CN116736852A - Vehicle obstacle avoidance planning method and system based on automatic driving - Google Patents
Vehicle obstacle avoidance planning method and system based on automatic driving Download PDFInfo
- Publication number
- CN116736852A CN116736852A CN202310679867.3A CN202310679867A CN116736852A CN 116736852 A CN116736852 A CN 116736852A CN 202310679867 A CN202310679867 A CN 202310679867A CN 116736852 A CN116736852 A CN 116736852A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- obstacle
- automatic driving
- obstacle avoidance
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims description 37
- 238000001914 filtration Methods 0.000 claims description 16
- 230000033001 locomotion Effects 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000007500 overflow downdraw method Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 13
- 238000011161 development Methods 0.000 description 7
- 230000018109 developmental process Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000005381 potential energy Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle obstacle avoidance planning method and system based on automatic driving, wherein the vehicle obstacle avoidance planning method based on automatic driving comprises the following steps: acquiring driving environment data of an automatic driving vehicle and vehicle self data by using a sensor and preprocessing the driving environment data and the vehicle self data; performing obstacle recognition on surrounding environment data based on the acquired driving environment data; constructing a vehicle dynamics model based on the acquired vehicle self data; judging a vehicle risk coefficient based on the identified obstacle and the vehicle dynamics model; performing obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient; and controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning. The invention can improve the active obstacle avoidance capability of the automatic driving vehicle, reduce the accident rate of the automatic driving vehicle in the running process, further improve the running safety of the vehicle and ensure the safety of a driver and passengers.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a vehicle obstacle avoidance planning method and system based on automatic driving.
Background
With the development of artificial intelligence technology and the continuous progress of the vehicle industry, the automatic driving technology of vehicles is also becoming mature. Autopilot, also called unmanned, refers to a technique in which a controller automatically and safely operates a vehicle to drive the vehicle on a road through cooperation of artificial intelligence, visual computing, a radar, a monitoring device, a global positioning system, and the like without driving the vehicle by a driver. The development of the automatic driving technology is not separated from the support of various advanced scientific technologies, the development of sensing and networking technologies can better provide rich, rapid and accurate environmental information for an automobile control system, the rapid development of a series of advanced technologies such as big data analysis technology, artificial intelligence and the like provides a more intelligent and efficient means for automobile control, and the automobile is integrated with the advanced technologies, so that various performances are greatly improved, and in sum, the rapid development of the emerging technology in the intelligent age provides a good opportunity for the development of the automobile driving automation industry.
The study of autopilot technology has focused mainly on four modules: environmental awareness, path planning, behavior decision and motion control, wherein the motion control is one of four core technical modules, and has extremely important roles in the development process of the automatic driving automobile. The active obstacle avoidance system of the automatic driving vehicle is the most effective and direct means for improving the driving safety, a safe and feasible path can be planned for the vehicle before the vehicle collides with an obstacle, traffic accidents are avoided, and track planning and tracking control are core technologies of the active obstacle avoidance system. However, because the situation on the road surface is complex, in the vehicle obstacle avoidance planning in the prior art, only the convex obstacle on the road surface can be often identified, the concave obstacles such as grooves, grooves and the like on the road surface cannot be identified, the obstacle on the road cannot be identified more comprehensively, and certain limitations exist.
For the problems in the related art, no effective solution has been proposed at present.
Disclosure of Invention
Aiming at the problems in the related art, the invention provides a vehicle obstacle avoidance planning method and system based on automatic driving, so as to overcome the technical problems in the prior related art.
For this purpose, the invention adopts the following specific technical scheme:
according to another aspect of the present invention, there is provided an automatic driving-based vehicle obstacle avoidance planning method, comprising the steps of:
s1, acquiring running environment data of an automatic driving vehicle and vehicle self data by using a sensor and preprocessing the running environment data;
s2, identifying obstacles on surrounding environment data based on the acquired driving environment data;
s3, constructing a vehicle dynamics model based on the acquired vehicle data;
s4, judging a vehicle risk coefficient based on the identified obstacle and the vehicle dynamics model;
s5, carrying out obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient;
s6, controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning.
Further, the obstacle recognition on the surrounding environment data based on the acquired surrounding environment data includes the following steps:
s11, acquiring front road data of an automatic driving vehicle through two ultrasonic sensors with different detection angles arranged on the front side of the automatic driving vehicle;
s12, collecting surrounding environment data of the automatic driving vehicle through laser radar sensors and vision sensors arranged around the automatic driving vehicle;
s13, denoising and filtering the acquired front road data and the acquired surrounding environment data.
Further, the obstacle recognition of the surrounding environment data based on the acquired running environment data includes the steps of:
s21, identifying a concave obstacle by adopting a cross detection method based on the front road data;
s22, identifying convex obstacles on the road based on the surrounding environment data.
