CN108674413B - Vehicle and pedestrian collision prevention method and system - Google Patents
Vehicle and pedestrian collision prevention method and system Download PDFInfo
- Publication number
- CN108674413B CN108674413B CN201810483121.4A CN201810483121A CN108674413B CN 108674413 B CN108674413 B CN 108674413B CN 201810483121 A CN201810483121 A CN 201810483121A CN 108674413 B CN108674413 B CN 108674413B
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- vehicle
- real
- collision
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
-
- 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
-
- 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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/08—Interaction between the driver and the control system
- B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
-
- 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
-
- 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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
- B60W50/08—Interaction between the driver and the control system
- B60W50/14—Means for informing the driver, warning the driver or prompting a driver intervention
- B60W2050/143—Alarm means
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
-
- 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
- B60W2554/00—Input parameters relating to objects
- B60W2554/80—Spatial relation or speed relative to objects
Abstract
The invention discloses a vehicle and pedestrian collision prevention method and system, which can predict and obtain the possible collision between a vehicle and a pedestrian and carry out collision early warning according to the coincidence condition of a first predicted motion area of the pedestrian and a second predicted position of the vehicle after a set time interval after the first predicted position of the pedestrian and the second predicted position of the vehicle after the set time interval are predicted by predicting the first predicted position of the pedestrian after the set time interval and correspondingly drawing a first predicted motion area of the pedestrian after the set time interval according to the real-time motion speed of the pedestrian. The invention can accurately predict and obtain the positions of the pedestrians and the vehicles by dynamically predicting the specific positions of the pedestrians and the vehicles after a period of time in the future, has higher prediction precision, and can effectively predict the possibility of collision between the vehicles and the pedestrians, thereby effectively preventing the collision between the pedestrians and the vehicles, achieving the aim of reducing traffic accidents, and being widely applied to the automobile intelligent technology.
Description
Technical Field
The invention relates to the field of driving assistance, in particular to a method and a system for preventing vehicle and pedestrian collision.
Background
With the great increase of the number of automobiles, safe driving technology of automobiles is also more and more important. The technology can remind a driver of paying attention to pedestrians and reduce the occurrence of collision accidents, so that the accuracy of the technology plays a vital role in safe driving of the automobile.
In the prior art, generally, a specific position of a pedestrian after a period of time is predicted, and then whether collision occurs is predicted by judging whether an intersection point exists between the position and a preset running track of a vehicle, so as to perform warning and reminding. Because the movement of the pedestrian has uncertainty, the accurate prediction algorithm still has error with the actual position, so the current technology has the problem of inaccurate prediction, and the condition that the pedestrian and the vehicle collide cannot be effectively prevented.
The noun explains:
UTM coordinate system: UTM is known as UNIVERSAL transform business or GRID SYSTEM, and chinese is known as UNIVERSAL TRANSVERSE ink card grid system, UTM coordinate is a planar rectangular coordinate, and this grid system and its projection are widely used in topographic map, as reference grid for satellite image and natural resource database, and for applications requiring precise positioning.
Disclosure of Invention
In order to solve the above technical problems, an object of the present invention is to provide a vehicle and pedestrian collision prevention method and system.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a vehicle pedestrian collision prevention method comprising the steps of:
according to the first real-time motion data of the pedestrian, dynamically predicting a first predicted position of the pedestrian after a set time interval by adopting a pedestrian prediction model;
according to the second real-time motion data of the vehicle, dynamically predicting a second predicted position of the vehicle after a set time interval by using a vehicle prediction model;
according to the first real-time moving speed of the pedestrian, a first prediction moving area of the pedestrian after a set time interval is described by combining the first prediction position;
and predicting and acquiring the possible collision between the vehicle and the pedestrian according to the coincidence condition of the first predicted motion area of the pedestrian and the second predicted position of the vehicle, and performing collision early warning.
Further, the method also comprises the following steps:
the first real-time motion data of the pedestrian and the second real-time motion data of the vehicle are synchronized based on the time stamp.
Further, the first real-time movement data and the first predicted position both include the position, speed and movement direction of the pedestrian, and the second real-time movement data and the second predicted position both include the position, speed and movement direction of the vehicle.
Further, the pedestrian prediction model and the vehicle prediction model are obtained by adopting multilayer perception neural network or nonlinear regression neural network dynamic training, and the hidden layer of the neural network is calculated by adopting a Bayesian regular algorithm or a back propagation algorithm.
Further, in the dynamic prediction process of the pedestrian prediction model and the vehicle prediction model, the corresponding first real-time motion data and the corresponding second real-time motion data are used as training data, and dynamic prediction is carried out in the following way:
carrying out data cleaning on the training data, and filtering noise data;
sampling the cleaned training data according to the set sampling frequency and the number of sampling points to obtain input data of the prediction model corresponding to different moments;
and inputting the input data at different moments into a multilayer perception neural network or a nonlinear regression neural network in real time for training, and dynamically predicting to obtain the real-time predicted position of the pedestrian or the vehicle after a set time interval.
Further, the dynamic prediction process is performed by performing prediction calculation in a UTM coordinate system, and before the steps of performing data cleaning on the training data and filtering out noise data, the dynamic prediction process further includes the following steps:
and converting the training data from the GPS coordinate system into the data of the UTM coordinate system.
Further, the first prediction motion area and the first real-time motion speed are in a positive correlation relationship.
Further, the step of predicting to obtain the collision which may occur between the vehicle and the pedestrian and performing collision early warning specifically comprises:
calculating and obtaining the distance D between the vehicle and the pedestrian according to the first predicted position and the second predicted position;
calculating to obtain the braking distance Ds of the vehicle according to the second real-time movement speed of the vehicle;
the method comprises the following steps of predicting and obtaining the possible collision of a vehicle and a pedestrian, and carrying out corresponding collision early warning:
judging that the vehicle and the pedestrian do not collide when D-Ds-L is larger than Dw under the first condition;
judging the possibility of collision between the vehicle and the pedestrian when Dw > D-Ds-L > Dd, and warning the vehicle to pay attention to the pedestrian;
judging that the vehicle and the pedestrian have collision danger when the D-Ds-L is less than the Dd, and warning the vehicle to take defensive measures and/or actively taking defensive measures;
wherein, L represents the distance of the vehicle body, Dw represents the radius of the warning area corresponding to the pedestrian, Dd represents the radius of the danger area corresponding to the pedestrian, and Dw and Dd are both the drawing parameters of the first predicted motion area corresponding to the pedestrian.
The other technical scheme adopted by the invention for solving the technical problem is as follows:
a vehicle pedestrian collision prevention system comprising:
at least one processor;
at least one memory for storing at least one program;
when the at least one program is executed by the at least one processor, the at least one processor causes the at least one processor to implement the vehicle pedestrian collision prevention method.
The invention has the beneficial effects that: according to the scheme, after the first predicted position of the pedestrian after the set time interval and the second predicted position of the vehicle after the set time interval are predicted, the real-time movement speed of the pedestrian is combined, the first predicted movement area of the pedestrian after the set time interval is correspondingly drawn, the collision which is possibly generated by the vehicle and the pedestrian can be predicted and obtained according to the superposition condition of the first predicted movement area of the pedestrian and the second predicted position of the vehicle, and the collision early warning is carried out. According to the scheme, the positions of the pedestrians and the vehicles can be accurately predicted and obtained by dynamically predicting the specific positions of the pedestrians and the vehicles in a future period of time, the prediction precision is high, the possibility of collision between the vehicles and the pedestrians can be effectively predicted, the collision between the pedestrians and the vehicles can be effectively prevented, and the purpose of reducing traffic accidents is achieved.
Drawings
FIG. 1 is a schematic flow chart of a vehicle pedestrian collision prevention method of the present invention;
FIG. 2 is a schematic illustration of a positional relationship between a first predicted movement area of a pedestrian and a vehicle in accordance with an embodiment of the present invention;
fig. 3 is a block diagram of the construction of a vehicle pedestrian collision prevention system of the present invention.
Detailed Description
Method embodiment
Referring to fig. 1, the present invention provides a vehicle pedestrian collision preventing method including the steps of:
according to the first real-time motion data of the pedestrian, dynamically predicting a first predicted position of the pedestrian after a set time interval delta t by adopting a pedestrian prediction model;
according to the second real-time motion data of the vehicle, dynamically predicting a second predicted position of the vehicle after a set time interval delta t by using a vehicle prediction model;
according to the first real-time moving speed of the pedestrian, a first prediction moving area of the pedestrian after a set time interval delta t is described by combining the first prediction position;
and predicting and acquiring the possible collision between the vehicle and the pedestrian according to the coincidence condition of the first predicted motion area of the pedestrian and the second predicted position of the vehicle, and performing collision early warning.
In the scheme, if a plurality of pedestrians are around the vehicle, the method is respectively adopted to predict the possibility of collision between the vehicle and each pedestrian, and collision early warning is carried out.
According to the method, a first predicted position of a pedestrian after a set time interval delta t and a second predicted position of a vehicle after the set time interval delta t are obtained through dynamic prediction, so that a first predicted movement region of the pedestrian after the set time interval delta t is drawn correspondingly in combination with the real-time movement speed of the pedestrian, the collision which is possibly caused by the vehicle and the pedestrian can be predicted and obtained according to the superposition condition of the first predicted movement region of the pedestrian and the second predicted position of the vehicle, and collision early warning is carried out. Since the vehicle driving path is generally determined and the position of the vehicle after the set time interval Δ t is relatively determined, the scheme mainly considers the uncertainty of the pedestrian motion to predict the collision. According to the scheme, the possibility of collision between the pedestrian and the vehicle can be accurately predicted by dynamically predicting the specific positions of the pedestrian and the vehicle after a period of time in the future and combining the uncertainty of the movement of the pedestrian, the prediction precision is high, and the possibility of collision between the vehicle and the pedestrian can be effectively predicted, so that the collision between the pedestrian and the vehicle can be effectively prevented, and the purpose of reducing traffic accidents is achieved.
Further as a preferred embodiment, the method further comprises the following steps:
the first real-time motion data of the pedestrian and the second real-time motion data of the vehicle are synchronized based on the time stamp. In the specific processing process, if the sampling frequency of the first real-time motion data is different from that of the second real-time motion data, the data needs to be converted into the same frequency, so that the vehicle data and the pedestrian data can be in one-to-one correspondence, and mathematical operation is facilitated. In actual test, the sampling frequency of the vehicle and the pedestrian is only required to be consistent.
Further as a preferred embodiment, the first real-time motion data and the first predicted position each include a position, a speed and a moving direction of the pedestrian, and the second real-time motion data and the second predicted position each include a position, a speed and a moving direction of the vehicle.
The pedestrian first real-time motion data can be acquired through a pedestrian mobile terminal, then is connected with an automobile and is sent to the automobile, and the pedestrian first real-time motion data can also be acquired through analysis and calculation after an image of the pedestrian is acquired in real time through an image sensor arranged on the automobile. And the second real-time motion data of the vehicle is directly acquired through the bus.
Preferably, in this embodiment, the pedestrian prediction model and the vehicle prediction model are both obtained by dynamic training using a multilayer perceptual neural network or a nonlinear regression neural network, and a hidden layer of the neural network is calculated using a bayesian regularization algorithm or a back propagation algorithm.
It should be noted that the pedestrian prediction model and the vehicle prediction model can be established in various ways, for example, in the simplest manner, the prediction is performed by using the GPS inertial navigation technology, and the operation state after the set time interval Δ t is obtained by calculation through a simple formula assuming that the vehicle and the pedestrian maintain the previous operation state. The motion prediction of pedestrians and vehicles can be realized.
Further preferably, in the dynamic prediction process of the pedestrian prediction model and the vehicle prediction model, the corresponding first real-time motion data and second real-time motion data are used as training data, and dynamic prediction is performed by the following method:
carrying out data cleaning on the training data, and filtering noise data;
sampling the cleaned training data according to the set sampling frequency and the number of sampling points to obtain input data of the prediction model corresponding to different moments;
and inputting the input data at different moments into a multilayer perception neural network or a nonlinear regression neural network in real time for training, and dynamically predicting to obtain the real-time predicted position of the pedestrian or the vehicle after a set time interval.
In this embodiment, a dynamic prediction mode is adopted for prediction, that is, input data of each training of the neural network is dynamic, so that a predicted position of a vehicle or a pedestrian after a set time interval Δ t at each moment is obtained.
For example, the sampling frequency is set to be 10Hz, the number of sampling points is 20, then training data within 2s before the current time is sampled during each training, which is equivalent to that the historical track within 2s is adopted as input data for dynamic prediction during each training, and the above steps are repeated in such a circulating way, so that the real-time position of the pedestrian/vehicle can be obtained through dynamic prediction at any moment, the problem of inaccurate prediction in the current static prediction technology is solved, and the real-time position of the pedestrian/vehicle can be effectively and accurately obtained through prediction.
Further as a preferred embodiment, the dynamic prediction process is performed by performing prediction calculation in a UTM coordinate system, and before the step of performing data cleaning on the training data and filtering out noise data, the method further includes the following steps:
and converting the training data from the GPS coordinate system into the data of the UTM coordinate system. Through calculation after conversion, the coordinate distribution condition of the training data can be visually observed, the interference of latitude can be avoided, and most importantly, the application and calculation of a physical mathematical formula are facilitated.
The first predicted movement region of the pedestrian can be set to be obtained based on probability theory algorithm drawing, and the normal distribution rule is conformed, so that the center of the region represents that the probability of collision is higher, the edge represents that the probability of collision is lower, when the coincidence position of the first predicted movement region of the pedestrian and the second predicted position of the vehicle is at the edge, the probability of collision is lower, and when the coincidence position is closer to the center region, the probability of collision is higher, therefore, the probability of collision between the pedestrian and the vehicle can be judged by calculating the coincidence condition of the first predicted movement region and the second predicted position.
In particular, the first predicted motion region may be depicted as a circle, a sector, or the like, primarily based on pedestrian motion data. For example, the first predicted movement region may be depicted as a circle based on the randomness of the pedestrian movement, or the first predicted movement region may be depicted as a sector based on the forward characteristic of the pedestrian movement, and then the first predicted movement region is subjected to coincidence calculation with the second predicted position, and mainly the coincidence situation of the circle region or the sector region with the second predicted position is calculated to perform collision prediction warning.
As a preferred embodiment of the present invention, the first predicted motion region is set to be circular. And the specific depiction is as follows:
the first prediction motion area and the first real-time motion speed are in positive correlation, and positive correlation coefficients of the first prediction motion area and the first real-time motion speed are read from a first coefficient database, wherein the first coefficient database is obtained by the following method:
a large amount of pedestrian movement speeds and corresponding movement areas are obtained and used as training data, after training is conducted, curve relations between the pedestrian movement speeds and the movement areas are obtained through fitting, and therefore after positive correlation coefficients between different movement speeds of pedestrians and the corresponding movement areas are obtained, a first coefficient database is established and obtained.
As another preferred example of the present invention, therefore, referring to fig. 2, the step of predicting to obtain a collision that may occur between a vehicle and a pedestrian and performing collision warning specifically includes:
calculating and obtaining the distance D between the vehicle and the pedestrian according to the first predicted position and the second predicted position;
calculating to obtain the braking distance Ds of the vehicle according to the second real-time movement speed of the vehicle;
the method comprises the following steps of predicting and obtaining the possible collision of a vehicle and a pedestrian, and carrying out corresponding collision early warning:
judging that the vehicle and the pedestrian do not collide when D-Ds-L is larger than Dw under the first condition;
judging the possibility of collision between the vehicle and the pedestrian when Dw > D-Ds-L > Dd, and warning the vehicle to pay attention to the pedestrian;
judging that the vehicle and the pedestrian have collision danger when the D-Ds-L is less than the Dd, and warning the vehicle to take defensive measures and/or actively taking defensive measures;
wherein, L represents the distance of the vehicle body, Dw represents the radius of the warning area corresponding to the pedestrian, Dd represents the radius of the danger area corresponding to the pedestrian, and Dw and Dd are both the drawing parameters of the first predicted motion area corresponding to the pedestrian.
The above embodiment mentions that the first predicted movement area of the pedestrian conforms to the probability distribution algorithm, different areas represent different collision possibilities, so that when the first predicted movement area is depicted, a warning area and a dangerous area can be depicted, the warning area represents that there is a possibility of collision, safety driving needs to be paid attention to, the dangerous area represents that there is a high possibility of collision, and the pedestrian enters the area, so that a vehicle needs to be warned to take defensive measures, and/or defensive measures need to be taken actively, for example, braking can be prompted by voice, or braking is performed actively, the driving direction is changed, the driving speed is changed, and the like.
System embodiment
Referring to fig. 3, the vehicle-pedestrian collision prevention system includes:
at least one processor 100;
at least one memory 200 for storing at least one program;
when the at least one program is executed by the at least one processor 100, the at least one processor 100 is caused to implement the vehicle pedestrian collision prevention method.
The vehicle and pedestrian collision prevention system can execute the vehicle and pedestrian collision prevention method provided by the method embodiment of the invention, can execute any combination of the implementation steps of the method embodiment, and has corresponding functions and beneficial effects of the method.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (8)
1. A vehicle pedestrian collision prevention method, characterized by comprising the steps of:
according to the first real-time motion data of the pedestrian, dynamically predicting a first predicted position of the pedestrian after a set time interval by adopting a pedestrian prediction model;
according to the second real-time motion data of the vehicle, dynamically predicting a second predicted position of the vehicle after a set time interval by using a vehicle prediction model;
according to the first real-time moving speed of the pedestrian, a first prediction moving area of the pedestrian after a set time interval is drawn by combining a first prediction position, and the first prediction moving area and the first real-time moving speed are in positive correlation;
predicting and obtaining the collision which is possibly generated between the vehicle and the pedestrian according to the superposition condition of the first predicted motion area of the pedestrian and the second predicted position of the vehicle, and carrying out collision early warning;
when the pedestrian prediction model and the vehicle prediction model carry out dynamic prediction, corresponding first real-time motion data and second real-time motion data are used as training data;
and when the pedestrian prediction model and the vehicle prediction model carry out dynamic prediction, acquiring input data of the prediction models corresponding to different moments from the training data, and training the prediction models by using the input data at different moments in real time.
2. The vehicular pedestrian collision prevention method according to claim 1, characterized by further comprising the steps of:
the first real-time motion data of the pedestrian and the second real-time motion data of the vehicle are synchronized based on the time stamp.
3. The vehicle pedestrian collision prevention method according to claim 1, wherein the first real-time motion data and the first predicted position each include a position, a speed, and a moving direction of the pedestrian, and the second real-time motion data and the second predicted position each include a position, a speed, and a moving direction of the vehicle.
4. The vehicle pedestrian collision prevention method according to claim 1, wherein the pedestrian prediction model and the vehicle prediction model are both obtained by dynamic training using a multilayer perceptual neural network or a nonlinear regression neural network, and a hidden layer of the neural network is calculated using a bayesian regularization algorithm or a back propagation algorithm.
5. The vehicle pedestrian collision prevention method according to claim 4, wherein the pedestrian prediction model and the vehicle prediction model use the corresponding first real-time motion data and the second real-time motion data as training data in a dynamic prediction process, and perform dynamic prediction by:
carrying out data cleaning on the training data, and filtering noise data;
sampling the cleaned training data according to the set sampling frequency and the number of sampling points to obtain input data of the prediction model corresponding to different moments;
and inputting the input data at different moments into a multilayer perception neural network or a nonlinear regression neural network in real time for training, and dynamically predicting to obtain the real-time predicted position of the pedestrian or the vehicle after a set time interval.
6. The method according to claim 5, wherein the dynamic prediction process is performed by prediction calculation under a UTM coordinate system, and the step of performing data cleaning on the training data and filtering out noise data further comprises the following steps:
and converting the training data from the GPS coordinate system into the data of the UTM coordinate system.
7. The method for preventing collision between a vehicle and a pedestrian according to claim 1, wherein the step of predicting to obtain the possible collision between the vehicle and the pedestrian and performing collision warning specifically comprises:
calculating and obtaining the distance D between the vehicle and the pedestrian according to the first predicted position and the second predicted position;
calculating to obtain the braking distance Ds of the vehicle according to the second real-time movement speed of the vehicle;
the method comprises the following steps of predicting and obtaining the possible collision of a vehicle and a pedestrian, and carrying out corresponding collision early warning:
judging that the vehicle and the pedestrian do not collide when D-Ds-L is larger than Dw under the first condition;
judging the possibility of collision between the vehicle and the pedestrian when Dw > D-Ds-L > Dd, and warning the vehicle to pay attention to the pedestrian;
judging that the vehicle and the pedestrian have collision danger when the D-Ds-L is less than the Dd, and warning the vehicle to take defensive measures and/or actively taking defensive measures;
wherein, L represents the distance of the vehicle body, Dw represents the radius of the warning area corresponding to the pedestrian, Dd represents the radius of the danger area corresponding to the pedestrian, and Dw and Dd are both the drawing parameters of the first predicted motion area corresponding to the pedestrian.
8. A vehicle pedestrian collision prevention system, comprising:
at least one processor;
at least one memory for storing at least one program;
when the at least one program is executed by the at least one processor, the at least one processor is caused to implement the vehicle pedestrian collision prevention method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810483121.4A CN108674413B (en) | 2018-05-18 | 2018-05-18 | Vehicle and pedestrian collision prevention method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810483121.4A CN108674413B (en) | 2018-05-18 | 2018-05-18 | Vehicle and pedestrian collision prevention method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108674413A CN108674413A (en) | 2018-10-19 |
| CN108674413B true CN108674413B (en) | 2021-02-19 |
Family
ID=63805338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810483121.4A Active CN108674413B (en) | 2018-05-18 | 2018-05-18 | Vehicle and pedestrian collision prevention method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108674413B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111223302B (en) * | 2018-11-23 | 2021-12-03 | 明创能源股份有限公司 | External coordinate real-time three-dimensional road condition auxiliary device for mobile carrier and system |
| US11260852B2 (en) * | 2019-03-26 | 2022-03-01 | GM Global Technology Operations LLC | Collision behavior recognition and avoidance |
| DE102019206945A1 (en) * | 2019-05-14 | 2020-11-19 | Volkswagen Aktiengesellschaft | Method for detecting a potential collision of a vehicle with a living being and a parking garage management system |
| CN110293968B (en) * | 2019-06-18 | 2021-09-28 | 百度在线网络技术（北京）有限公司 | Control method, device and equipment for automatic driving vehicle and readable storage medium |
| CN110443310B (en) * | 2019-08-07 | 2022-08-09 | 浙江大华技术股份有限公司 | Updating method of comparison analysis system, server and computer storage medium |
| JP2021092980A (en) * | 2019-12-10 | 2021-06-17 | 本田技研工業株式会社 | Information presentation device for automatic driving vehicle |
| CN111081067B (en) * | 2019-12-27 | 2021-07-20 | 武汉大学 | Vehicle collision early warning system and method based on IGA-BP neural network under vehicle networking environment |
| CN111634292B (en) * | 2020-05-18 | 2021-10-19 | 北京踏歌智行科技有限公司 | Collision prediction method for mining area |
| CN111739344B (en) * | 2020-06-29 | 2022-11-08 | 北京百度网讯科技有限公司 | Early warning method and device and electronic equipment |
| CN112349144B (en) * | 2020-11-10 | 2022-04-19 | 中科海微(北京)科技有限公司 | Monocular vision-based vehicle collision early warning method and system |
| CN114511999B (en) * | 2020-11-17 | 2023-09-01 | 宇通客车股份有限公司 | Pedestrian behavior prediction method and device |
| CN114228707B (en) * | 2021-12-22 | 2023-04-25 | 广西科技大学 | Anti-collision method and system for unmanned vehicle |
| CN114187764B (en) * | 2022-02-16 | 2022-05-13 | 深圳佑驾创新科技有限公司 | Method for rapidly detecting collision risk degree aiming at VRU (virtual Router Unit) passing scene |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102010011497A1 (en) * | 2010-03-16 | 2011-09-22 | GM Global Technology Operations LLC , (n. d. Ges. d. Staates Delaware) | Method for avoiding or mitigating a collision, control device for a driver assistance system and vehicle |
| DE112010005448B4 (en) * | 2010-04-05 | 2020-07-23 | Toyota Jidosha Kabushiki Kaisha | Collision assessment device for a vehicle |
| US8922431B2 (en) * | 2010-04-13 | 2014-12-30 | Becker Research And Development (Proprietary) Limited | Apparatus, a system and a method for collission avoidance |
| CN102765365B (en) * | 2011-05-06 | 2014-07-30 | 香港生产力促进局 | Pedestrian detection method based on machine vision and pedestrian anti-collision warning system based on machine vision |
| JP5668862B2 (en) * | 2011-09-14 | 2015-02-12 | トヨタ自動車株式会社 | Driving support device and driving support method |
| CN105761488B (en) * | 2016-03-30 | 2018-11-23 | 湖南大学 | Real-time extreme learning machine Short-time Traffic Flow Forecasting Methods based on fusion |
-
2018
- 2018-05-18 CN CN201810483121.4A patent/CN108674413B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108674413A (en) | 2018-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108674413B (en) | Vehicle and pedestrian collision prevention method and system | |
| Åsljung et al. | Using extreme value theory for vehicle level safety validation and implications for autonomous vehicles | |
| Wachenfeld et al. | The worst-time-to-collision metric for situation identification | |
| Shirazi et al. | Looking at intersections: a survey of intersection monitoring, behavior and safety analysis of recent studies | |
| CN111489588B (en) | Vehicle driving risk early warning method and device, equipment and storage medium | |
| Jiménez et al. | An improved method to calculate the time-to-collision of two vehicles | |
| CN110097786B (en) | Vehicle-vehicle collision detection method based on V2X and application system | |
| Batz et al. | Recognition of dangerous situations within a cooperative group of vehicles | |
| US11648936B2 (en) | Method and apparatus for controlling vehicle | |
| Kim et al. | Probabilistic threat assessment with environment description and rule-based multi-traffic prediction for integrated risk management system | |
| CN105374231A (en) | Early warning method, device and system | |
| CN107274721B (en) | Multi-vehicle cooperative positioning method in intelligent transportation system | |
| CN113844444A (en) | Vehicle forward collision early warning method and device, electronic equipment and vehicle | |
| CN111469837A (en) | Vehicle collision prediction method and device | |
| CN109345870B (en) | Early warning method and device for preventing vehicle collision | |
| US20220089192A1 (en) | Risk estimation in autonomous driving environments | |
| WO2016107876A1 (en) | Vehicular motion monitoring method | |
| EP3696791A1 (en) | Prediction of intention of path deviance for vehicles | |
| CN110745119A (en) | Anti-collision method and device | |
| CN111710188B (en) | Vehicle alarm prompting method, device, electronic equipment and storage medium | |
| CN110491167A (en) | A kind of section alarming method for power and equipment | |
| CN116088538B (en) | Vehicle track information generation method, device, equipment and computer readable medium | |
| CN116546067A (en) | Internet of vehicles formation method, system and medium based on hong Mongolian system | |
| CN111731285B (en) | Vehicle anti-collision method and device based on V2X technology | |
| CN111583633B (en) | Early warning method and device for vehicle collision |
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 | ||
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: 510000 No.8 Songgang street, Cencun, Tianhe District, Guangzhou City, Guangdong Province Patentee after: GUANGZHOU XPENG AUTOMOBILE TECHNOLOGY Co.,Ltd. Address before: Room 333, Guangzhou, China Patentee before: GUANGZHOU XPENG AUTOMOBILE TECHNOLOGY Co.,Ltd. |