation method on the smoothness degree of the whole tracking process, and extends to n latitude:
6. the error model-based comfortable unmanned vehicle path tracking evaluation method of claim 4, further comprising a comfort evaluation module:
calculating the actual yaw rate E of the vehicleyawYaw rate E generated by neutral steering of the vehicle along a reference pathrefThe difference value of (a) is used as a new factor for evaluating the riding attribute; where k is the weight coefficient for comfort, i.e.:
Ecomfort=k(Eyaw-Eref)
the neutral steering yaw rate is a reference yaw rate, and can be calculated by the following formula:
Eref=v/r
wherein r is the curvature radius 7 of the current reference point, v is the speed of the mass center of the rear axle of the vehicle, a is the distance between the future reference point X and the current reference point Y, b is the distance between the current reference point Y and the past reference point, and k is the weight coefficient of comfort level; v is the vehicle speed, c is the distance between the future reference point X and the past reference point; x, Y, Z are expressed by (x1, y1), (x2, y2), and (x3, y 3).
7. The utility model provides a comfortable unmanned vehicles route tracking evaluation system based on error model which characterized in that includes:
a memory;
one or more processors, and
one or more programs stored in the memory and configured to be executed by the one or more processors, the programs when executed by the processors implementing the unmanned vehicle path tracking quantitative assessment method of any of claims 1-3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010939349.7A CN112102616A (en) | 2020-09-09 | 2020-09-09 | Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010939349.7A CN112102616A (en) | 2020-09-09 | 2020-09-09 | Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112102616A true CN112102616A (en) | 2020-12-18 |
Family
ID=73751127
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010939349.7A Pending CN112102616A (en) | 2020-09-09 | 2020-09-09 | Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112102616A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102358287A (en) * | 2011-09-05 | 2012-02-22 | 北京航空航天大学 | Trajectory tracking control method used for automatic driving robot of vehicle |
CN103234763A (en) * | 2013-04-09 | 2013-08-07 | 北京理工大学 | System and method for quantitatively evaluating unmanned vehicles |
CN107037818A (en) * | 2017-06-22 | 2017-08-11 | 奇瑞汽车股份有限公司 | A kind of path following method of pilotless automobile |
CN108108885A (en) * | 2017-12-15 | 2018-06-01 | 湖北汽车工业学院 | A kind of pilotless automobile tracking trajectory capacity evaluation method |
CN108958258A (en) * | 2018-07-25 | 2018-12-07 | 吉林大学 | A kind of track follow-up control method, control system and the relevant apparatus of unmanned vehicle |
CN109606379A (en) * | 2018-11-22 | 2019-04-12 | 江苏大学 | A kind of distributed driving automatic driving vehicle path trace fault tolerant control method |
CN110001637A (en) * | 2019-04-10 | 2019-07-12 | 吉林大学 | A kind of pilotless automobile path following control device and control method based on multiple spot tracking |
-
2020
- 2020-09-09 CN CN202010939349.7A patent/CN112102616A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102358287A (en) * | 2011-09-05 | 2012-02-22 | 北京航空航天大学 | Trajectory tracking control method used for automatic driving robot of vehicle |
CN103234763A (en) * | 2013-04-09 | 2013-08-07 | 北京理工大学 | System and method for quantitatively evaluating unmanned vehicles |
CN107037818A (en) * | 2017-06-22 | 2017-08-11 | 奇瑞汽车股份有限公司 | A kind of path following method of pilotless automobile |
CN108108885A (en) * | 2017-12-15 | 2018-06-01 | 湖北汽车工业学院 | A kind of pilotless automobile tracking trajectory capacity evaluation method |
CN108958258A (en) * | 2018-07-25 | 2018-12-07 | 吉林大学 | A kind of track follow-up control method, control system and the relevant apparatus of unmanned vehicle |
CN109606379A (en) * | 2018-11-22 | 2019-04-12 | 江苏大学 | A kind of distributed driving automatic driving vehicle path trace fault tolerant control method |
CN110001637A (en) * | 2019-04-10 | 2019-07-12 | 吉林大学 | A kind of pilotless automobile path following control device and control method based on multiple spot tracking |
Non-Patent Citations (2)
Title |
---|
PENG H 等: "A Theoretical and Experimental Study on Vehicle Lateral Control", 《AMERICAN CONTROL CONFERENCE》 * |
谭启凡: "四轮独立转向独立驱动车辆路径跟踪控制策略研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110001637B (en) | Unmanned vehicle path tracking control device and control method based on multipoint tracking | |
CN114655248A (en) | Transverse control method and device for automatic driving vehicle and vehicle | |
CN108995657A (en) | Operate the method and data processing system of automatic driving vehicle | |
CN109866752A (en) | Double mode parallel vehicles track following driving system and method based on PREDICTIVE CONTROL | |
CN113044064B (en) | Vehicle self-adaptive automatic driving decision method and system based on meta reinforcement learning | |
US20210165409A1 (en) | Adaptive Control of Autonomous or Semi-Autonomous Vehicle | |
CN114005280A (en) | Vehicle track prediction method based on uncertainty estimation | |
CN111238521A (en) | Path planning method and system for unmanned vehicle | |
Huang et al. | Conditional predictive behavior planning with inverse reinforcement learning for human-like autonomous driving | |
CN107169223A (en) | Vehicle parameter instruction method based on vehicle handling stability test evaluation system | |
CN112418646A (en) | Vehicle comfort evaluation method and device and readable storage medium | |
CN111338379B (en) | Collaborative path planning method for automatically controlled aircraft | |
Zhang et al. | An autonomous overtaking maneuver based on relative position information | |
CN109556609A (en) | A kind of collision prevention method and device based on artificial intelligence | |
CN111722306A (en) | Typhoon landing intensity prediction method and system based on TCN network model | |
CN113741464B (en) | Automatic driving speed control framework based on space-time data reinforcement learning | |
CN114384916A (en) | Adaptive decision-making method and system for off-road vehicle path planning | |
Dai et al. | A bargaining game-based human–machine shared driving control authority allocation strategy | |
CN112102616A (en) | Comfortable unmanned vehicle path tracking evaluation method, system and device based on error model | |
CN111123948B (en) | Vehicle multidimensional sensing fusion control method and system and automobile | |
CN114997048A (en) | Automatic driving vehicle lane keeping method based on TD3 algorithm improved by exploration strategy | |
WO2023045791A1 (en) | Lane keeping method and apparatus, device, medium, and system | |
CN115494849A (en) | Navigation control method and system for automatic driving vehicle | |
CN109035759A (en) | A kind of guardrail check and evaluation method | |
CN111275328A (en) | RNGRU (radio network unit) position marker part matching method based on comprehensive grey correlation sequence |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201218 |