Further, the identifying the concave obstacle by adopting the cross detection method based on the front road data comprises the following steps:
s211, defining two detection lines formed by two ultrasonic sensors with different detection angles on the front side of an automatic driving vehicle as a detection line I and a detection line II respectively;
s212, acquiring the vertical distance h between two ultrasonic sensors with different detection angles and the ground and the detection angles of the two ultrasonic sensors;
s213, based on the obtained vertical distance h and the detection angle, calculating the distance between the two ultrasonic sensors and the ground along the detection direction by using a trigonometric function, wherein the distances are respectively marked as OA and OB;
s214, calculating the sizes of the OA and the OB in real time, and when the calculated values of the OA and the OB float within a preset range, indicating that a concave obstacle does not appear in front of the automatic driving vehicle; when the calculated values of OA and OB appear to jump beyond a preset range, it is indicated that a concave obstacle appears in front of the autonomous vehicle.
Further, the identifying the convex obstacle on the road based on the surrounding environment data includes the steps of:
s221, transmitting convex obstacle information acquired by a laser radar sensor to a vision sensor, processing image information acquired by the vision sensor to obtain outline information of the convex obstacle, and identifying the obstacle;
s222, analyzing shadow features at the bottom of the obstacle, establishing an interested region, establishing a current region, and detecting whether the front convex obstacle is a dynamic obstacle or not;
s223, performing time fusion and space fusion on information acquired by a laser radar sensor and a vision sensor by using a data fusion method;
s224, fusing the time data and the space data fused by the combined Kalman filtering algorithm again, and determining the motion state of the front convex obstacle.
Further, the step of transmitting the convex obstacle information acquired by the laser radar sensor to the vision sensor, and processing the image information acquired by the vision sensor to obtain the outline information of the convex obstacle, and the step of identifying the obstacle comprises the following steps:
s2211, newly dividing the acquired image by using a ground dividing algorithm, and separating a road data part from a convex obstacle part;
s2212, carrying out feature extraction on the point cloud data of the convex obstacle part to obtain feature information;
s2213, classifying the convex obstacle by using a support vector machine classifier based on the extracted characteristic information, and determining the type of the convex obstacle.
Further, the determining the vehicle risk coefficient based on the identified obstacle and the vehicle dynamics model includes the steps of:
s41, acquiring the distance between the automatic driving vehicle and the obstacle;
s42, predicting the intersection point position and time of the automatic driving vehicle and the obstacle according to the form speed and direction of the automatic driving vehicle by a speed prediction method, and judging the risk coefficient of the automatic driving vehicle.
Further, the obstacle avoidance path planning for the automatic driving vehicle based on the vehicle risk coefficient comprises the following steps:
s51, based onArtificial potential field methodPlanning a coordinate of a next point to be reached by the automatic driving vehicle, and determining a turning radius required by the automatic driving vehicle when the automatic driving vehicle reaches the next coordinate point based on the pose of the previous moment;
s52, if the turning radius is smaller than the minimum turning radius of the automatic driving, calculating the coordinate of the next point according to the minimum turning radius of the automatic driving, and if the turning radius is larger than the minimum turning radius of the automatic driving, calculating the coordinate of the next point according to the turning radius;
s53, judging the relation between the calculated next coordinate point and the coordinate of the target point, if the two are coincident, finishing path planning, and if the two are not coincident, returningArtificial potential field methodAnd continuing to calculate until the coordinates of the target point are reached.
Further, the step of controlling the autonomous driving vehicle to perform obstacle avoidance driving according to the obstacle avoidance path planning includes the following steps:
s61, controlling the corresponding angle and the corresponding rotating speed of the wheel rotation of the automatic driving vehicle based on obstacle avoidance path planning;
s62, continuously acquiring the change of the running environment data in the moving obstacle avoidance process of the automatic driving vehicle, and adjusting the obstacle avoidance path and updating the control instruction.
According to another aspect of the present invention, there is provided an autonomous driving-based vehicle obstacle avoidance planning system, the system comprising: the system comprises a data acquisition module, an obstacle recognition module, a model construction module, a risk coefficient judgment module, a path planning module and a vehicle control module;
the data acquisition module is used for acquiring and preprocessing the running environment data of the automatic driving vehicle and the data of the vehicle by using the sensor;
the obstacle recognition module is used for recognizing obstacles on the surrounding environment data based on the acquired driving environment data;
the model construction module is used for constructing a vehicle dynamics model based on the acquired vehicle self data;
the risk coefficient judging module is used for judging the risk coefficient of the vehicle based on the identified obstacle and the vehicle dynamics model;
the path planning module is used for carrying out obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient;
the vehicle control module is used for controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning.
The beneficial effects of the invention are as follows:
1. according to the invention, the sensor is used for acquiring the running environment data of the automatic driving vehicle, identifying the obstacles around the automatic driving vehicle, judging the danger coefficient based on the vehicle dynamics model and the identified obstacles, and planning the obstacle avoidance path of the automatic driving vehicle, so that the active obstacle avoidance capability of the automatic driving vehicle can be improved, the accident rate of the automatic driving vehicle in the running process can be reduced, the running safety of the vehicle can be improved, and the safety of a driver and passengers can be ensured.
2. According to the invention, through the identification of the concave obstacle and the convex obstacle on the road, the comprehensive identification of the obstacle on the road can be realized, the automatic driving vehicle can be helped to more accurately sense the surrounding environment condition, the safer and more efficient running of the automatic driving vehicle can be realized, the running efficiency of the automatic driving vehicle is improved, and the cost is saved.
3. The invention is realized byArtificial potential field methodPlanning obstacle avoidance path for autonomous vehicles such thatArtificial potential field methodReasonable obstacle avoidance paths can be planned according to the distribution of the obstacles and the starting direction of the vehicle, and the curve is smooth and soft, so that the obstacle avoidance paths can be planned according to the movement characteristics and the steering characteristics of the automatic driving vehicle, and the trafficability of the automatic driving vehicle is guaranteed.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a vehicle obstacle avoidance planning method and system based on autopilot according to an embodiment of the present invention.
Detailed Description
For the purpose of further illustrating the various embodiments, the present invention provides the accompanying drawings, which are a part of the disclosure of the present invention, and which are mainly used to illustrate the embodiments and, together with the description, serve to explain the principles of the embodiments, and with reference to these descriptions, one skilled in the art will recognize other possible implementations and advantages of the present invention, wherein elements are not drawn to scale, and like reference numerals are generally used to designate like elements.
According to the embodiment of the invention, a vehicle obstacle avoidance planning method and system based on automatic driving are provided.
The invention will be further described with reference to the accompanying drawings and detailed description, as shown in fig. 1, according to an embodiment of the invention, there is provided an automatic driving-based vehicle obstacle avoidance planning method, which includes the steps of:
s1, acquiring running environment data of an automatic driving vehicle and vehicle self data by using a sensor and preprocessing the running environment data;
specifically, the obstacle recognition on the surrounding environment data based on the acquired surrounding environment data includes the following steps:
s11, acquiring front road data of an automatic driving vehicle through two ultrasonic sensors with different detection angles arranged on the front side of the automatic driving vehicle;
s12, collecting surrounding environment data of the automatic driving vehicle through laser radar sensors and vision sensors arranged around the automatic driving vehicle;
s13, denoising and filtering the acquired front road data and the acquired surrounding environment data.
Specifically, denoising and filtering are performed on the front road data and the surrounding environment data, so that the environment sensing capability and the driving safety of the automatic driving vehicle can be improved; the denoising and filtering processes can be generally performed by the following methods:
and (5) average value filtering: pixel values around a pixel are averaged to reduce random noise in the image.
Median filtering: taking the median value of the pixel values around a certain pixel as the pixel value to eliminate impulse noise in the image.
Gaussian filtering: pixels in the image are weighted averaged with a gaussian function to eliminate gaussian noise.
Wavelet transform denoising: the time and frequency characteristics of the signal are analyzed by wavelet transformation, the signal is decomposed into a plurality of components, and high-frequency noise is removed.
Edge preserving filtering: the edge information in the image is reserved, and the non-edge information is removed so as to reduce noise in the image.
Wavelet transform denoising: the image is converted to the wavelet domain, reducing noise by removing high frequency components in the wavelet coefficients.
The method can be selected and combined according to actual conditions, for example, the method can remove random noise by means of mean filtering or median filtering, and then retain important information in the image by means of edge retaining filtering or wavelet transformation denoising.
S2, identifying obstacles on surrounding environment data based on the acquired driving environment data;
specifically, the obstacle recognition for the surrounding environment data based on the acquired driving environment data includes the following steps:
s21, identifying a concave obstacle by adopting a cross detection method based on the front road data;
specifically, the identifying the concave obstacle by adopting the cross detection method based on the front road data comprises the following steps:
s211, defining two detection lines formed by two ultrasonic sensors with different detection angles on the front side of an automatic driving vehicle as a detection line I and a detection line II respectively;
specifically, the ultrasonic sensors on the front side of the mobile driving vehicle are all inclined downward and perform ultrasonic detection toward the front of the vehicle.
S212, acquiring the vertical distance h between two ultrasonic sensors with different detection angles and the ground and the detection angles of the two ultrasonic sensors; the detection angles of the two ultrasonic sensors are respectively set to alpha and beta;
s213, based on the obtained vertical distance h and the detection angle, calculating the distance between the two ultrasonic sensors and the ground along the detection direction by using a trigonometric function, wherein the distances are respectively marked as OA and OB;
specifically, a calculation formula for obtaining the distance between the two ultrasonic sensors and the ground along the detection direction based on the obtained vertical distance h and the detection angle by using a trigonometric function is as follows:
OA＝h/cosα
OB＝h/cos(α+β)
s214, calculating the sizes of the OA and the OB in real time, and when the calculated values of the OA and the OB float within a preset range, indicating that a concave obstacle does not appear in front of the automatic driving vehicle; when the calculated values of OA and OB appear to jump beyond a preset range, it is indicated that a concave obstacle appears in front of the autonomous vehicle.
S22, identifying convex obstacles on the road based on the surrounding environment data;
specifically, the identifying the convex obstacle on the road based on the surrounding environment data includes the following steps:
s221, transmitting convex obstacle information acquired by a laser radar sensor to a vision sensor, processing image information acquired by the vision sensor to obtain outline information of the convex obstacle, and identifying the obstacle;
specifically, the step of transmitting the convex obstacle information acquired by the laser radar sensor to the vision sensor, and processing the image information acquired by the vision sensor to obtain the outline information of the convex obstacle, and the step of identifying the obstacle comprises the following steps:
s2211, newly dividing the acquired image by using a ground dividing algorithm, and separating a road data part from a convex obstacle part;
specifically, the ground segmentation algorithm is generally based on an image processing technology and a machine learning method, wherein the ground segmentation algorithm based on the image processing technology mainly depends on the characteristics of colors, textures, shapes and the like of images, and the distribution condition of ground and non-ground pixel points can be obtained by performing operations such as preprocessing, filtering, threshold segmentation and the like on the images.
S2212, carrying out feature extraction on the point cloud data of the convex obstacle part to obtain feature information;
specifically, feature extraction is performed by calculating shape features such as curvature, normal vector, feature points and the like of the point cloud data; the curvature may be used to describe a degree of curvature of the point cloud data, the normal vector may be used to describe a surface normal direction of the point cloud data, and the feature point may be used to describe a local feature of the point cloud data. These shape features may help identify objects in the point cloud data, such as vehicles, pedestrians, etc.
S2213, classifying the convex obstacle by using a support vector machine classifier based on the extracted characteristic information, and determining the type of the convex obstacle.
S222, analyzing shadow features at the bottom of the obstacle, establishing an interested region, establishing a current region, and detecting whether the front convex obstacle is a dynamic obstacle or not;
specifically, the analysis of the shadow features of the bottom of the obstacle is specifically the shadow segmentation of the obstacle, which can be implemented by a threshold improvement algorithm of local gray statistics, and the specific algorithm is as follows:
selecting N local areas with the same size from near to far in the effective area of the road;
respectively carrying out gray value statistics on each local area to obtain gray average value and variance of each local area;
removing the local area with the local area variance value larger than 100;
calculating the gray mean mu and variance sigma of the N areas according to the gray mean and variance of each local area;
wherein,,
wherein N represents the number of local areas; mu (mu) i Representing the gray average value of the ith local area; sigma (sigma) i Representing the variance of the i-th local region.
Since the road surface gray level is always larger than the obstacle shadow gray level, the minimum road surface gray level can be selected to perform threshold segmentation on the obstacle shadow, namely t=μ -3σ.
And taking T as a threshold value to carry out binarization processing on the acquired image.
S223, performing time fusion and space fusion on information acquired by a laser radar sensor and a vision sensor by using a data fusion method;
s224, fusing the time data and the space data fused by the combined Kalman filtering algorithm again, and determining the motion state of the front convex obstacle.
S3, constructing a vehicle dynamics model based on the acquired vehicle data;
specifically, constructing a vehicle dynamics model based on the acquired vehicle self data includes the steps of:
defining the state of the vehicle, wherein the state of the vehicle comprises position, speed, acceleration, steering angle, vehicle body inclination angle and the like, and the state variables need to be determined according to actual conditions;
determining a control amount of the vehicle, the control amount of the vehicle including acceleration, braking force, steering angle, and the like; these control amounts may be acquired by sensors, controllers, etc. of the vehicle;
determining dynamics of the vehicle, the dynamics of the vehicle including mass, inertia, coefficient of friction, and the like; these characteristics need to be determined experimentally or by simulation;
establishing a motion equation, and establishing the motion equation of the vehicle according to the dynamics characteristics and the control quantity of the vehicle;
model parameters are determined, and various parameters in the model, such as the mass, inertia, friction coefficient and the like of the vehicle, are determined according to experimental or simulation methods and the like.
S4, judging a vehicle risk coefficient based on the identified obstacle and the vehicle dynamics model;
specifically, the determining the vehicle risk coefficient based on the identified obstacle and the vehicle dynamics model includes the following steps:
s41, acquiring the distance between the automatic driving vehicle and the obstacle;
s42, predicting the intersection point position and time of the automatic driving vehicle and the obstacle according to the form speed and direction of the automatic driving vehicle by a speed prediction method, and judging the risk coefficient of the automatic driving vehicle.
Specifically, predicting the intersection point position and time of the autonomous vehicle and the obstacle by the speed prediction method includes the steps of:
determining the position and speed of the vehicle and the obstacle;
calculating a distance between the vehicle and the obstacle;
from the speeds of the vehicle and the obstacle, and the distance between them, their closest distance and intersection time can be predicted.
S5, carrying out obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient;
specifically, the obstacle avoidance path planning for the automatic driving vehicle based on the vehicle risk coefficient includes the following steps:
s51, based onArtificial potential field methodPlanning a coordinate of a next point to be reached by the automatic driving vehicle, and determining a turning radius required by the automatic driving vehicle when the automatic driving vehicle reaches the next coordinate point based on the pose of the previous moment;
in particular, the method comprises the steps of,artificial potential field methodThe vehicle is regarded as a particle, and a potential field is built around the particle, so that a model algorithm is built. The artificial potential field method comprises the following steps: in a certain space, a repulsive force field and a gravitational field are defined and simultaneously exist, wherein the repulsive force field and the gravitational field are defined as being generated by an obstacle, and the gravitational field is defined as being generated by a destination point. The "gravitational field" causes the vehicle to advance toward the target point, while the "repulsive field" causes the vehicle to avoid the obstacle.
Wherein, the basic principle formula of the artificial potential field method is as follows:
P(θ)＝f{d(θ,θ 0 ),d[R(θ),0],d r }
in θ 0 Represents a target pose vector, θ represents the current pose of the vehicle, f represents a potential energy function, represents the potential energy of the vehicle at the current position θ, d (θ, θ) 0 ) Represents θ 0 And θ, d [ R (θ), 0]Represents the shortest distance between the vehicle and the obstacle, d r Represents a predetermined minimum influence range value, and P (θ) represents a variable d (θ, θ) 0 ) And is d [ R (θ), 0]Is a single reduction function of (2).
In a gravitational fieldIn the 'middle', the force exerted on the vehicle is determined by the target point and the 'attractive force' of the target point; the magnitude of the force and the vehicle
Inversely proportional to the distance of the target point, i.e. the farther the distance, the weaker the "attractive force" is.
In the "repulsive field", the force to which the vehicle is subjected is determined by the obstacle and its "repulsive force"; the magnitude of the force and the vehicle
Proportional to the distance of the obstacle, i.e. the closer the distance, the greater the "repulsive force" is.
According to the invention, on the basis of influence of the motion law and the motion characteristic of the vehicle on the path planning, the factor of the turning radius of the vehicle is considered on the obstacle avoidance path planning. The artificial potential field method can plan a reasonable obstacle avoidance path according to the distribution of obstacles and the starting direction of the vehicle, and the curve is smooth and soft, so that the obstacle avoidance path can be planned according to the motion characteristic and the steering characteristic of the intelligent vehicle, and the trafficability of the vehicle is ensured. S52, if the turning radius is smaller than the minimum turning radius of the automatic driving, calculating the coordinate of the next point according to the minimum turning radius of the automatic driving, and if the turning radius is larger than the minimum turning radius of the automatic driving, calculating the coordinate of the next point according to the turning radius;
s53, judging the relation between the calculated next coordinate point and the coordinate of the target point, if the two are coincident, finishing path planning, and if the two are not coincident, returningArtificial potential field methodAnd continuing to calculate until the coordinates of the target point are reached.
S6, controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning;
specifically, the step of controlling the autonomous driving vehicle to perform obstacle avoidance driving according to the obstacle avoidance path planning includes the following steps:
s61, controlling the corresponding angle and the corresponding rotating speed of the wheel rotation of the automatic driving vehicle based on obstacle avoidance path planning;
s62, continuously acquiring the change of the running environment data in the moving obstacle avoidance process of the automatic driving vehicle, and adjusting the obstacle avoidance path and updating the control instruction.
According to one embodiment of the present invention, there is provided an autonomous-based vehicle obstacle avoidance planning system, the system comprising: the system comprises a data acquisition module, an obstacle recognition module, a model construction module, a risk coefficient judgment module, a path planning module and a vehicle control module;
the data acquisition module is used for acquiring and preprocessing the running environment data of the automatic driving vehicle and the data of the vehicle by using the sensor;
the obstacle recognition module is used for recognizing obstacles on the surrounding environment data based on the acquired driving environment data;
the model construction module is used for constructing a vehicle dynamics model based on the acquired vehicle self data;
the risk coefficient judging module is used for judging the risk coefficient of the vehicle based on the identified obstacle and the vehicle dynamics model;
the path planning module is used for carrying out obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient;
the vehicle control module is used for controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning.
In summary, by means of the above technical solution of the present invention, the present invention obtains the driving environment data of the autonomous vehicle by using the sensor, identifies the obstacle around the autonomous vehicle, then determines the risk coefficient based on the vehicle dynamics model and the identified obstacle, and performs obstacle avoidance path planning on the autonomous vehicle, so as to improve the active obstacle avoidance capability of the autonomous vehicle, reduce the accident rate of the autonomous vehicle in the driving process, further improve the driving safety of the vehicle, and ensure the safety of the driver and the passengers; according to the invention, through the identification of the concave obstacle and the convex obstacle on the road, the comprehensive identification of the obstacle on the road can be realized, the automatic driving vehicle can be helped to more accurately sense the surrounding environment condition, the safer and more efficient running of the automatic driving vehicle can be realized, the running efficiency of the automatic driving vehicle is improved, and the cost is saved; the invention is realized byArtificial potential field methodPlanning obstacle avoidance path for autonomous vehicles such thatArtificial potential field methodReasonable obstacle avoidance paths can be planned according to the distribution of the obstacles and the starting direction of the vehicle, and the curve is smooth and soft, so that the obstacle avoidance paths can be planned according to the movement characteristics and the steering characteristics of the automatic driving vehicle, and the trafficability of the automatic driving vehicle is guaranteed.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. An automatic driving-based vehicle obstacle avoidance planning method is characterized by comprising the following steps:
s1, acquiring running environment data of an automatic driving vehicle and vehicle self data by using a sensor and preprocessing the running environment data;
s2, identifying obstacles on surrounding environment data based on the acquired driving environment data;
s3, constructing a vehicle dynamics model based on the acquired vehicle data;
s4, judging a vehicle risk coefficient based on the identified obstacle and the vehicle dynamics model;
s5, carrying out obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient;
s6, controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning.
2. The method for automatically driving-based obstacle avoidance planning of a vehicle according to claim 1, wherein the identifying the obstacle based on the acquired surrounding data comprises the steps of:
s11, acquiring front road data of an automatic driving vehicle through two ultrasonic sensors with different detection angles arranged on the front side of the automatic driving vehicle;
s12, collecting surrounding environment data of the automatic driving vehicle through laser radar sensors and vision sensors arranged around the automatic driving vehicle;
s13, denoising and filtering the acquired front road data and the acquired surrounding environment data.
3. The vehicle obstacle avoidance planning method based on automatic driving according to claim 2, characterized in that the obstacle recognition of the surrounding environment data based on the acquired running environment data comprises the steps of:
s21, identifying a concave obstacle by adopting a cross detection method based on the front road data;
s22, identifying convex obstacles on the road based on the surrounding environment data.
4. A vehicle obstacle avoidance planning method based on autopilot according to claim 3 wherein the identifying of concave obstacles using cross detection based on the forward road data comprises the steps of:
s211, defining two detection lines formed by two ultrasonic sensors with different detection angles on the front side of an automatic driving vehicle as a detection line I and a detection line II respectively;
s212, acquiring the vertical distance h between two ultrasonic sensors with different detection angles and the ground and the detection angles of the two ultrasonic sensors;
s213, based on the obtained vertical distance h and the detection angle, calculating the distance between the two ultrasonic sensors and the ground along the detection direction by using a trigonometric function, wherein the distances are respectively marked as OA and OB;
s214, calculating the sizes of the OA and the OB in real time, and when the calculated values of the OA and the OB float within a preset range, indicating that a concave obstacle does not appear in front of the automatic driving vehicle; when the calculated values of OA and OB appear to jump beyond a preset range, it is indicated that a concave obstacle appears in front of the autonomous vehicle.
5. A vehicle obstacle avoidance planning method based on autopilot as claimed in claim 3 wherein said identifying a convex obstacle on the roadway based on said ambient data comprises the steps of:
s221, transmitting convex obstacle information acquired by a laser radar sensor to a vision sensor, processing image information acquired by the vision sensor to obtain outline information of the convex obstacle, and identifying the obstacle;
s222, analyzing shadow features at the bottom of the obstacle, establishing an interested region, establishing a current region, and detecting whether the front convex obstacle is a dynamic obstacle or not;
s223, performing time fusion and space fusion on information acquired by a laser radar sensor and a vision sensor by using a data fusion method;
s224, fusing the time data and the space data fused by the combined Kalman filtering algorithm again, and determining the motion state of the front convex obstacle.
6. The method for planning obstacle avoidance planning of a vehicle based on automatic driving according to claim 5, wherein the step of transmitting the convex obstacle information acquired by the laser radar sensor to the vision sensor, processing the image information acquired by the vision sensor to obtain the contour information of the convex obstacle, and identifying the obstacle comprises the steps of:
s2211, newly dividing the acquired image by using a ground dividing algorithm, and separating a road data part from a convex obstacle part;
s2212, carrying out feature extraction on the point cloud data of the convex obstacle part to obtain feature information;
s2213, classifying the convex obstacle by using a support vector machine classifier based on the extracted characteristic information, and determining the type of the convex obstacle.
7. The method for automatically driving-based vehicle obstacle avoidance planning of claim 1, wherein the determining the vehicle risk factor based on the identified obstacle and the vehicle dynamics model comprises the steps of:
s41, acquiring the distance between the automatic driving vehicle and the obstacle;
s42, predicting the intersection point position and time of the automatic driving vehicle and the obstacle according to the form speed and direction of the automatic driving vehicle by a speed prediction method, and judging the risk coefficient of the automatic driving vehicle.
8. The method for planning obstacle avoidance of an autonomous vehicle according to claim 1, wherein the planning of the obstacle avoidance path of the autonomous vehicle based on the vehicle risk coefficient comprises the steps of:
s51, based onArtificial potential field methodPlanning a coordinate of a next point to be reached by the automatic driving vehicle, and determining a turning radius required by the automatic driving vehicle when the automatic driving vehicle reaches the next coordinate point based on the pose of the previous moment;
s52, if the turning radius is smaller than the minimum turning radius of the automatic driving, calculating the coordinate of the next point according to the minimum turning radius of the automatic driving, and if the turning radius is larger than the minimum turning radius of the automatic driving, calculating the coordinate of the next point according to the turning radius;
s53, judging the relation between the calculated next coordinate point and the coordinate of the target point, if the two are coincident, finishing path planning, and if the two are not coincident, returningArtificial potential field methodAnd continuing to calculate until the coordinates of the target point are reached.
9. The method for planning obstacle avoidance of an autopilot-based vehicle according to claim 1, wherein said controlling the autopilot-based vehicle to travel on obstacle avoidance path planning comprises the steps of:
s61, controlling the corresponding angle and the corresponding rotating speed of the wheel rotation of the automatic driving vehicle based on obstacle avoidance path planning;
s62, continuously acquiring the change of the running environment data in the moving obstacle avoidance process of the automatic driving vehicle, and adjusting the obstacle avoidance path and updating the control instruction.
10. An autopilot-based vehicle obstacle avoidance planning system for implementing the autopilot-based vehicle obstacle avoidance planning method of any one of claims 1-9, the system comprising: the system comprises a data acquisition module, an obstacle recognition module, a model construction module, a risk coefficient judgment module, a path planning module and a vehicle control module;
the data acquisition module is used for acquiring and preprocessing the running environment data of the automatic driving vehicle and the data of the vehicle by using the sensor;
the obstacle recognition module is used for recognizing obstacles on the surrounding environment data based on the acquired driving environment data;
the model construction module is used for constructing a vehicle dynamics model based on the acquired vehicle self data;
the risk coefficient judging module is used for judging the risk coefficient of the vehicle based on the identified obstacle and the vehicle dynamics model;
the path planning module is used for carrying out obstacle avoidance path planning on the automatic driving vehicle based on the vehicle risk coefficient;
the vehicle control module is used for controlling the automatic driving vehicle to carry out obstacle avoidance driving according to the obstacle avoidance path planning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310679867.3A CN116736852B (en) | 2023-06-08 | 2023-06-08 | Vehicle obstacle avoidance planning method and system based on automatic driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310679867.3A CN116736852B (en) | 2023-06-08 | 2023-06-08 | Vehicle obstacle avoidance planning method and system based on automatic driving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116736852A true CN116736852A (en) | 2023-09-12 |
| CN116736852B CN116736852B (en) | 2024-06-25 |
Family
ID=87916249
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310679867.3A Active CN116736852B (en) | 2023-06-08 | 2023-06-08 | Vehicle obstacle avoidance planning method and system based on automatic driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116736852B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117284281A (en) * | 2023-09-26 | 2023-12-26 | 上海大学 | Vehicle-mounted AR-HUD intelligent driving system based on laser radar |
| CN118377025A (en) * | 2024-06-24 | 2024-07-23 | 蚂蚁侠科技(深圳)有限公司 | Automatic driving adjustment and measurement method based on millimeter radar wave |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113848914A (en) * | 2021-09-28 | 2021-12-28 | 福州大学 | Collision coefficient artificial potential field method local path planning method in dynamic environment |
| CN114879207A (en) * | 2022-04-28 | 2022-08-09 | 湖北汽车工业学院 | Ultrasonic obstacle avoidance method for L4-level automatic driving vehicle |
| CN115416650A (en) * | 2022-09-16 | 2022-12-02 | 湖北汽车工业学院 | Intelligent driving obstacle avoidance system of vehicle |
| CN115620263A (en) * | 2022-10-25 | 2023-01-17 | 四川吉利学院 | Intelligent vehicle obstacle detection method based on image fusion of camera and laser radar |
-
2023
- 2023-06-08 CN CN202310679867.3A patent/CN116736852B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113848914A (en) * | 2021-09-28 | 2021-12-28 | 福州大学 | Collision coefficient artificial potential field method local path planning method in dynamic environment |
| CN114879207A (en) * | 2022-04-28 | 2022-08-09 | 湖北汽车工业学院 | Ultrasonic obstacle avoidance method for L4-level automatic driving vehicle |
| CN115416650A (en) * | 2022-09-16 | 2022-12-02 | 湖北汽车工业学院 | Intelligent driving obstacle avoidance system of vehicle |
| CN115620263A (en) * | 2022-10-25 | 2023-01-17 | 四川吉利学院 | Intelligent vehicle obstacle detection method based on image fusion of camera and laser radar |
Non-Patent Citations (2)
| Title |
|---|
| 李柠等: "履带式机器人中基于交叉探测法的凹形障碍物识别", 科技信息, no. 17, pages 82 - 83 * |
| 王伟: "基于信息融合的机器人障碍物检测与道路分割", 中国优秀硕士学位论文全文数据库（电子期刊）, pages 140 - 217 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117284281A (en) * | 2023-09-26 | 2023-12-26 | 上海大学 | Vehicle-mounted AR-HUD intelligent driving system based on laser radar |
| CN117284281B (en) * | 2023-09-26 | 2024-03-12 | 上海大学 | Vehicle-mounted AR-HUD intelligent driving system based on laser radar |
| CN118377025A (en) * | 2024-06-24 | 2024-07-23 | 蚂蚁侠科技(深圳)有限公司 | Automatic driving adjustment and measurement method based on millimeter radar wave |
| CN118377025B (en) * | 2024-06-24 | 2024-10-11 | 蚂蚁侠科技(深圳)有限公司 | Automatic driving adjustment and measurement method based on millimeter radar wave |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116736852B (en) | 2024-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116736852B (en) | Vehicle obstacle avoidance planning method and system based on automatic driving | |
| US9904855B2 (en) | Atomic scenes for scalable traffic scene recognition in monocular videos | |
| US11731639B2 (en) | Method and apparatus for lane detection on a vehicle travel surface | |
| CN111806433B (en) | Obstacle avoidance method, device and equipment for automatically driven vehicle | |
| CN111551957B (en) | Park low-speed automatic cruise and emergency braking system based on laser radar sensing | |
| CN109919074A (en) | A kind of the vehicle cognitive method and device of view-based access control model cognition technology | |
| US20240001962A1 (en) | Electronic control device | |
| CN112674646B (en) | Self-adaptive welting operation method based on multi-algorithm fusion and robot | |
| CN113119960B (en) | System and method for providing vehicle safety distance and speed warning under road-slip condition | |
| CN115416650A (en) | Intelligent driving obstacle avoidance system of vehicle | |
| CN115230729A (en) | Automatic driving obstacle avoidance method and system and storage medium | |
| CN116611603B (en) | Vehicle path scheduling method, device, computer and storage medium | |
| Javadi et al. | A robust vision-based lane boundaries detection approach for intelligent vehicles | |
| Kuan et al. | Pothole detection and avoidance via deep learning on edge devices | |
| CN116859962A (en) | Method and system for avoiding landing obstacle of aircraft | |
| CN112328651B (en) | Traffic target identification method based on millimeter wave radar data statistical characteristics | |
| CN115083199A (en) | Parking space information determination method and related equipment thereof | |
| CN114842660B (en) | Unmanned lane track prediction method and device and electronic equipment | |
| CN114779764B (en) | Vehicle reinforcement learning movement planning method based on driving risk analysis | |
| CN116215518A (en) | Unmanned mining card front road collision risk prediction and quantification method | |
| Kalms et al. | Robust lane recognition for autonomous driving | |
| He et al. | Monocular based lane-change on scaled-down autonomous vehicles | |
| CN111474940A (en) | Water drop sequencing algorithm for determining effective driving area of intelligent driving vehicle based on vehicle body posture | |
| Li et al. | Lane keeping of intelligent vehicle under crosswind based on visual navigation | |
| Lin et al. | Vehicle vision robust detection and recognition method |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240524 Address after: Room 109, No. 5, Lane 1688, Shenchang Road, Minhang District, Shanghai, 200335 Applicant after: Junjiu Electronic Technology (Shanghai) Co.,Ltd. Country or region after: China Address before: 225000 Wenchang East Road, Jiangdu District, Yangzhou, Jiangsu Province, No. 88 Applicant before: Yangzhou (Jiangdu) New Energy Automobile Industry Research Institute of Jiangsu University Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |