CN111650945B - Dynamic obstacle anticollision method - Google Patents
Dynamic obstacle anticollision method Download PDFInfo
- Publication number
- CN111650945B CN111650945B CN202010598009.2A CN202010598009A CN111650945B CN 111650945 B CN111650945 B CN 111650945B CN 202010598009 A CN202010598009 A CN 202010598009A CN 111650945 B CN111650945 B CN 111650945B
- Authority
- CN
- China
- Prior art keywords
- dynamic obstacle
- obstacle
- speed
- vehicle
- track
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 36
- 230000001133 acceleration Effects 0.000 claims description 20
- 230000009467 reduction Effects 0.000 claims description 8
- 230000004888 barrier function Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 description 10
- 230000004927 fusion Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000008447 perception Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/0088—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
- G05D1/0248—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means in combination with a laser
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- 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
- G05D1/0255—Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Optics & Photonics (AREA)
- Acoustics & Sound (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Electromagnetism (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
The application provides a dynamic obstacle anticollision method, which comprises the following steps: predicting linear prediction track information of the dynamic obstacle in a preset duration according to the instantaneous speed and the current position information of the dynamic obstacle; when the linear prediction track is intersected with the planned path and when the linear prediction track and the planned path are not in opposite directions, determining the region attribute of the dynamic obstacle according to the current position information of the vehicle and the current position information of the dynamic obstacle; according to the region attribute of the dynamic obstacle and the age of the dynamic obstacle, carrying out speed planning through a trapezoidal speed planning algorithm; the age of the dynamic obstacle is the time length counted from the moment of acquiring the obstacle information; and generating a control signal according to the planned speed, and sending the control signal to an actuator for execution. Thus, the safety of the vehicle is ensured.
Description
Technical Field
The application relates to the field of data processing, in particular to a dynamic obstacle anticollision method.
Background
In recent years, the unmanned technology is becoming more mature depending on the rapid development of artificial intelligence technology, and the daily life of people is gradually changed from the aspects of travel modes, unmanned logistics, special operation and the like. How to ensure the safety of an unmanned vehicle in an environment with dynamic obstacles such as pedestrians or vehicles is a difficult problem. In order to improve the safety of the unmanned vehicle, a set of dynamic obstacle anti-collision method with strong robustness and high operation efficiency needs to be developed, so that the unmanned vehicle can safely avoid dynamic pedestrians or vehicles.
The existing dynamic obstacle anticollision method firstly obtains the predicted track of the obstacle in a future period of time through a probability track model method, and the method considers the yaw rate and the uncertainty of the speed of the obstacle in track prediction and predicts the possible future running track of the obstacle. And then in the motion planning, judging the intersection of the future running track of the vehicle and the predicted track, and planning the deceleration or stopping according to the collision time during the intersection so as to realize the anti-collision.
The existing dynamic obstacle anticollision method is generally limited by perception precision, calculation performance and planning strategies, and mainly has three-point problems: (1) When the perceived obstacle data shake is larger, the problem that the planned self-vehicle track consistency is poor due to the larger shake of the tail section of the predicted track is caused, and the safety and the anti-collision cannot be ensured; (2) The algorithm complexity is high, the calculation time is long, and the method is not applicable to a vehicle-mounted calculation platform with limited calculation resources; (3) The method for planning the collision time can plan a more aggressive result and occupy the original running track of the pedestrian or the vehicle, so that the unsafe feeling of the pedestrian or the vehicle can be caused.
Disclosure of Invention
The embodiment of the application aims to provide a dynamic obstacle anticollision method, which solves the problems that safety anticollision cannot be ensured, algorithm complexity is high, calculation time is long, and the method is not suitable for a vehicle-mounted calculation platform with limited calculation resources and is unsafe for people or vehicles in the prior art.
To solve the above problems, in a first aspect, the present application provides a dynamic obstacle avoidance method, including:
acquiring obstacle information; the obstacle information includes dynamic obstacle information including an instantaneous speed and current position information of the dynamic obstacle;
predicting linear prediction track information of the dynamic obstacle in a preset duration according to the instantaneous speed and the current position information of the dynamic obstacle; the linear prediction track information comprises the direction and the length of a linear prediction track;
judging whether the linear predicted track is intersected with a planned path or not according to the direction of the linear predicted track, the length of the linear predicted track and the planned path;
when the linear predicted track is intersected with the planned path, judging whether the linear predicted track and the planned path are in opposite directions or not according to the direction of the linear predicted track and the planned path;
when the linear predicted track and the planned path are not in opposite directions, determining the region attribute of the dynamic obstacle according to the current position information of the vehicle and the current position information of the dynamic obstacle;
according to the regional attribute of the dynamic obstacle and the age of the dynamic obstacle, performing speed planning through a trapezoidal speed planning algorithm; the age of the dynamic barrier is the time length counted from the clustering moment;
and generating a control signal according to the planned speed, and sending the control signal to an actuator for execution.
In a possible implementation manner, the predicting, according to the instantaneous speed and the current position information of the dynamic obstacle, the linear predicted track information of the dynamic obstacle within the preset duration specifically includes:
determining the origin of the linear prediction track according to the current position information of the dynamic obstacle;
determining the direction of the linear prediction track according to the direction of the instantaneous speed of the dynamic obstacle;
and determining the length of the linear prediction track according to the product of the instantaneous speed of the dynamic obstacle and the prediction duration.
In one possible implementation, the method further includes:
when the linear predicted track and the local planning path are in non-opposite directions, determining collision points according to the linear predicted track information and the planning path;
determining a deceleration distance according to the current position information of the vehicle and a preset first acceleration threshold value; the deceleration distance is a distance for decelerating and driving the vehicle from the current position by a preset first acceleration threshold value;
and planning the distance between the deceleration distance and the collision point through a trapezoidal speed planning algorithm.
In one possible implementation manner, the speed planning specifically includes:
when the region attribute of the dynamic obstacle is a first region, judging whether the age of the dynamic obstacle is longer than a preset time period, and when the age of the dynamic obstacle is longer than the preset time period, calling a trapezoidal speed planning algorithm to plan parking;
when the area attribute of the dynamic obstacle is a second area, judging whether the age of the dynamic obstacle is longer than a preset time period, and when the age of the dynamic obstacle is longer than the preset time period, calling a trapezoidal speed planning algorithm to plan for speed reduction.
In a possible implementation manner, after predicting linear predicted track information of the dynamic obstacle within a preset duration according to the instantaneous speed and the current position information of the dynamic obstacle, the method further includes:
discretizing the linear prediction track to obtain cell points of a plurality of virtual barriers; and the cell points of the virtual obstacle are contour points of the dynamic obstacle on the linear prediction track.
In one possible implementation manner, the trapezoidal speed planning algorithm specifically includes:
calculating a pre-aiming distance according to the speed information of the vehicle, a preset second acceleration threshold value, a safety distance and a preset distance constant;
judging whether the pre-aiming distance is intersected with a cell point of the virtual obstacle or not;
when the pre-aiming distance intersects with the cell point of the virtual obstacle, determining the cell point of the intersected virtual obstacle, and setting the speed of the road point within the safe distance before the cell point of the intersected virtual obstacle and the road point between the cell point of the intersected virtual obstacle and the end point of the planned path to be 0.
In one possible implementation manner, the calculating the pre-aiming distance according to the speed information of the vehicle, the preset second acceleration threshold value, the safety distance and the preset distance constant specifically includes:
the square of the speed information of the vehicle is divided by a preset second acceleration threshold value, and the safety distance and a preset distance constant are added to obtain the preset aiming distance.
In a second aspect, the application provides an apparatus comprising a memory for storing a program and a processor for performing the method of any of the first aspects.
In a third aspect, the application provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method according to any of the first aspects.
In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, which when executed by a processor performs a method according to any of the first aspects.
The dynamic obstacle anticollision method provided by the embodiment of the application has the following technical effects:
(1) The application predicts the track of the dynamic obstacle by using the linear prediction model, discretizes the predicted track into track points which are used as cell points of the virtual obstacle, simplifies the obstacle prediction model, simplifies the time domain solving problem into the space anti-collision problem, and does not have the safety problem of anti-collision failure caused by the jitter of collision time.
(2) The linear prediction model algorithm disclosed by the application is low in complexity and suitable for a vehicle-mounted computing platform with limited computing resources.
(3) The regional division method and the obstacle age method effectively filter dynamic obstacles in non-dangerous regions, and reduce false triggering of an anti-collision function.
(4) According to the application, when the self-vehicle track is interfered to the running track of dynamic obstacles such as pedestrians or vehicles, the vehicle can be stopped and waited until the obstacles are far away and the vehicle can continue to run without collision risk, so that the safety is effectively ensured.
