CN112330959A - Optimal peer-to-peer collision avoidance method for unmanned vehicle - Google Patents

Optimal peer-to-peer collision avoidance method for unmanned vehicle Download PDF

Info

Publication number
CN112330959A
CN112330959A CN202011185044.8A CN202011185044A CN112330959A CN 112330959 A CN112330959 A CN 112330959A CN 202011185044 A CN202011185044 A CN 202011185044A CN 112330959 A CN112330959 A CN 112330959A
Authority
CN
China
Prior art keywords
vehicle
collision
automobile
vehicles
curve
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
Application number
CN202011185044.8A
Other languages
Chinese (zh)
Other versions
CN112330959B (en
Inventor
高天寒
张温玉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fantasy Technology Shanghai Co ltd
Original Assignee
Northeastern University China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northeastern University China filed Critical Northeastern University China
Priority to CN202011185044.8A priority Critical patent/CN112330959B/en
Publication of CN112330959A publication Critical patent/CN112330959A/en
Application granted granted Critical
Publication of CN112330959B publication Critical patent/CN112330959B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention provides an optimal equal collision avoidance method for an unmanned vehicle, which comprises the steps of firstly carrying out abstract processing on the vehicle, and then providing a vehicle optimal equal collision avoidance speed set formula; the vehicle follows different collision treatment principles according to the environment of the vehicle; for a plurality of vehicles passing through the curve, different collision avoidance logics are selected according to the positions of the vehicles. In addition, the method divides the lane for the vehicle at the turning position, and provides corresponding collision avoidance logic according to the position of the vehicle, so that the possible collision at the turning position is avoided as much as possible. The support is used as an effective support for the advance of the unmanned vehicle, can effectively avoid the collision possibly generated in the advance process of the unmanned vehicle, and is very suitable for the aspects of collision detection and avoidance of the unmanned vehicle.

Description

Optimal peer-to-peer collision avoidance method for unmanned vehicle
Technical Field
The invention relates to the technical field of unmanned vehicle control, in particular to an optimal peer-to-peer collision avoidance method for an unmanned vehicle.
Background
Unmanned driving has a profound influence on the automobile industry and even the transportation industry as a future research direction of automobiles. The coming of the unmanned automobile can liberate both hands of people, reduce the frequency of traffic accidents and ensure the safety of people. Future automobiles have been not limited to only one vehicle, but have been developed to a new generation of internet terminals. The unmanned automobile integrates sensing, decision making, control and feedback into one system, so that the driving maneuverability and safety of the automobile can be ensured by separating the automobile from a driver. The motion behavior of the unmanned vehicle is divided into three levels: motion selection, steering and movement, where steering behavior is described in terms of geometric calculations of vectors representing the required steering force. As one of the steering behaviors, the collision avoidance behavior gives the automobile the ability to move flexibly on a crowded road by avoiding obstacles.
Collision avoidance processing techniques for automobiles are an important part of any automobile control system.
Statistics show that automobile rear-end collisions account for the vast majority of road traffic accidents. The reason for this is that the vehicle speed is too fast, and the brain and body of a person are not in time to react, so a safe collision avoidance system is undoubtedly important for an unmanned vehicle. However, the Advanced Collision Avoidance (ACA) technology commonly used in the current autodrive vehicle is still in the primary stage, and relies on the network to perform collision recognition and avoidance, and most of the avoidance means are only simple emergency braking, which is still far from the autodrive vehicle that we imagine. The ACA technology still has certain problems at present, and the technology depends on the security and the stability of a network and has a single avoidance means. The present invention thus proposes another collision avoidance concept, without communication and information sharing between vehicles, to achieve a distributed, intended and mathematically reliable, guaranteed collision-free motion by having the unmanned vehicle follow established constraints.
Disclosure of Invention
In view of the above-described deficiencies of the prior art, the present invention provides an optimal peer-to-peer collision avoidance method for an unmanned vehicle.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: an optimal peer-to-peer collision avoidance method for an unmanned vehicle, comprising the steps of:
step 1: abstract processing a vehicle into a rectangle with a width of b and a length of a, taking p as the center of the rectangle, and representing the vehicle as R (p, a, b);
step 2: calculating an optimal peer-to-peer collision avoidance speed set among vehicles, wherein the optimal peer-to-peer collision avoidance speed set among the vehicles refers to a set of optimal allowed speeds which can be selected by the vehicles when collision avoidance is carried out, and the process is as follows:
step 2.1: calculating a set of speed obstacles for vehicle A
Figure BDA0002751190000000011
Comprises the following steps:
Figure BDA0002751190000000021
where V is the current speed of the vehicle A, t is a time variable, τ is A and the set V is selectedATime at which the inner velocity will collide with B, aAAnd aBLength and width of vehicle A, respectively, bAAnd bBThe length and width of vehicle B, respectively; if the vehicle A selects the speed in the set, the vehicle A collides with the vehicle B within the time tau;
the same principle calculates the set of speed obstacles of the vehicle B
Figure BDA0002751190000000022
The current speeds of the known vehicles A and B are v, respectivelyAAnd vBAccording to the definition of the speed obstacle we can get if
Figure BDA0002751190000000023
Or
Figure BDA0002751190000000024
A and B keep the current speed forward and will collide for at least time tau; on the contrary, if
Figure BDA0002751190000000025
The vehicles a and B can avoid collision at least for the time τ;
step 2.2: calculating a set of collision avoidance speeds of the vehicle:
use of
Figure BDA0002751190000000026
Minkowski sum representing the sets X and Y, then for any set VBIf v isB∈VBAnd is
Figure BDA0002751190000000027
A and B keep the current speed forward and will collide for at least time tau; selectable set of speeds V for any vehicle B at this timeBIn order to ensure that the vehicle A and the vehicle B do not collide at least within the time tau, the collision avoiding speed set of the vehicle A
Figure BDA0002751190000000028
Is defined as:
Figure BDA0002751190000000029
the same principle is used to obtain the collision avoidance speed set of the vehicle B as
Figure BDA00027511900000000210
Step 2.3: calculating an optimal equivalent collision avoidance speed set of the vehicle:
rational selection of the speed set V of the vehicles A and BA、VBSo that the vehicles A and B are equally maximally protected against collision, i.e.
Figure BDA00027511900000000211
And is
Figure BDA00027511900000000212
Respectively from VA、VBTo select the closest optimum speed
Figure BDA00027511900000000213
And
Figure BDA00027511900000000214
speed pairs, sets
Figure BDA00027511900000000215
And
Figure BDA00027511900000000216
sets of optimal equal-to-equal collision avoidance velocities for vehicles A and B, respectively, that contain more nearly optimal velocities
Figure BDA0002751190000000031
And
Figure BDA0002751190000000032
while avoiding speed pairs V for all other peer-to-peer collisionsAAnd VBComprises the following steps:
Figure BDA0002751190000000033
and step 3: set a limited field of view for the vehicle: setting a collision time range, wherein the vehicle only considers the collision within the time range from the current moment and ignores any collision which occurs after the time range from the current moment;
the limited view is set for the vehicle, namely the vehicle only has one limited view and is not allowed to react to all other vehicles outside the view, and the limited view is called the view.
And 4, step 4: setting a principle of avoiding collision of different vehicles under different scenes;
the different scenes comprise two scenes, namely a straight road scene and a curve scene;
the collision avoidance principle of the straight line road scene is as follows:
when the automobile detects that the automobile is about to collide with the relatively-running automobile in front, the automobile automation system controls the automobiles to be away from each other, so that the responsibility of avoiding collision among the automobiles is equally assumed;
if the vehicle detects that the vehicle is about to collide with the tail of the automobile running in parallel in front, the automobile automation system controls the automobile to avoid the collision in an effort no matter whether the front vehicle avoids the collision or not;
therefore, on a straight road, the vehicles in front can only avoid the guardrails on the two sides and the fault vehicles on the road, and pay attention to whether the potential collision between the vehicles behind needs to be avoided or not at intervals, and the speed reduction or the direction change of the vehicles behind cannot be abandoned in order to avoid the collision of the vehicles behind.
The collision avoidance principle of the curve scene is as follows:
if the vehicle keeps the original state and continues to advance, when the collision is about to occur within the detected collision time range:
the vehicle positioned at the forefront only passes through a curve as soon as possible, the vehicle condition at the rear is not concerned, and the automobile automation system cannot control the automobile to avoid;
when a vehicle at the rear side detects that collision is about to occur, in order to safely pass through a curve, the automobile automation system controls the automobile to avoid potential collision of a front vehicle;
when the vehicle on the inner side of the curve is ahead of the vehicle on the outer side, the vehicle only takes charge of avoiding the collision possibly occurring with the inner side of the curve, and neglects the influence of other vehicles on the vehicle, and the automobile automatic system only detects the distance between the vehicle and the curve on the left side and does not pay attention to the rear vehicle condition; the vehicle automatic system is responsible for avoiding the outside vehicle at the same time only when the vehicle lags behind the outside vehicle, and the vehicle automatic system detects the distance between the vehicle and the outside vehicle and controls the vehicle and the outside vehicle to keep a certain distance so as to avoid possible collision;
the vehicle positioned at the outermost side of the curve adds the curve at the outer side into a collision range needing to be avoided, and only takes charge of avoiding the possible collision with the outer side of the curve when the vehicle is ahead of the vehicle at the inner side, and neglects the influence of other vehicles on the vehicle, so that the automobile automatic system only detects the distance between the vehicle and the outer side of the curve without paying attention to the rear vehicle condition; only when the automobile lags behind the inner side vehicle, the automobile can be simultaneously responsible for avoiding the inner side vehicle, and at the moment, the automatic system can simultaneously detect the distance between the automobile and the inner side vehicle, control the automobile and the inner side vehicle to keep a certain distance, and avoid the possible collision as much as possible.
The method for detecting the impending collision of the vehicle comprises the following steps:
the video camera, the radar sensor and the laser range finder which are arranged in the unmanned vehicle are used for knowing the surrounding traffic condition, so that the two vehicles are determined to be about to collide when the distance between the two vehicles is less than or equal to a threshold value.
And 5: and (4) sensing the collision in real time and avoiding the collision according to the collision avoiding principle in the step 4.
Adopt the produced beneficial effect of above-mentioned technical scheme to lie in: the method provided by the invention analyzes the set principle of the unmanned vehicle, adjusts the collision processing principle followed by a plurality of vehicles meeting on a straight road, and the vehicles can make corresponding reactions according to the surrounding road conditions. In addition, the method divides the lane for the vehicle at the turning position, and provides corresponding collision avoidance logic according to the position of the vehicle, so that the possible collision at the turning position is avoided as much as possible. The support is used as an effective support for the advance of the unmanned vehicle, can effectively avoid the collision possibly occurring in the advance process of the unmanned vehicle, is beneficial to the research and development of the unmanned vehicle, provides reference for the design of a collision avoidance system in the future, and can be applied to more research fields.
Drawings
FIG. 1 is a flow chart of an optimal peer-to-peer collision avoidance method for an unmanned vehicle in an embodiment of the present invention;
FIG. 2 is a flowchart of a method for calculating a set of optimal allowable speeds to be selected by a vehicle in avoiding a collision according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a vehicle turning position distribution according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Considering that most roads can tolerate at most three vehicles to go ahead in parallel, fig. 3 shows a position distribution diagram of at most three vehicles passing through a turn of 90 ° to the left at the same time.
As shown in fig. 1, an optimal peer-to-peer collision avoidance method for an unmanned vehicle in the present embodiment includes the steps of:
step 1: abstract processing a vehicle into a rectangle with a width of b and a length of a, taking p as the center of the rectangle, and representing the vehicle as R (p, a, b);
step 2: calculating an optimal peer-to-peer collision avoidance speed set among vehicles, wherein the optimal peer-to-peer collision avoidance speed set among the vehicles refers to a set of optimal allowed speeds which can be selected by the vehicles when collision avoidance is carried out, and the process is as follows:
step 2.1: calculating a set of speed obstacles for vehicle A
Figure BDA0002751190000000041
Comprises the following steps:
Figure BDA0002751190000000042
where V is the current speed of the vehicle A, t is a time variable, τ is A and the set V is selectedATime at which the inner velocity will collide with B, aAAnd aBLength and width of vehicle A, respectively, bAAnd bBThe length and width of vehicle B, respectively; if the vehicle A selects the speed in the set, the vehicle A collides with the vehicle B within the time tau;
the same principle calculates the set of speed obstacles of the vehicle B
Figure BDA0002751190000000051
The current speeds of the known vehicles A and B are v, respectivelyAAnd vBAccording to the definition of the speed obstacle we can get if
Figure BDA0002751190000000052
Or
Figure BDA0002751190000000053
A and B maintain the current speedAdvance, will collide for at least time τ; on the contrary, if
Figure BDA0002751190000000054
The vehicles a and B can avoid collision at least for the time τ;
step 2.2: calculating a set of collision avoidance speeds of the vehicle:
use of
Figure BDA0002751190000000055
Minkowski sum representing the sets X and Y, then for any set VBIf v isB∈VBAnd is
Figure BDA0002751190000000056
A and B keep the current speed forward and will collide for at least time tau; selectable set of speeds V for any vehicle B at this timeBIn order to ensure that the vehicle A and the vehicle B do not collide at least within the time tau, the collision avoiding speed set of the vehicle A
Figure BDA0002751190000000057
Is defined as:
Figure BDA0002751190000000058
the same principle is used to obtain the collision avoidance speed set of the vehicle B as
Figure BDA0002751190000000059
Step 2.3: calculating an optimal equivalent collision avoidance speed set of the vehicle:
rational selection of the speed set V of the vehicles A and BA、VBSo that the vehicles A and B are equally maximally protected against collision, i.e.
Figure BDA00027511900000000510
And is
Figure BDA00027511900000000511
Respectively from VA、VBTo select the closest optimum speed
Figure BDA00027511900000000512
And
Figure BDA00027511900000000513
speed pairs, sets
Figure BDA00027511900000000514
And
Figure BDA00027511900000000515
sets of optimal equal-to-equal collision avoidance velocities for vehicles A and B, respectively, that contain more nearly optimal velocities
Figure BDA00027511900000000516
And
Figure BDA00027511900000000517
while avoiding speed pairs V for all other peer-to-peer collisionsAAnd VBComprises the following steps:
Figure BDA0002751190000000061
in this example, the flow of step 1 and step 2 is shown in fig. 2.
And step 3: set a limited field of view for the vehicle: setting a collision time range, wherein the vehicle only considers the collision within the time range from the current moment and ignores any collision which occurs after the time range from the current moment;
the limited view is set for the vehicle, namely the vehicle only has one limited view and is not allowed to react to all other vehicles outside the view, and the limited view is called the view. A smaller time range threshold value can ensure that the vehicle keeps the original motion state of the vehicle as much as possible when avoiding collision and does not deviate from the established advancing route greatly.
And 4, step 4: setting a principle of avoiding collision of different vehicles under different scenes;
in this embodiment, the principle of avoiding collision in a curve situation is taken as an example for explanation, and the turning position distribution of the vehicle is schematically shown in fig. 3: to make the fastest pass through a curve, when only one vehicle passes, it will continue to advance from 1. When two or more automobiles pass through the curve, in order to create the chance of overtaking, the principle that one automobile passes through one lane is implemented before the curve is passed through. The position distribution when two vehicles pass through is as follows: car 1 is at 1, 4 and car 2 is at 2, 5. The possible locations of car 1 when three cars pass are distributed as 1, 4, 7, car 2 may be at 2, 5, 7, and car 3 may be at 3, 6, 9.
Scenario one: when only one vehicle is on the curve;
the car 1 is located at 1 and the car 1 is only responsible for evading possible collisions with the left-hand bend.
Scenario two: when two vehicles are on the curve;
car 1 is located at 1 and car 2 is located at 2. The automobile 1 is only responsible for avoiding possible collision with the left curve, and the automobile 2 is completely responsible for avoiding possible collision with the automobile 1;
car 1 is located at 1 and car 2 is located at 5. The automobile 1 is only responsible for avoiding possible collision with the left curve, and the automobile 2 is completely responsible for avoiding possible collision with the automobile 1;
car 1 is located at 4 and car 2 is located at 2. The automobile 2 does not collide within the collision time tau, the original state is kept, and the automobile 1 is in charge of possible collision between the left curve and the automobile 2;
car 1 is located at 4 and car 2 is located at 5. The automobile 1 is only responsible for avoiding possible collision with the left-side curve, and the automobile 2 is completely responsible for avoiding possible collision with the automobile 1.
Scenario three: when there are three vehicles on the curve;
car 1 is located at 1, car 2 is located at 2, and car 3 is located at 3. The automobile 1 is only responsible for avoiding possible collision with the left curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right curve;
car 1 is located at 1, car 2 is located at 2, and car 3 is located at 6. The automobile 1 is only responsible for avoiding possible collision with the left side curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and when the automobile 3 advances to the position 6, if the automobile is detected to be still within the time range of collision with the automobile 2, the automobile 2 and the right side curve start to be responsible for possible collision;
car 1 is located at 1, car 2 is located at 2, and car 3 is located at 9. The automobile 1 is only responsible for avoiding possible collision with the left side curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and when the automobile 3 advances to the position 6, if the automobile is detected to be still within the time range of collision with the automobile 2, the automobile 2 and the right side curve start to be responsible for possible collision;
car 1 is located at 1, car 2 is located at 5, and car 3 is located at 3. The automobile 1 is only responsible for avoiding the possible collision with the left curve, the automobile 3 is responsible for avoiding the possible collision with the right curve, and the automobile 2 is also responsible for avoiding the possible collision with the automobiles 1 and 3;
car 1 is located at 1, car 2 is located at 5, and car 3 is located at 6. The automobile 1 is only responsible for avoiding possible collision with the left curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right curve;
car 1 is located at 1, car 2 is located at 5, and car 3 is located at 9. The automobile 1 is only responsible for avoiding possible collision with the left curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right curve;
car 1 is located at 1, car 2 is located at 8, and car 3 is located at 3. The automobile 1 is only responsible for avoiding possible collision with the left side curve, the automobile 3 is responsible for possible collision with the right side curve, and the automobile 2 starts to be responsible for avoiding possible collision with the automobiles 1 and 3 when detecting that the automobile is still within the collision time range with the automobiles 1 and 3 when advancing to the position 5;
car 1 is located at 1, car 2 is located at 8, and car 3 is located at 6. The automobile 1 is only responsible for avoiding the possible collision with the left curve, the automobile 3 is responsible for avoiding the possible collision with the right curve, and the automobile 2 is also responsible for avoiding the possible collision with the automobiles 1 and 3;
car 1 is located at 1, car 2 is located at 8, and car 3 is located at 9. The automobile 1 is only responsible for avoiding the possible collision with the left side curve, the automobile 2 and the automobile 3 continue to advance, if the automobile 2 reaches the position 5 first, the automobile 2 is completely responsible for the possible collision with the automobile 1, and the automobile 3 is responsible for the possible collision with the automobile 2 and the right side curve; on the contrary, if the automobile 3 arrives at the position 6 first, the automobile 3 is responsible for avoiding the collision possibly occurring between the automobile 1 and the automobile 3 while the automobile 3 is responsible for avoiding the collision possibly occurring between the automobile 3 and the right-side curve;
car 1 is located at 4, car 2 is located at 2, and car 3 is located at 3. The automobile 2 cannot collide within the collision time tau and keeps the original state to move forward, the automobile 1 is in charge of avoiding the possible collision between the automobile 2 and the left side curve, and the automobile 3 is in charge of avoiding the possible collision between the automobile 2 and the right side curve;
car 1 is located at 4, car 2 is located at 2, and car 3 is located at 6. The automobile 2 cannot collide within the collision time tau and keeps the original state to move forward, the automobile 1 is in charge of avoiding the possible collision between the automobile 2 and the left side curve, and the automobile 3 is in charge of avoiding the possible collision between the automobile 2 and the right side curve;
car 1 is located at 4, car 2 is located at 2, and car 3 is located at 9. The automobile 2 cannot collide within the collision time tau and keeps moving forward in the original state, the automobile 1 is responsible for avoiding possible collision between the automobile 2 and the left side curve, and the automobile 3 starts to be responsible for avoiding possible collision between the automobile 2 and the right side curve if the automobile 2 is still within the collision time range when the automobile 3 moves forward to the position 6;
car 1 is located at 4, car 2 is located at 5, and car 3 is located at 3. The automobile 1 is only responsible for avoiding the possible collision with the left curve, the automobile 3 is responsible for avoiding the possible collision with the right curve, and the automobile 2 is also responsible for avoiding the possible collision with the automobiles 1 and 3;
car 1 is located at 4, car 2 is located at 5, and car 3 is located at 6. The automobile 1 is only responsible for avoiding possible collision with the left curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right curve;
car 1 is located at 4, car 2 is located at 5, and car 3 is located at 9. The automobile 1 is only responsible for avoiding possible collision with the left curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right curve;
car 1 is located at 4, car 2 is located at 8, and car 3 is located at 3. The automobile 1 is only responsible for avoiding possible collision with the left side curve, the automobile 3 is only responsible for possible collision with the right side curve, and the automobile 2 is simultaneously responsible for avoiding possible collision with the automobile 1 and the automobile 3 when the automobile 3 is still in the collision time range, otherwise, the automobile 1 is only responsible for avoiding possible collision with the automobile 1;
car 1 is located at 4, car 2 is located at 8, and car 3 is located at 6. The automobile 1 is only responsible for avoiding the possible collision with the left curve, the automobile 3 is responsible for avoiding the possible collision with the right curve, and the automobile 2 is also responsible for avoiding the possible collision with the automobiles 1 and 3;
car 1 is located at 4, car 2 is located at 8, and car 3 is located at 9. The automobile 1 is only responsible for avoiding possible collision with the left curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right curve;
car 1 is located at 7, car 2 is located at 2, and car 3 is located at 3. The automobile 2 cannot collide within the collision time tau and keeps the original state to move forward, and the automobile 3 is responsible for avoiding possible collision between the right curve and the automobile 2; when the automobile 1 moves to the position 4, if the automobile 2 still is in the collision time range, the automobile starts to take charge of avoiding the possible collision between the left curve and the automobile 2;
car 1 is located at 7, car 2 is located at 2, and car 3 is located at 6. The automobile 2 cannot collide within the collision time tau and keeps the original state to move forward, and the automobile 3 is responsible for avoiding possible collision between the right curve and the automobile 2; when the automobile 1 moves to the position 4, if the automobile 2 still is in the collision time range, the automobile starts to take charge of avoiding the possible collision between the left curve and the automobile 2;
car 1 is located at 7, car 2 is located at 2, and car 3 is located at 9. The automobile 2 does not collide within the collision time tau and keeps the original state to move forward; when the automobile 1 moves forward to the position 4, if the automobile 2 still is in the collision time range, the automobile starts to take charge of avoiding the possible collision between the left curve and the automobile 2; when the automobile 3 moves to the position 6, if the automobile 2 still stays in the collision time range, the automobile starts to take charge of avoiding the possible collision between the automobile 2 and the right curve;
car 1 is located at 7, car 2 is located at 5, and car 3 is located at 3. The automobile 3 is responsible for avoiding the collision possibly occurring between the automobile and the right curve; the automobile 2 is responsible for avoiding the collision possibly occurring with the automobile 3; the automobile 1 is in charge of avoiding the collision which can happen between the automobile 2 and the left curve;
car 1 is located at 7, car 2 is located at 5, and car 3 is located at 6. The automobile 2 cannot collide within the collision time tau and keeps the original state to move forward, the automobile 1 is in charge of avoiding the possible collision between the automobile 2 and the left side curve, and the automobile 3 is in charge of avoiding the possible collision between the automobile 2 and the right side curve;
car 1 is located at 7, car 2 is located at 5, and car 3 is located at 9. The automobile 2 cannot collide within the collision time tau and keeps the original state to move forward, the automobile 1 is in charge of avoiding the possible collision between the automobile 2 and the left side curve, and the automobile 3 is in charge of avoiding the possible collision between the automobile 2 and the right side curve;
car 1 is located at 7, car 2 is located at 8, and car 3 is located at 3. The automobile 1 is only responsible for avoiding possible collision with the left side curve, the automobile 3 is only responsible for possible collision with the right side curve, and the automobile 2 is simultaneously responsible for avoiding possible collision with the automobile 1 and the automobile 3 when the automobile 3 is still in the collision time range, otherwise, the automobile 1 is only responsible for avoiding possible collision with the automobile 1;
car 1 is located at 7, car 2 is located at 8, and car 3 is located at 6. The automobile 1 is only responsible for avoiding the possible collision with the left curve, the automobile 3 is responsible for avoiding the possible collision with the right curve, and the automobile 2 is also responsible for avoiding the possible collision with the automobiles 1 and 3;
car 1 is located at 7, car 2 is located at 8, and car 3 is located at 9. The automobile 1 is only responsible for avoiding possible collision with the left side curve, the automobile 2 is completely responsible for possible collision with the automobile 1, and the automobile 3 is responsible for possible collision with the automobile 2 and the right side curve.
And 5: and (4) sensing collision in real time by the vehicle and avoiding collision according to the collision avoidance logic in the step 4.
Each vehicle senses whether collision exists in the visual field in real time, analyzes the environment of the vehicle when the collision possibly occurs is detected, calculates the optimal equivalent collision prevention speed set by adopting different collision prevention logics, and selects the optimal speed in the set to perform real-time collision prevention.
The method for detecting the impending collision of the vehicle comprises the following steps: the video camera, the radar sensor and the laser range finder which are arranged in the unmanned vehicle are used for knowing the surrounding traffic condition, so that the two vehicles are determined to be about to collide when the distance between the two vehicles is less than or equal to a threshold value.

Claims (6)

1. An optimal peer-to-peer collision avoidance method for an unmanned vehicle, comprising the steps of:
step 1: abstract processing a vehicle into a rectangle with a width of b and a length of a, taking p as the center of the rectangle, and representing the vehicle as R (p, a, b);
step 2: calculating an optimal set of collision avoidance speeds between vehicles;
and step 3: set a limited field of view for the vehicle: setting a collision time range, wherein the vehicle only considers the collision within the time range from the current moment and ignores any collision which occurs after the time range from the current moment;
and 4, step 4: setting a principle of avoiding collision of different vehicles under different scenes;
and 5: and (4) sensing the collision in real time and avoiding the collision according to the collision avoiding principle in the step 4.
2. An optimal peer-to-peer collision avoidance method for unmanned vehicles according to claim 1, characterized in that: the process of the step 2 is as follows:
step 2.1: calculating a set of speed obstacles for vehicle A
Figure FDA0002751189990000011
Comprises the following steps:
Figure FDA0002751189990000012
where V is the current speed of the vehicle A, t is a time variable, τ is A and the set V is selectedATime at which the inner velocity will collide with B, aAAnd aBLength and width of vehicle A, respectively, bAAnd bBThe length and width of vehicle B, respectively; if the vehicle A selects the speed in the set, the vehicle A collides with the vehicle B within the time tau;
the same principle calculates the set of speed obstacles of the vehicle B
Figure FDA0002751189990000013
The current speeds of the known vehicles A and B are v, respectivelyAAnd vBAccording to the definition of the speed obstacle we can get if
Figure FDA0002751189990000014
Or
Figure FDA0002751189990000015
A and B keep the current speed forward and will collide for at least time tau; on the contrary, if
Figure FDA0002751189990000016
The vehicles a and B can avoid collision at least for the time τ;
step 2.2: calculating a set of collision avoidance speeds of the vehicle:
use of
Figure FDA0002751189990000017
Minkowski sum representing the sets X and Y, then for any set VBIf v isB∈VBAnd is
Figure FDA0002751189990000018
A and B keep the current speed forward and will collide for at least time tau; selectable set of speeds V for any vehicle B at this timeBIn order to ensure that the vehicle A and the vehicle B do not collide at least within the time tau, the collision avoiding speed set of the vehicle A
Figure FDA0002751189990000021
Is defined as:
Figure FDA0002751189990000022
the same principle is used to obtain the collision avoidance speed set of the vehicle B as
Figure FDA0002751189990000023
Step 2.3: calculating an optimal equivalent collision avoidance speed set of the vehicle:
rational selection of the speed set V of the vehicles A and BA、VBSo that the vehicles A and B are equally maximally protected against collision, i.e.
Figure FDA0002751189990000024
And is
Figure FDA0002751189990000025
Respectively from VA、VBTo select the closest optimum speed
Figure FDA0002751189990000026
And
Figure FDA0002751189990000027
speed pairs, sets
Figure FDA0002751189990000028
And
Figure FDA0002751189990000029
sets of optimal equal-to-equal collision avoidance velocities for vehicles A and B, respectively, that contain more nearly optimal velocities
Figure FDA00027511899900000210
And
Figure FDA00027511899900000211
while avoiding speed pairs V for all other peer-to-peer collisionsAAnd VBComprises the following steps:
Figure FDA00027511899900000212
3. an optimal peer-to-peer collision avoidance method for unmanned vehicles according to claim 1, characterized in that: the different scenes of the step 4 comprise two scenes, namely a straight road scene and a curve scene.
4. An optimal peer-to-peer collision avoidance method for unmanned vehicles according to claim 3, characterized in that: the collision avoidance principle of the straight line road scene is as follows:
when the automobile detects that the automobile is about to collide with the relatively-running automobile in front, the automobile automation system controls the automobiles to be away from each other, so that the responsibility of avoiding collision among the automobiles is equally assumed;
if the vehicle detects that the vehicle is about to collide with the tail of the automobile running in parallel in front, the automobile automation system controls the automobile to avoid the collision in an effort no matter whether the front vehicle avoids the collision or not;
therefore, on a straight road, the vehicles in front can only avoid the guardrails on the two sides and the fault vehicles on the road, and pay attention to whether the potential collision between the vehicles behind needs to be avoided or not at intervals, and the speed reduction or the direction change of the vehicles behind cannot be abandoned in order to avoid the collision of the vehicles behind.
5. An optimal peer-to-peer collision avoidance method for unmanned vehicles according to claim 3, characterized in that: the collision avoidance principle of the curve scene is as follows:
if the vehicle keeps the original state and continues to advance, when the collision is about to occur within the detected collision time range:
the vehicle positioned at the forefront only passes through a curve as soon as possible, the vehicle condition at the rear is not concerned, and the automobile automation system cannot control the automobile to avoid;
when a vehicle at the rear side detects that collision is about to occur, in order to safely pass through a curve, the automobile automation system controls the automobile to avoid potential collision of a front vehicle;
when the vehicle on the inner side of the curve is ahead of the vehicle on the outer side, the vehicle only takes charge of avoiding the collision possibly occurring with the inner side of the curve, and neglects the influence of other vehicles on the vehicle, and the automobile automatic system only detects the distance between the vehicle and the curve on the left side and does not pay attention to the rear vehicle condition; the vehicle automatic system is responsible for avoiding the outside vehicle at the same time only when the vehicle lags behind the outside vehicle, and the vehicle automatic system detects the distance between the vehicle and the outside vehicle and controls the vehicle and the outside vehicle to keep a certain distance so as to avoid possible collision;
the vehicle positioned at the outermost side of the curve adds the curve at the outer side into a collision range needing to be avoided, and only takes charge of avoiding the possible collision with the outer side of the curve when the vehicle is ahead of the vehicle at the inner side, and neglects the influence of other vehicles on the vehicle, so that the automobile automatic system only detects the distance between the vehicle and the outer side of the curve without paying attention to the rear vehicle condition; only when the automobile lags behind the inner side vehicle, the automobile can be simultaneously responsible for avoiding the inner side vehicle, and at the moment, the automatic system can simultaneously detect the distance between the automobile and the inner side vehicle, control the automobile and the inner side vehicle to keep a certain distance, and avoid the possible collision as much as possible.
6. An optimal peer-to-peer collision avoidance method for unmanned vehicles according to claim 4 or 5, wherein the method of detecting an impending collision of a vehicle is:
the video camera, the radar sensor and the laser range finder which are arranged in the unmanned vehicle are used for knowing the surrounding traffic condition, so that the two vehicles are determined to be about to collide when the distance between the two vehicles is less than or equal to a threshold value.
CN202011185044.8A 2020-10-30 2020-10-30 Optimal peer-to-peer collision avoidance method for unmanned vehicle Active CN112330959B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011185044.8A CN112330959B (en) 2020-10-30 2020-10-30 Optimal peer-to-peer collision avoidance method for unmanned vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011185044.8A CN112330959B (en) 2020-10-30 2020-10-30 Optimal peer-to-peer collision avoidance method for unmanned vehicle

Publications (2)

Publication Number Publication Date
CN112330959A true CN112330959A (en) 2021-02-05
CN112330959B CN112330959B (en) 2022-02-22

Family

ID=74297337

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011185044.8A Active CN112330959B (en) 2020-10-30 2020-10-30 Optimal peer-to-peer collision avoidance method for unmanned vehicle

Country Status (1)

Country Link
CN (1) CN112330959B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034928A (en) * 2021-02-26 2021-06-25 山东交通学院 Accompanying system of road maintenance safety warning robot

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103722565A (en) * 2014-01-23 2014-04-16 哈尔滨工业大学 Self-collision monitoring system for humanoid robot and monitoring method thereof
CN108664020A (en) * 2018-04-11 2018-10-16 上海大学 A kind of unmanned boat dynamic obstacle avoidance algorithm based on Speed Obstacles method and dynamic window method
CN109416886A (en) * 2016-06-22 2019-03-01 库卡德国有限公司 Check method, Unpiloted carrier vehicle and the system with multiple Unpiloted carrier vehicles of the collision between two Unpiloted carrier vehicles
WO2019060679A1 (en) * 2017-09-22 2019-03-28 Locus Robotics Corp. Dynamic window approach using optimal reciprocal collision avoidance cost-critic
WO2019062630A1 (en) * 2017-09-30 2019-04-04 蔚来汽车有限公司 Forward collision avoidance method and system for vehicle
CN111045422A (en) * 2018-10-11 2020-04-21 顾泽苍 Control method for automatically driving and importing 'machine intelligence acquisition' model

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103722565A (en) * 2014-01-23 2014-04-16 哈尔滨工业大学 Self-collision monitoring system for humanoid robot and monitoring method thereof
CN109416886A (en) * 2016-06-22 2019-03-01 库卡德国有限公司 Check method, Unpiloted carrier vehicle and the system with multiple Unpiloted carrier vehicles of the collision between two Unpiloted carrier vehicles
WO2019060679A1 (en) * 2017-09-22 2019-03-28 Locus Robotics Corp. Dynamic window approach using optimal reciprocal collision avoidance cost-critic
WO2019062630A1 (en) * 2017-09-30 2019-04-04 蔚来汽车有限公司 Forward collision avoidance method and system for vehicle
CN108664020A (en) * 2018-04-11 2018-10-16 上海大学 A kind of unmanned boat dynamic obstacle avoidance algorithm based on Speed Obstacles method and dynamic window method
CN111045422A (en) * 2018-10-11 2020-04-21 顾泽苍 Control method for automatically driving and importing 'machine intelligence acquisition' model

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
GAO, TH: "The Virtual Museum Based on HoloLens and Vuforia", 《2018 4TH ANNUAL INTERNATIONAL CONFERENCE ON NETWORK AND INFORMATION SYSTEMS FOR COMPUTERS (ICNISC 2018)》 *
刘中常等: "基于预测窗的轮式移动机器人最优避障避碰算法", 《控制理论与应用》 *
徐鹏等: "改进互动速度障碍的多机器人协同避障", 《自动化与仪表》 *
高天寒: "基于MAP框架的移动IPv6关键技术研究", 《中国博士学位论文全文数据库信息科技辑》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034928A (en) * 2021-02-26 2021-06-25 山东交通学院 Accompanying system of road maintenance safety warning robot

Also Published As

Publication number Publication date
CN112330959B (en) 2022-02-22

Similar Documents

Publication Publication Date Title
EP2902290B1 (en) System for accommodating a pedestrian during autonomous vehicle operation
CN107792064B (en) Vehicle control device
EP2743901B1 (en) Driving assistance device
US8606479B2 (en) Vehicle drive assist system
US7191049B2 (en) Vehicle drive assist system
CN107709122B (en) Prospective control system for motor vehicle
JP5949961B2 (en) Driving assistance device
US7711466B2 (en) Vehicle driving control unit
CN112896157A (en) Defensive driving control method, device and system, vehicle-mounted terminal and storage medium
US20090037052A1 (en) Vehicle drive assist system
CN112193246B (en) Vehicle and method for performing inter-vehicle distance control
CN109843679B (en) Vehicle control device
JP2010018062A (en) Vehicle driving support device
CN110758392A (en) Autonomous lane change control method for passenger car
CN112477856B (en) Arbitration method and system for emergency steering
JP2007269312A (en) Driving operation auxiliary device for vehicle
CN112224202B (en) Multi-vehicle cooperative collision avoidance system and method under emergency working condition
JP2017136968A (en) Vehicle control device
CN112330959B (en) Optimal peer-to-peer collision avoidance method for unmanned vehicle
CN108064207B (en) Vehicle control device
CN108407807B (en) Steering collision avoidance system of commercial vehicle and control method thereof
CN113619575A (en) Emergency steering system based on millimeter wave radar
CN112068559A (en) Method and system for controlling deviation of unmanned vehicle, vehicle and storage medium
JP2019191882A (en) Platoon travel controller
CN110758391A (en) Control method of self-adaptive cruise ACC system for passenger car

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20230323

Address after: Room I, 30/F, Feizhou International, No. 899, Lingling Road, Xuhui District, Shanghai, 200030

Patentee after: Fantasy Technology (Shanghai) Co.,Ltd.

Address before: 110819 No. 3 lane, Heping Road, Heping District, Shenyang, Liaoning 11

Patentee before: Northeastern University