Drawings
FIG. 1 is a schematic diagram of a software system according to a first embodiment of the present application;
FIG. 2 is a view of a dynamic obstacle avoidance scene provided in accordance with an embodiment of the present application;
FIG. 3 is a schematic diagram of a dynamic obstacle avoidance process according to an embodiment of the present application;
FIG. 4 is a graph showing the effect of a trapezoidal speed planning algorithm according to a first embodiment of the present application;
fig. 5 shows an embodiment of the present application.
Detailed Description
The application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be noted that, for convenience of description, only the portions related to the present application are shown in the drawings.
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
Fig. 1 is a schematic diagram of a software system according to an embodiment of the application. The system consists of a vehicle-mounted sensor, a vehicle-mounted operation unit and an actuator, wherein a dynamic obstacle anticollision function is integrated in the vehicle-mounted operation unit. The dynamic obstacle anticollision function relates to a plurality of key technologies such as environment perception, behavior prediction, speed planning in motion planning and the like, and a specific implementation scheme will be described in detail below.
Firstly, the fusion positioning obtains the vehicle state from the vehicle-mounted sensor, and the sensing fusion obtains the surrounding information of the vehicle from the vehicle-mounted sensor, and mainly comprises the vehicle speed and the position and speed information of surrounding obstacles. Secondly, behavior prediction predicts a straight-line predicted trajectory of the dynamic obstacle in a future period of time based on the position and instantaneous speed of the dynamic obstacle. And thirdly, judging whether the motion planning is intersected with the linear prediction track of the dynamic obstacle on the basis of the planned path, if so, planning to slow down or stop the vehicle near the intersection point, and correspondingly generating acceleration and deceleration requirements by the control module. Finally, the executor actively controls the accelerator and the brake system of the vehicle after receiving the acceleration and deceleration requirement, and adjusts the speed of the vehicle, so that the collision between the vehicle and the dynamic obstacle is avoided.
Fig. 2 is a view of a dynamic obstacle collision avoidance scene provided in an embodiment of the present application, in order to avoid the problem of larger jitter at the end of a predicted track in the prior art, the present application converts a linear predicted track of a dynamic obstacle in a time domain into an obstacle collision avoidance problem in a two-dimensional space by using a linear prediction method, so as to reduce collision risk caused by jitter of collision time. The contour of the virtual obstacle is represented by obtaining linear predicted trajectory information of the dynamic obstacle through the instantaneous speed and age of the dynamic obstacle, sampling the linear predicted trajectory at fixed time intervals to obtain discrete trajectory points (dashed lines in fig. 2), and taking the trajectory points as cell points of the virtual obstacle.
In addition, the speed calculation of the perception fusion on the low-speed obstacle is not accurate enough, and particularly the course of the low-speed target can change greatly, so that the predicted output track swing is obvious. To reduce the problem of predictive jitter, an obstacle is considered a dynamic obstacle when its speed exceeds 0.2m/s, and the dynamic obstacle is tracked to be given an age attribute for downstream algorithms. Here the age attribute is the duration counted from the clustering moment. The clustering is one of the acquired laser point cloud processing modes for the perception fusion, for example, the laser point cloud clustering can be carried out through a point cloud European clustering algorithm.
Fig. 3 is a schematic diagram of a dynamic obstacle avoidance process according to an embodiment of the present application. The execution main body of the application is a vehicle-mounted operation unit, the application is applied to an unmanned vehicle, as shown in fig. 3, and the application comprises the following steps:
step 110, obtaining obstacle information; the obstacle information includes dynamic obstacle information including an instantaneous speed and current position information of the dynamic obstacle.
Specifically, various vehicle-mounted sensors such as a laser radar, an ultrasonic radar, a camera and a vision module are mounted on a vehicle, sensing fusion is carried out on various environmental information acquired by the vehicle-mounted sensors, for example, after point cloud data are removed from the ground, point clouds on the ground represent real obstacles, and static obstacle information and dynamic obstacle information are determined after clustering. Here, in order to reduce the shake of the straight-line predicted trajectory direction information, an obstacle speed exceeding a preset speed threshold, such as 0.2m/s, is determined as a dynamic obstacle, otherwise considered as a static obstacle.
Step 120, predicting linear prediction track information of the dynamic obstacle in a preset duration according to the instantaneous speed and the current position information of the dynamic obstacle; the linear prediction trajectory information includes a direction and a length of the linear prediction trajectory.
The method is limited by the characteristics of the sensor and the algorithm boundary, the perceived and fused output obstacle position, speed and yaw rate information have certain deviation, if the obstacle track is predicted by adopting a probability track model, larger jitter occurs at the end section of the predicted track, so that the downstream decision and motion planning have poor consistency, and the safety cannot be ensured. In addition, the algorithm for predicting the obstacle track by the probability track model has high complexity, long calculation time consumption and is not suitable for a vehicle-mounted platform with limited calculation resources.
The complexity of the linear model prediction algorithm is low, the jitter of the end section of the prediction track is small, and the linear model prediction algorithm can basically accord with the real motion track of the obstacle when the high-frequency real-time prediction is performed and the prediction duration is not long. The application therefore adopts linear model prediction, i.e. predicts the linear track of the dynamic obstacle in the future period of time according to the instantaneous speed of the dynamic obstacle.
Wherein step 120 comprises: firstly, determining an origin of a linear prediction track according to current position information of a dynamic obstacle;secondly, determining the direction of a linear prediction track according to the direction of the instantaneous speed of the dynamic obstacle; and finally, determining the length of the linear prediction track according to the product of the instantaneous speed of the dynamic obstacle and the prediction duration. I.e. the direction of the straight track is the current speed direction, the length S is determined by the instantaneous speed V of the obstacle obj Is determined by the predicted time period t, i.e. s=v obj ×t。
And 130, judging whether the linear predicted track is intersected with the planned path according to the direction of the linear predicted track, the length of the linear predicted track and the planned path.
Specifically, during the running process of the vehicle, the motion planning can plan the path in advance according to the fusion positioning and sensing fusion data, and can judge whether the linear prediction track is intersected with the path according to the direction and length of the linear prediction track and the planned path. When intersecting, step 140 is performed.
And 140, judging whether the linear predicted track and the local planned path are opposite directions or not according to the direction of the linear predicted track and the local planned path when the linear predicted track and the planned path are intersected.
Specifically, when the straight-line predicted trajectory intersects the planned path, the determination is continued as to whether the straight-line predicted trajectory is in the opposite direction or the same direction, and the direction of the planned path is the same direction.
And 150, determining the region attribute of the dynamic obstacle according to the current position information of the vehicle and the current position information of the dynamic obstacle when the linear predicted track and the planned path are not opposite directions.
Specifically, when the straight-line predicted trajectory is in the same direction as the planned path, it may be determined in which region of the vehicle the dynamic obstacle is located, i.e., the region attribute, which includes the first region, the second region, and the third region, and then step 160 is performed. Referring to fig. 2, in the vehicle coordinate system, a midpoint position of a rear axle of the vehicle is taken as an origin, a longitudinal direction of the vehicle is taken as an x-axis direction, a transverse direction of the vehicle is taken as a y-axis direction, a vertical upward direction is taken as a z-axis direction, regions around the vehicle can be divided according to a rigid body structure of the vehicle under the vehicle body coordinate system, a first region is a parking region, a second region is a deceleration region, and a third region is a blind region. The parking areas and the speed reduction areas are divided according to threat degrees of obstacles in different directions, for example, the areas outside the speed reduction areas in front and behind, namely, the areas outside the speed reduction areas, namely, the parking areas, have the greatest threat of the obstacles, so that parking is needed; the obstruction of the rear blind area should not be taken into consideration; the dead zone is in transition with the parking zone, so that shaking of decision of whether to park or not due to shaking of the position of an obstacle is avoided. Therefore, three areas around the vehicle are determined, and different speed planning is conveniently carried out on dynamic obstacles in different areas.
Further, when the linear predicted trajectory and the planned path are in opposite directions, determining collision points according to the linear predicted trajectory information and the planned path; then determining a deceleration distance according to the current position information of the vehicle and a preset first acceleration threshold value; and planning the deceleration distance and the distance between collision points through a trapezoidal speed planning algorithm. The deceleration distance is a distance for the vehicle to run at a deceleration with a preset first deceleration threshold value from the current position. The preset first deceleration threshold value is related to the type of the vehicle and the current speed information of the vehicle, for example, if the current speed information of the vehicle is 60km/h when the vehicle is medium-sized, the preset first acceleration threshold value is-0.5 m/s, and corresponding first acceleration threshold values are preset after each specific speed information of each vehicle type, the corresponding relation between the speed information and the first acceleration information can be stored in a motion planning in a first comparison table mode, and when the planning is performed through a trapezoidal speed planning algorithm, the first acceleration threshold value can be determined through searching the first comparison table after the current speed information of the vehicle is determined.
Therefore, if the vehicle is a dynamic obstacle opposite to the planned path of the vehicle, the vehicle can be planned to be decelerated to a lower speed in advance by a certain distance, and the vehicle can be enabled to run at a constant speed at the lower speed, so that collision is avoided.
Step 160, performing speed planning through a trapezoidal speed planning algorithm according to the region attribute of the dynamic obstacle and the age of the dynamic obstacle; the age of a dynamic obstacle is the length of time counted from the acquisition of obstacle information.
Specifically, when the region attribute of the dynamic obstacle is a first region, judging whether the age of the dynamic obstacle is longer than a preset time period, and when the age of the dynamic obstacle is longer than the preset time period, calling a trapezoidal speed planning algorithm to plan parking; when the area attribute of the dynamic obstacle is the second area, judging whether the age of the dynamic obstacle is longer than a preset time period, and when the age of the dynamic obstacle is longer than the preset time period, calling a trapezoidal speed planning algorithm to plan for speed reduction.
Further, after step 120, the present application may further include: discretizing the linear prediction track to obtain cell points of a plurality of virtual barriers. Subsequently, speed planning can be performed in a trapezoidal speed planning algorithm through cell points of the virtual obstacle.
Referring to fig. 4, v represents a speed after planning by a trapezoidal speed planning algorithm, S represents an accumulated distance corresponding to the planned speed, and in the trapezoidal speed planning algorithm, a passing speed or a parking position of the vehicle can be planned according to current position information of the obstacle. The trapezoidal programming algorithm is described below:
specifically, firstly, calculating a pre-aiming distance according to speed information of a vehicle, a preset second acceleration threshold value, a safety distance and a preset distance constant; secondly, judging whether the pretightening distance is intersected with a cell point of the virtual barrier or not; and finally, when the pre-aiming distance is intersected with the cell point of the virtual obstacle, determining the cell point of the intersected virtual obstacle, and setting the speed of the road point in the safety distance before the cell point of the intersected virtual obstacle and the road point between the cell point of the intersected virtual obstacle and the end point of the planned path to be 0. Then, the road points between the current position information of the vehicle and the safe distance of the cell points of the intersected virtual obstacle can be smoothed, so that a new planning path can be obtained through motion planning, and the new planning path is output to the control.
Therefore, through the trapezoidal speed planning algorithm, when the cell point of the virtual obstacle is within the pre-aiming distance, the path of the road point before the safe distance of the cell point of the intersected virtual obstacle and the road point between the current position information of the vehicle can be planned, so that the vehicle is planned to stop before reaching the safe distance of the cell point of the virtual obstacle.
The square of the speed information of the vehicle is divided by a preset second acceleration threshold value, and the safety distance and a preset distance constant are added to obtain the preset aiming distance. The specific formula is as follows:
wherein Dist is a pre-aiming distance, v is speed information of the vehicle, the speed information can be obtained in real time through a vehicle-mounted sensor, a is a preset second, safe is a safety distance, and Offset is a set distance constant. The preset second acceleration is generally-0.2 to-1 m/s 2 The safety distance is a configurable distance of how much distance before the obstacle is parked, is related to the model of the vehicle, is a set value, and the distance constant is usually the distance from the rear axle to the front suspension of the vehicle body blind area part, is related to the model of the vehicle, and is a set value.
When the cell point of the virtual obstacle occupies the road in front of the vehicle in the parking area, the vehicle can park and wait in a safe distance until the dynamic obstacle is far away and no collision risk exists, the vehicle can continue to run, the vehicle can not interfere with the original track of the dynamic obstacle, the situation that the vehicle is accelerated to rush through the original collision point can not occur, and the safety is effectively ensured.
And step 170, generating a control signal according to the planned speed, and sending the control signal to the actuator so as to enable the actuator to execute.
Specifically, a control module in the vehicle-mounted operation unit generates control signals according to the planned speed, and controls the control signals, wherein the control signals comprise steering control signals and torque control signals, the steering of the vehicle is controlled through the steering control signals, the speed of the vehicle is controlled through the torque control signals, and the control signals are sent to the actuator, so that the actuator controls the steering of the vehicle according to the steering control signals, and the speed of the vehicle is controlled according to the torque control signals.
The application is further described below in connection with fig. 3 and 5.
Step 201, start.
That is, after the vehicle is electrified, the vehicle-mounted sensor, the vehicle-mounted operation unit and the actuator are started to start working.
Step 202, determining whether the straight-line predicted trajectory intersects with the path planned by the vehicle, executing step 203 when intersecting, and executing step 210 when not intersecting.
The specific description in step 120 may be referred to, where the linear prediction track information may be obtained by using a linear prediction model, and it may be determined whether the linear prediction track and the path planned by the vehicle intersect, and for the intersection or the non-intersection, different steps are executed respectively.
Step 203, determining whether the path planned by the own vehicle and the linear prediction track are in opposite directions, if so, executing step 204, and if not, executing step 205.
Step 204, planning deceleration in advance within the deceleration distance.
When the deceleration is planned in advance within the deceleration distance, the planning can be performed according to a trapezoidal speed planning algorithm.
Step 205, determining the region attribute of the dynamic barrier.
The regional attribute comprises three types of parking areas, speed reduction areas and blind areas. Step 206 is performed when the dynamic obstacle is in the parking area, step 207 is performed when the dynamic obstacle is in the deceleration area, and step 210 is performed when the dynamic obstacle is in the blind area.
Step 206, when the dynamic obstacle is in the parking area, judging whether the dynamic obstacle is older than t seconds.
Wherein t seconds is a preset time period, when the vehicle is in the parking area, the step 208 is executed when the obstacle age is greater than t seconds, and when the obstacle age is not greater than t seconds, the step 210 is executed.
Step 207, when the dynamic obstacle is in the deceleration zone, it is determined whether the dynamic obstacle is older than t seconds, and when the dynamic obstacle is older than t seconds, step 209 is executed, and when the dynamic obstacle is older than t seconds, step 210 is executed.
Step 208, planning to park.
Step 209, planning to reduce the speed.
Step 210, disregard.
Step 211, when parking is planned or speed reduction is planned, path planning is performed according to a trapezoidal speed planning algorithm.
Step 212, end.
The dynamic obstacle anticollision method provided by the embodiment of the application has the following technical effects:
(1) The application predicts the track of the dynamic obstacle by using the linear prediction model, discretizes the predicted track into track points which are used as cell points of the virtual obstacle, simplifies the obstacle prediction model, simplifies the time domain solving problem into the space anti-collision problem, and does not have the safety problem of anti-collision failure caused by the jitter of collision time.
(2) The linear prediction model algorithm disclosed by the application is low in complexity and suitable for a vehicle-mounted computing platform with limited computing resources.
(3) The regional division method and the obstacle age method effectively filter dynamic obstacles in non-dangerous regions, and reduce false triggering of an anti-collision function.
(4) According to the application, when the self-vehicle track is interfered to the running track of dynamic obstacles such as pedestrians or vehicles, the vehicle can be stopped and waited until the obstacles are far away and the vehicle can continue to run without collision risk, so that the safety is effectively ensured.
The second embodiment of the application provides an apparatus, which includes a memory and a processor, where the memory is configured to store a program, and the memory may be connected to the processor through a bus. The memory may be non-volatile memory, such as a hard disk drive and flash memory, in which software programs and device drivers are stored. The software program can execute various functions of the method provided by the embodiment of the application; the device driver may be a network and interface driver. The processor is configured to execute a software program, where the software program is executed to implement the method provided in the first embodiment of the present application.
A third embodiment of the present application provides a computer program product containing instructions, which when executed on a computer, cause the computer to perform the method provided by the first embodiment of the present application.
The fourth embodiment of the present application provides a computer readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the method provided in the first embodiment of the present application is implemented.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of function in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The foregoing detailed description of the application has been presented for purposes of illustration and description, and it should be understood that the application is not limited to the particular embodiments disclosed, but is intended to cover all modifications, equivalents, alternatives, and improvements within the spirit and principles of the application.
Claims (9)
1. A method of dynamic obstacle avoidance, the method comprising:
acquiring obstacle information; the obstacle information includes dynamic obstacle information including an instantaneous speed and current position information of the dynamic obstacle;
predicting linear prediction track information of the dynamic obstacle in a preset duration according to the instantaneous speed and the current position information of the dynamic obstacle; the linear prediction track information comprises the direction and the length of a linear prediction track;
judging whether the linear predicted track is intersected with a planned path or not according to the direction of the linear predicted track, the length of the linear predicted track and the planned path;
when the linear predicted track is intersected with the planned path, judging whether the linear predicted track and the planned path are in opposite directions or not according to the direction of the linear predicted track and the planned path;
when the linear predicted track and the planned path are not opposite directions, determining the region attribute of the dynamic obstacle according to the current position information of the vehicle and the current position information of the dynamic obstacle; the area comprises a first area and a second area, wherein the first area has a parking attribute, and the second area has a deceleration attribute;
according to the regional attribute of the dynamic obstacle and the age of the dynamic obstacle, performing speed planning through a trapezoidal speed planning algorithm; the age of the dynamic barrier is the time length counted from the clustering moment;
generating a control signal according to the planned speed, and sending the control signal to an executor for execution;
the speed planning specifically includes:
when the area where the dynamic obstacle is located is a first area, judging whether the age of the dynamic obstacle is longer than a preset time period, and when the age of the dynamic obstacle is longer than the preset time period, calling a trapezoidal speed planning algorithm to plan parking;
when the area where the dynamic obstacle is located is a second area, judging whether the age of the dynamic obstacle is longer than a preset time period, and when the age of the dynamic obstacle is longer than the preset time period, calling a trapezoidal speed planning algorithm to plan for speed reduction.
2. The method of claim 1, wherein predicting linear predicted trajectory information of the dynamic obstacle within a preset duration according to the instantaneous speed and the current position information of the dynamic obstacle specifically comprises:
determining the origin of the linear prediction track according to the current position information of the dynamic obstacle;
determining the direction of the linear prediction track according to the direction of the instantaneous speed of the dynamic obstacle;
and determining the length of the linear prediction track according to the product of the instantaneous speed of the dynamic obstacle and the prediction duration.
3. The method according to claim 1, wherein the method further comprises:
when the linear predicted track and the local planning path are in non-opposite directions, determining collision points according to the linear predicted track information and the planning path;
determining a deceleration distance according to the current position information of the vehicle and a preset first acceleration threshold value; the deceleration distance is a distance for decelerating and driving the vehicle from the current position by a preset first acceleration threshold value;
and planning the distance between the deceleration distance and the collision point through a trapezoidal speed planning algorithm.
4. The method according to claim 1, wherein after predicting linear predicted trajectory information of the dynamic obstacle within a preset time period according to the instantaneous speed and the current position information of the dynamic obstacle, the method further comprises:
discretizing the linear prediction track to obtain cell points of a plurality of virtual barriers; and the cell points of the virtual obstacle are contour points of the dynamic obstacle on the linear prediction track.
5. The method according to claim 4, wherein the trapezoidal speed planning algorithm is specifically:
calculating a pre-aiming distance according to the speed information of the vehicle, a preset second acceleration threshold value, a safety distance and a preset distance constant;
judging whether the pre-aiming distance is intersected with a cell point of the virtual obstacle or not;
when the pre-aiming distance intersects with the cell point of the virtual obstacle, determining the cell point of the intersected virtual obstacle, and setting the speed of the road point within the safe distance before the cell point of the intersected virtual obstacle and the road point between the cell point of the intersected virtual obstacle and the end point of the planned path to be 0.
6. The method according to claim 5, wherein calculating the pre-aiming distance based on the speed information of the vehicle, the preset second acceleration threshold, the safety distance, and the preset distance constant comprises:
the square of the speed information of the vehicle is divided by a preset second acceleration threshold value, and the safety distance and a preset distance constant are added to obtain the preset aiming distance.
7. An apparatus comprising a memory for storing a program and a processor for performing the method of any of claims 1-6.
8. A computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of any of claims 1-6.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, performs the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010598009.2A CN111650945B (en) | 2020-06-28 | 2020-06-28 | Dynamic obstacle anticollision method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010598009.2A CN111650945B (en) | 2020-06-28 | 2020-06-28 | Dynamic obstacle anticollision method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111650945A CN111650945A (en) | 2020-09-11 |
| CN111650945B true CN111650945B (en) | 2023-10-24 |
Family
ID=72343134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010598009.2A Active CN111650945B (en) | 2020-06-28 | 2020-06-28 | Dynamic obstacle anticollision method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111650945B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112631276A (en) * | 2020-12-03 | 2021-04-09 | 长沙行深智能科技有限公司 | Unmanned vehicle dynamic obstacle decision method, system, medium and equipment |
| CN112327888B (en) * | 2021-01-07 | 2021-03-30 | 中智行科技有限公司 | Path planning method and device, electronic equipment and storage medium |
| CN113110413B (en) * | 2021-03-10 | 2022-11-08 | 成都永奉科技有限公司 | Following robot, following control method and following control system |
| CN113156947B (en) * | 2021-04-14 | 2024-03-08 | 武汉理工大学 | Method for planning path of ship in dynamic environment |
| CN113867365B (en) * | 2021-10-28 | 2024-05-14 | 广州文远知行科技有限公司 | Method and device for determining variable acceleration of unmanned vehicle and related equipment |
| CN114323044A (en) * | 2021-12-16 | 2022-04-12 | 中汽创智科技有限公司 | Path planning method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017165687A1 (en) * | 2016-03-24 | 2017-09-28 | Honda Motor Co., Ltd. | System and method for trajectory planning for unexpected pedestrians |
| CN109712421A (en) * | 2019-02-22 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | The speed planning method, apparatus and storage medium of automatic driving vehicle |
| CN110389581A (en) * | 2018-04-17 | 2019-10-29 | 百度（美国）有限责任公司 | Method for the prediction locus for automatic driving vehicle dyspoiesis object |
| CN111289008A (en) * | 2020-04-28 | 2020-06-16 | 南京维思科汽车科技有限公司 | Local path planning algorithm for unmanned vehicle |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10800408B2 (en) * | 2018-05-24 | 2020-10-13 | Baidu Usa Llc | Determining driving paths for autonomous driving that avoid moving obstacles |
-
2020
- 2020-06-28 CN CN202010598009.2A patent/CN111650945B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017165687A1 (en) * | 2016-03-24 | 2017-09-28 | Honda Motor Co., Ltd. | System and method for trajectory planning for unexpected pedestrians |
| CN110389581A (en) * | 2018-04-17 | 2019-10-29 | 百度（美国）有限责任公司 | Method for the prediction locus for automatic driving vehicle dyspoiesis object |
| CN109712421A (en) * | 2019-02-22 | 2019-05-03 | 百度在线网络技术（北京）有限公司 | The speed planning method, apparatus and storage medium of automatic driving vehicle |
| CN111289008A (en) * | 2020-04-28 | 2020-06-16 | 南京维思科汽车科技有限公司 | Local path planning algorithm for unmanned vehicle |
Non-Patent Citations (2)
| Title |
|---|
| 基于中间速度的智能车辆梯形速度规划方法;段建民 等;《计算机工程》;20180815;第44卷(第8期);全文 * |
| 基于梯形规划曲线的智能车速度规划算法研究;曹波 等;《计算机科学》;20191031;第46卷(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111650945A (en) | 2020-09-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111650945B (en) | Dynamic obstacle anticollision method | |
| CN109017786B (en) | Vehicle obstacle avoidance method | |
| US11809194B2 (en) | Target abnormality determination device | |
| US11267474B2 (en) | Vehicle control device, vehicle control method, and storage medium | |
| JP6592074B2 (en) | Vehicle control device, vehicle control method, program, and information acquisition device | |
| CN107848531B (en) | Vehicle control device, vehicle control method, and medium storing vehicle control program | |
| JP7295012B2 (en) | Vehicle control system and vehicle control method | |
| US20220080961A1 (en) | Control system and control method for sampling based planning of possible trajectories for motor vehicles | |
| CN111661055B (en) | Lane changing control method and system for automatic driving vehicle | |
| US20190217861A1 (en) | Travel control device and travel control method | |
| JP6838525B2 (en) | Vehicle control device | |
| JP7193656B2 (en) | Control unit and method for recognizing intruding or exiting vehicles | |
| JP7032178B2 (en) | Vehicle control device | |
| CN114258366A (en) | Polyline profile representation for autonomous vehicles | |
| CN108983787B (en) | Road driving method | |
| JP2021020580A (en) | Vehicle control device, vehicle control method, and program | |
| JP2020147139A (en) | Vehicle control device, vehicle control method, and program | |
| JP2019153028A (en) | Vehicle controller | |
| JP7379033B2 (en) | Driving support method and driving support device | |
| JP2021041754A (en) | Operation control method and operation control apparatus | |
| CN117227714A (en) | Control method and system for turning avoidance of automatic driving vehicle | |
| CN117261938A (en) | Path planning method, path planning device, vehicle and storage medium | |
| CN114162115B (en) | Vehicle collision risk monitoring method and domain controller for intelligent driving | |
| JP6838769B2 (en) | Surrounding environment recognition device, display control device | |
| CN115848363A (en) | Collision avoidance and loss reduction trajectory planning 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 | ||
| CB02 | Change of applicant information |
Address after: B4-006, maker Plaza, 338 East Street, Huilongguan town, Changping District, Beijing 100096 Applicant after: Beijing Idriverplus Technology Co.,Ltd. Address before: B4-006, maker Plaza, 338 East Street, Huilongguan town, Changping District, Beijing 100096 Applicant before: Beijing Idriverplus Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |