CN114063079B - Target confidence coefficient acquisition method and device, radar system and electronic device - Google Patents

Target confidence coefficient acquisition method and device, radar system and electronic device Download PDF

Info

Publication number
CN114063079B
CN114063079B CN202111186684.5A CN202111186684A CN114063079B CN 114063079 B CN114063079 B CN 114063079B CN 202111186684 A CN202111186684 A CN 202111186684A CN 114063079 B CN114063079 B CN 114063079B
Authority
CN
China
Prior art keywords
target
detected
predicted
current frame
acquiring
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
Application number
CN202111186684.5A
Other languages
Chinese (zh)
Other versions
CN114063079A (en
Inventor
洪帅鑫
王明辉
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.)
Freetech Intelligent Systems Co Ltd
Original Assignee
Freetech Intelligent Systems Co Ltd
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 Freetech Intelligent Systems Co Ltd filed Critical Freetech Intelligent Systems Co Ltd
Priority to CN202111186684.5A priority Critical patent/CN114063079B/en
Publication of CN114063079A publication Critical patent/CN114063079A/en
Application granted granted Critical
Publication of CN114063079B publication Critical patent/CN114063079B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/04—Systems determining presence of a target
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/04—Systems determining the presence of a target
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/50—Systems of measurement based on relative movement of target
    • G01S17/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88—Lidar systems specially adapted for specific applications
    • G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
    • G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The application relates to a target confidence coefficient acquisition method, a target confidence coefficient acquisition device, a radar system and an electronic device, wherein the target confidence coefficient acquisition method comprises the following steps: acquiring track data of a target to be detected; determining the shielded state of the target to be detected of the current frame based on the track data; if the current frame of target to be detected is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of the predicted existence state of the target to be detected from the previous frame to the current frame; and acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability. By the method and the device, the problem that the accuracy of the target confidence coefficient is low is solved, and the technical effect of accurately acquiring the target confidence coefficient is achieved.

Description

Target confidence coefficient acquisition method and device, radar system and electronic device
Technical Field
The application relates to the field of intelligent driving, in particular to a target confidence obtaining method and device, a radar system and an electronic device.
Background
With the popularization and development of Advanced Driver Assistance (ADAS) in vehicles, more sophisticated advanced driver assistance functions place higher demands on sensing external environmental information. The vehicle-mounted millimeter wave radar plays an important role in the whole advanced driver assistance system and has the characteristic of all weather. In advanced driver assistance systems, multi-sensor fusion techniques are important. The multi-sensor fusion technology fuses target information output by various sensors together to form fused target information, and then makes a decision by using the fused target information to realize an auxiliary driving function. However, each sensor outputs some false target information at the same time as outputting real target information. Therefore, the fusion technique requires that each sensor provide the target confidence while outputting the target information. The target confidence can represent the probability that the current target is a real target, and an algorithm in the fusion technology can be further realized only when the current target is the real target. Therefore, the accuracy of the target confidence information is particularly important in advanced driver assistance techniques.
When the vehicle-mounted radar detects target information, the current target is shielded by other targets, and at the moment, the vehicle-mounted radar judges that the target disappears in a detection visual field, so that the confidence coefficient of the target is rapidly reduced, the realization of an internal algorithm of a multi-sensor fusion technology is influenced, and the normal operation of a driving auxiliary function is influenced.
Aiming at the problem that the accuracy of the target confidence coefficient is low in the related technology, no effective solution is provided at present.
Disclosure of Invention
The embodiment provides a target confidence coefficient obtaining method and device, a radar system and an electronic device, so as to solve the problem that the accuracy of a target confidence coefficient is low in the related art.
In a first aspect, in this embodiment, a method for obtaining a target confidence level is provided, including:
acquiring track data of a target to be detected;
determining the shielded state of the target to be detected of the current frame based on the track data;
if the target to be detected of the current frame is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted target existence and a predicted target nonexistence, the predicted existence indicates that a vehicle sensor detects the target to be detected at the corresponding moment, and the predicted nonexistence indicates that the vehicle sensor does not detect the target to be detected at the corresponding moment;
and acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability.
In one embodiment, the acquiring the track data of the target to be detected includes: detecting whether the point track exists in the current frame of the target to be detected based on the track data; if the point trace exists, acquiring the correlation quality score of the track data and the point trace of the target to be detected in the current frame; and acquiring the second target confidence of the target to be detected in the current frame based on the associated quality score.
In one embodiment of the above, the obtaining the associated quality score of the track data and the point track of the target to be detected in the current frame includes: if the target to be detected has a plurality of point tracks in the current frame, acquiring the associated parameters of each point track and the track data, wherein the associated parameters at least comprise the distance relationship, the speed relationship and the track service life between each point track and the track data; determining the association quality score based on the association parameter.
In one embodiment, the change probability includes a first parameter, a second parameter, a third parameter, and a fourth parameter; the first parameter is the probability that the target to be detected exists in the predicted target of the previous frame and the predicted target of the current frame; the second parameter is the probability that the target to be detected exists in the predicted target of the previous frame and does not exist in the predicted target of the current frame; the third parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and exists in the predicted target of the current frame; the fourth parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and does not exist in the predicted target of the current frame.
In one embodiment, if the current frame is blocked, the method further includes, after obtaining a first target confidence of the target to be detected in a previous frame and a change probability that a predicted existing state of the target to be detected changes from the previous frame to the current frame: and reducing the first parameter to a first preset value and increasing the second parameter to a second preset value.
In one embodiment of the foregoing, after determining the occluded state of the target to be detected in the current frame based on the track data, the method further includes: acquiring radar detection probability; if the target to be detected is not shielded, acquiring the likelihood ratio of the target to be detected based on the radar detection probability; obtaining the second target confidence based on the likelihood ratio.
In one embodiment, the obtaining the radar detection probability further includes: if the target to be detected is blocked, reducing the radar detection probability by a third preset value to obtain a corrected detection probability; and acquiring the likelihood ratio of the target to be detected based on the corrected detection probability.
In a second aspect, in this embodiment, there is provided an object confidence obtaining apparatus, including:
the acquisition module is used for acquiring track data of the target to be detected;
the occlusion detection module is used for determining the occluded state of the target to be detected of the current frame based on the track data;
the position detection module is used for acquiring a first target confidence coefficient of a target to be detected in a previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the vehicle sensor detects the target to be detected at a corresponding moment, and the predicted object nonexistence indicates that the vehicle sensor does not detect the target to be detected at the corresponding moment;
and the calculation module is used for acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability.
In a third aspect, there is provided in this embodiment a radar system comprising: the system comprises a detection radar and a control unit, wherein the detection radar is connected with the control unit; the detection radar is used for acquiring track information and track information of the target to be detected; the control unit is used for determining the shielded state of the target to be detected of the current frame based on the track data; if the target to be detected in the current frame is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the target to be detected is detected by a vehicle sensor at a corresponding moment, and the predicted nonexistence indicates that the target to be detected is not detected by the vehicle sensor at a corresponding moment; acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability; the control unit is further configured to implement the target confidence level obtaining method according to the first aspect.
In a fourth aspect, in this embodiment, there is provided an electronic apparatus, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the target confidence obtaining method according to the first aspect when executing the computer program.
Compared with the related art, the target confidence coefficient obtaining method provided by the embodiment obtains the track data of the target to be detected; determining the shielded state of the target to be detected of the current frame based on the track data; if the target to be detected in the current frame is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the target to be detected is detected by a vehicle sensor at a corresponding moment, and the predicted nonexistence indicates that the target to be detected is not detected by the vehicle sensor at a corresponding moment; and acquiring the second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability, so that the problem of low accuracy of the target confidence coefficient is solved, and the technical effect of accurately acquiring the target confidence coefficient is realized.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more concise and understandable description of the application, and features, objects, and advantages of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a block diagram of a hardware structure of a terminal of a target confidence obtaining method according to an embodiment of the present application;
FIG. 2 is a flow chart of a target confidence acquisition method according to an embodiment of the application;
FIG. 3 is a schematic diagram illustrating an occluded target according to the target confidence obtaining method of the embodiment of the present application;
fig. 4 is a schematic diagram of a conversion model of whether an event occurs in a target to be detected according to the target confidence obtaining method in the embodiment of the present application;
FIG. 5 is a flow diagram of a target confidence acquisition method according to another embodiment of the present application;
fig. 6 is a block diagram of a target confidence obtaining apparatus according to an embodiment of the present application.
Detailed Description
For a clearer understanding of the objects, aspects and advantages of the present application, reference is made to the following description and accompanying drawings.
Unless defined otherwise, technical or scientific terms used herein shall have the same general meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The use of the terms "a" and "an" and "the" and similar referents in the context of describing the invention (including a reference to the context of the specification and claims) are to be construed to cover both the singular and the plural, as well as the singular and plural. The terms "comprises," "comprising," "has," "having," and any variations thereof, as referred to in this application, are intended to cover non-exclusive inclusions; for example, a process, method, and system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or modules, but may include other steps or modules (elements) not listed or inherent to such process, method, article, or apparatus. Reference throughout this application to "connected," "coupled," and the like is not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Reference to "a plurality" in this application means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. In general, the character "/" indicates a relationship in which the objects associated before and after are an "or". The terms "first," "second," "third," and the like in this application are used for distinguishing between similar items and not necessarily for describing a particular sequential or chronological order.
The method embodiment provided in this embodiment may be executed in a terminal, a computer, a radar, and similar computing device, and is preferably applicable to a millimeter wave radar system, and fig. 1 is a hardware configuration block diagram of a terminal according to the target confidence obtaining method of the embodiment of the present application. As shown in fig. 1, the terminal includes a radio frequency device, a processor, an internal bus, a memory, and a network interface. The radio frequency device includes an antenna module that can transmit and receive electromagnetic waves through an antenna. The number of processors may be one or more (only one shown in fig. 1). The processor is used for receiving electromagnetic wave signals and interacting with the vehicle control system through the internal bus. The network interface is used for network connection, and the network may be a wireless network, an ethernet network, a mobile communication network, or the like. The memory is used for storing the electromagnetic wave data processed by the processor and a computer program, and the computer program includes a software program and a module of application software, such as a computer program corresponding to the target confidence degree obtaining method in the embodiment.
Automatic driving is an important application in an intelligent traffic system, and requires that vehicles detect surrounding vehicles and avoid according to the surrounding vehicles so as to avoid traffic accidents. Sensors commonly used in vehicles today include cameras and radars. When the radar detects target information, the current target is shielded by other targets. At this time, the radar considers that the target disappears in the detection range, and correspondingly, the target confidence coefficient is sharply reduced, however, the current target is only shielded by other targets, so that the radar is not detected, and the target still exists in the radar detection range, and obviously, the sharp reduction of the target confidence coefficient is not in accordance with the actual situation and is inaccurate. Therefore, a method for improving the accuracy of the confidence of the target is needed.
In this embodiment, a target confidence obtaining method is provided, and fig. 2 is a flowchart of the target confidence obtaining method according to the embodiment of the present application, and as shown in fig. 2, the flowchart includes the following steps:
step S201, acquiring track data of the target to be detected.
Specifically, the track data of the target to be detected is obtained through a radar, and the radar can be a millimeter wave radar and also can be a laser radar. Objects to be detected include, but are not limited to, pedestrians, vehicles, traffic signs, obstacles, and the like.
In the field of object detection, it is assumed that the data of the object under study detected by the radar sensor include a position x and a velocity v, and the predicted value is an x value and a v value at the next time according to the x value and the v value at the time. The measured values are obtained from sensors, i.e. radars, which are not accurate because of the presence of various errors, and therefore they are weighted and fused to obtain a new quantity, called the state value. A trace of points is a so-called measurement, which is a series of points detected by the radar on the same object. And the state value corresponding to the flight path is the weighted fusion of the point path and the predicted path.
Step S202, determining the shielded state of the target to be detected of the current frame based on the track data.
Specifically, the track data includes multiple frames of data, each frame corresponds to a detection time of the radar, and the current frame represents the current time. Whether the target to be detected is shielded in the current frame can be determined according to the track data. For example, whether the target to be detected is blocked can be judged by calling track data of the current frame, namely the track data of the current moment, combining the track data of other targets at the same moment and according to the distance and the angle between the track frame of the target to be detected and the track frames of other targets to the current vehicle. It should be emphasized that the occlusion state in the embodiment of the present application refers to occlusion in a short time, that is, the object is occluded within a preset time period. Fig. 3 is a schematic diagram of a target being blocked according to the target confidence obtaining method in the embodiment of the present application, as shown in fig. 3, at this time, a target to be detected is blocked by an obstacle.
Step S203, if the target to be detected in the current frame is blocked, obtaining a first target confidence of the target to be detected in the previous frame and a change probability of the predicted existence state of the target to be detected from the previous frame to the current frame.
Specifically, the target confidence of the initial frame may be set manually to an initial value. When the target confidence coefficient is calculated for the next frame of the initial frame, the initial value of the initial frame is used as the first target confidence coefficient for calculation. And taking the calculation result of the previous frame as the first target confidence coefficient for each subsequent frame, and performing iterative calculation.
The predicted presence state comprises a predicted target presence and a predicted target absence, the predicted presence indicates that the vehicle sensor detects the target to be detected at the corresponding time, and the predicted target absence indicates that the vehicle sensor does not detect the target to be detected at the corresponding time. The vehicle sensor may be a radar sensor, an image sensor, or the like. Preferably, the vehicle sensor is a millimeter wave radar. Predicting the existence of the target refers to that the millimeter wave radar detects the target to be detected. Generally, once an object to be detected is detected by the millimeter wave radar, the object may be detected by the millimeter wave radar for a period of time thereafter. However, since the system can only make a decision based on the presence or absence of the target at the present time, what is obtained is a predicted value, i.e., the predicted target will be detected again in the next period of time. If the predicted target does not exist, the target is not detected at the current time and in the next period of time. For example: the subject vehicle travels suddenly and then turns suddenly and is then blocked by a metal reflective panel or a building. Although the target disappears at the moment when the target turns, the system cannot judge whether the target appears again behind the target according to data at this moment, that is, the target is only temporarily blocked by other vehicles or the vehicle turns and does not appear in the field of view any more, so that judgment of the absence of the predicted target is made and the probability of the absence of the predicted target is given, which can be set by self-definition. At the next moment, the probability that the target does not exist at the next moment is also given. Since both the present time and the next time are predictions of the presence or absence of a target, there are four possibilities from the present time to the next time, that is, a state transition of the presence of a target to the presence of a target, a state transition of the presence of a target to the absence of a target, a state transition of the absence of a target to the presence of a target, and a state transition of the absence of a target to the absence of a target. According to actual needs, different probabilities, namely, the change probabilities, can be configured for the four cases. It should be emphasized that the driving view in this embodiment is not a radar detection view, and the above scenario is also taken as an example: the subject vehicle travels suddenly and then turns suddenly and is then blocked by a metal reflective panel or a building. At this time, although the radar does not detect the vehicle, the vehicle actually still exists in the detection range of the radar.
And step S204, acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability.
Through the steps, the target confidence obtaining method of the application judges whether the target to be detected is shielded or not, and if the current frame of the target to be detected is shielded, the second target confidence of the current frame is determined according to the probability of the target to be detected in the radar detection range of the previous frame and the probability of the change from the existing state of the previous frame to the existing state of the current frame. According to the method, the confidence of the target when the target is occluded can be evaluated, and because the confidence is related to the confidence of the target in the previous frame, the problem of rapid reduction of the confidence caused by occlusion is avoided, and the accuracy of the confidence of the target is improved.
In one embodiment, the acquiring the track data of the target to be detected includes: detecting whether the point track exists in the current frame of the target to be detected based on the track data; if the point trace exists, acquiring the correlation quality score of the track data and the point trace of the target to be detected in the current frame; and acquiring the second target confidence of the target to be detected in the current frame based on the associated quality score.
Specifically, assuming that the radar detects the object of interest as a position x and a velocity v, the position x and the velocity v are measured values detected by the radar, the object of interest is an object to be detected, and the measured values are traces. If the current frame is the kth frame, acquiring track data of the kth frame, detecting whether the point track exists in the kth frame or not based on the track data, if the point track exists, associating the track data with the point track, acquiring an association quality score of the track data of the target to be detected in the kth frame and the point track of the target to be detected, and acquiring a first target confidence coefficient of the target to be detected in the kth frame according to the association quality score. In the process of acquiring the track data, the track frame of each target is acquired, correspondingly, the position and the speed of each point on the track frame can be acquired, the position and the speed of each point on the track frame are matched with the track of the point detected at the same moment through a preset algorithm, if the point on the track frame can be matched with the track of the point, the track of the point is associated, and the associated quality score is calculated. If the point on the track frame can not be matched with the point track, the point track is not associated. If the flight path of the kth frame is associated with the point path, calculating an associated quality score q, and further calculating a likelihood ratio, wherein the calculation formula of the likelihood ratio is as follows:
Lk＝1-Pd+d*q*ρ
wherein, L iskRepresenting the likelihood ratio, P, of the k-th framedAnd representing radar detection probability, representing clutter density, and calculating the correlation quality fraction q by considering the number of track correlation points, the track and the like. Then, according to the conditional probability, obtaining the probability of the occurrence of the event of the target to be detected in the previous frame, namely a second target confidence:
Figure BDA0003299522620000081
and judging whether the track data of the current frame exists in the point track or not, so that the reliability of the data can be preliminarily judged. Because the flight path is a state value, the flight path is generally data obtained according to the point path and the predicted value of the research object. Therefore, if the point trace exists in the target to be detected in the k-th frame, it is indicated that the data reliability of the k-th frame is high, and the second target confidence with high accuracy can be obtained based on the preset target confidence calculation method and the associated quality scores of the flight trace and the point trace.
In one embodiment of the above, the obtaining the associated quality score of the track data and the point track of the target to be detected in the current frame includes: if the target to be detected has a plurality of point tracks in the current frame, acquiring the associated parameters of each point track and the track data, wherein the associated parameters at least comprise the distance relationship, the speed relationship and the track service life between each point track and the track data; determining the association quality score based on the association parameter.
Specifically, for the same target, the radar detects a series of trace data about the position and speed of the target at different times. However, due to the existence of errors, the radar faces the same target at the same time, and may not scan the point trace or scan a plurality of point traces. Therefore, in this embodiment, the traces are evaluated according to the position relationship, the distance relationship, the speed relationship, and the like between each trace and each track to obtain the evaluation result between each trace and each track of the target, the evaluation result reflects the accuracy of the trace, the trace with the highest accuracy is taken as the standard trace of the frame, the association quality score is calculated according to the evaluation result of the standard trace, which is helpful for improving the accuracy of the association quality score, and further, when the second target confidence coefficient of the target is calculated based on the association quality score, the accuracy of the second target confidence coefficient is improved.
The markov transition matrix method is a method for analyzing development trend by using a transition probability matrix. The method considers that some factors of a system are in transition, the nth result is only influenced by the (n-1) th result, and is only related to the current state and is not related to other conditions. In markov analysis, the concept of state transition is introduced. The state refers to a state in which an objective thing may appear or exist; state transition refers to the probability of an objective thing transitioning from one state to another.
In one embodiment, the change probability includes a first parameter, a second parameter, a third parameter, and a fourth parameter; the first parameter is the probability that the target to be detected exists in the predicted target of the previous frame and the predicted target of the current frame; the second parameter is the probability that the target to be detected exists in the predicted target of the previous frame and does not exist in the predicted target of the current frame; the third parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and exists in the predicted target of the current frame; the fourth parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and does not exist in the predicted target of the current frame.
Specifically, the change probability represents the probability of the markov state transition, which is suitable for the radar application scenario in the present application. Two situations should be included for whether the target to be detected has such an event at a certain time. One condition is that the target to be detected exists at the moment and can pass through chikIt is shown that,namely, the target to be detected exists at the moment k; in another case, the target to be detected does not exist at the moment, and the target to be detected can pass through
Figure BDA0003299522620000091
Indicating that the target to be detected does not exist at time k. And modeling the conversion of whether the event occurrence probability exists in the target to be detected or not by utilizing Markov state transition. Further, at adjacent moments, the existence state of the target to be detected includes four conditions, the first condition is that the target to be detected exists at the previous moment and exists at the current moment; the second situation is that the target to be detected exists at the previous moment and does not exist at the current moment; the third situation is that the target to be detected does not exist at the previous moment and exists at the current moment; the fourth case is that the object to be detected does not exist at the previous time and does not exist at the present time. Fig. 4 is a schematic diagram of a conversion model of whether an event occurs in a target to be detected according to the target confidence obtaining method in the embodiment of the present application. As shown in fig. 4. Wherein the first parameter a11A second parameter a12A third parameter a21A fourth parameter a22Respectively, represent the probability of transition between states. The first parameter represents the probability that the target to be detected exists in the k-1 frame and the k frame exists, and the second parameter represents the probability that the target to be detected exists in the k-1 frame and the k frame does not exist; the third parameter represents the probability that the target to be detected does not exist in the k-1 th frame and exists in the k-1 th frame, and the fourth parameter represents the probability that the target to be detected does not exist in the k-1 th frame and exists in the k-1 th frame. The relationship between the parameters is:
a11+a12＝1；
a21+a22＝1；
preferably, a11Is taken to be 0.98, a12Is taken to be 0.02, a21Is taken to be 0, a22The value of (c) takes 1. The logic set by the numerical value is that the target to be detected exists at the moment k-1, so that the target still exists at the moment k with higher probability; when the target to be detected does not exist at the moment k-1, the probability of the existence of the target to be detected at the moment k is extremely low theoretically. The specific numerical value can be adjusted according to actual conditions.
According to the four parameters, a predicted value P (x) of the occurrence probability of the event of the target to be detected at the moment k can be obtainedk|k-1) And the predicted value of the target non-existence event occurrence probability
Figure BDA0003299522620000092
The calculation method comprises the following steps:
Figure BDA0003299522620000093
Figure BDA0003299522620000094
wherein, P (χ)k-1) And the probability of the existence of the target to be detected at the moment k-1 is shown, and the probability is the same as the first target confidence coefficient of the target at the moment k-1.
Figure BDA0003299522620000095
Representing the probability that the object to be detected does not exist at time k-1.
In one embodiment, if the current frame is blocked, the method further includes, after obtaining a first target confidence of the target to be detected in a previous frame and a change probability between the target to be detected and a radar detection range from the previous frame to the current frame: and reducing the first parameter to a first preset value and increasing the second parameter to a second preset value.
Specifically, in the markov state transition process, the probability of converting the target existence into the target nonexistence is properly improved, and the probability of converting the target existence into the target existence is reduced.
In one embodiment of the above, after determining the occluded state of the target to be detected in the current frame based on the track data, the method further includes: acquiring radar detection probability; if the target to be detected is not shielded, acquiring the likelihood ratio of the target to be detected based on the radar detection probability; obtaining the second target confidence based on the likelihood ratio.
Specifically, if the target to be detected is not blocked, the formula L is based onk＝1-PdObtaining a likelihood ratio, wherein PdAnd the radar detection probability is represented and is a system preset value. According to conditional probability and formula
Figure BDA0003299522620000101
Figure BDA0003299522620000102
And obtaining a second target confidence of the current frame.
In one embodiment, the obtaining the radar detection probability further includes: if the target to be detected is shielded, reducing the radar detection probability by a third preset value to obtain a corrected detection probability; and acquiring the likelihood ratio of the target to be detected based on the corrected detection probability.
Specifically, if the radar is shielded, parameter adjustment is carried out, the numerical value of the second radar detection probability is used for obtaining the corrected detection probability, and then the formula L is used for correcting the detection probabilityk＝1-RdCalculating a likelihood ratio and based on the likelihood ratio and the formula
Figure BDA0003299522620000103
A second target confidence is calculated.
Through the steps, the target confidence coefficient obtaining method combines the thought method of Markov state transition with the target existing state change in the radar detection process, avoids the situation that the confidence coefficient is rapidly reduced, and improves the accuracy of the target confidence coefficient obtained in the state that the target is shielded.
The present embodiment is described and illustrated below by means of preferred embodiments.
Fig. 5 is a flowchart of a target confidence obtaining method according to another embodiment of the present application, and as shown in fig. 5, in the track association stage, track association scores of all the point tracks associated with the current track and the track are obtained, and a track association score with the smallest track association score is taken as the track association score. The track association score is an evaluation result obtained based on the association parameters, and the smaller the track association score is, the better the association effect is. The track association of the point track and the track takes the position relation, the speed relation, the track service life and other factors into consideration. Meanwhile, the number of track-associated points needs to be counted.
Defining a target presence event, the event comprising: chi shapekIndicating that the target exists at the time k;
Figure BDA0003299522620000104
indicating that the target does not exist at time k. Modeling the transition of whether there is a probability of occurrence of an event for a target using Markov state transitions, wherein a11、a12、a21、a22Respectively represent the probability of transition between states and satisfy the following relations:
a11+a12＝1
a21+a22＝1
at this time, a predicted value of the target occurrence probability and a predicted value of the target non-occurrence probability may be obtained, as follows:
Figure BDA0003299522620000111
Figure BDA0003299522620000112
if the flight path is associated with the point path, calculating the point path association quality q by utilizing the flight path association score, and further calculating the likelihood ratio, wherein the likelihood ratio under the condition is calculated as follows:
Lk＝1-Pd+Pd*q*ρ
wherein P isdAnd representing radar detection probability, representing clutter density, and considering the number of track associated points, track associated points and the like in the calculation of the associated quality q. Then according to the conditional probability, obtaining the probability of the k target of the current frame to have the event, namely the target positionReliability:
Figure BDA0003299522620000113
if the track is not associated with the trace point, performing logic of whether the track is blocked, namely whether the track exists in an intersection unit occupied by other tracks, wherein the schematic diagram of being blocked is as follows:
if the current frame target is blocked, the occurrence probability of the target event of the current frame is equal to the predicted value of the occurrence probability of the target event, namely the target confidence:
P(χk)＝P(χk|k-1)
in one embodiment, in the Markov state transition process, the probability of converting the target existence into the target nonexistence is properly improved, and the probability of converting the target existence into the target existence is simultaneously reduced, namely, the a is improved12And decrease a11The operation may reduce the prediction result of the target confidence, and then execute the following formula to obtain the target confidence:
Figure BDA0003299522620000114
P(χk)＝P(χk|k-1)
wherein, a11And a21The parameter adjustment can be carried out according to specific scenes.
If not, the likelihood ratio is calculated as follows: l isk＝1-Pd；
Then, according to the conditional probability, obtaining the probability of the occurrence of the event of the k target of the current frame, namely the target confidence:
Figure BDA0003299522620000115
it should be noted that the steps illustrated in the above-described flow diagrams or in the flow diagrams of the figures may be performed in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flow diagrams, in some cases, the steps illustrated or described may be performed in an order different than here.
The present embodiment further provides a target confidence obtaining device, which is used to implement the foregoing embodiments and preferred embodiments, and the description of the device that has been already made is omitted. The terms "module," "unit," "subunit," and the like as used below may implement a combination of software and/or hardware for a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware or a combination of software and hardware is also possible and contemplated.
Fig. 6 is a block diagram of a target confidence obtaining apparatus according to an embodiment of the present application, and as shown in fig. 6, the apparatus includes:
the acquisition module 10 is used for acquiring the track data of the target to be detected;
the occlusion detection module 20 is configured to determine, based on the track data, an occluded state of the target to be detected in the current frame;
the position detection module 30 is configured to, if the target to be detected in the current frame is blocked, obtain a first target confidence of the target to be detected in a previous frame and a change probability that a predicted presence state of the target to be detected from the previous frame to the current frame changes, where the predicted presence state includes a predicted target presence and a predicted target absence, the predicted presence indicates that the target to be detected is detected by the vehicle sensor at a corresponding time, and the predicted target absence indicates that the target to be detected is not detected by the vehicle sensor at the corresponding time;
and the calculating module 40 is configured to obtain a second target confidence of the target to be detected in the current frame based on the first target confidence and the change probability.
The position detection module 30 is further configured to detect whether a point trace exists in the current frame of the target to be detected based on the track data; if the point trace exists, acquiring the correlation quality score of the track data and the point trace of the target to be detected in the current frame; and acquiring the second target confidence of the target to be detected in the current frame based on the associated quality score.
The position detection module 30 is further configured to, if the target to be detected has a plurality of the point tracks in the current frame, obtain associated parameters of each of the point tracks and the track data, where the associated parameters at least include a distance relationship, a speed relationship, and a track life between each of the point tracks and the track data; determining the association quality score based on the association parameter.
The position detection module 30 is further configured to obtain a first parameter, a second parameter, a third parameter, and a fourth parameter, where the first parameter is a probability that the target to be detected exists in a predicted target of a previous frame and a predicted target of a current frame; the second parameter is the probability that the target to be detected exists in the predicted target of the previous frame and does not exist in the predicted target of the current frame; the third parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and exists in the predicted target of the current frame; the fourth parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and does not exist in the predicted target of the current frame.
The position detecting module 30 is further configured to reduce the first parameter to a first preset value and increase the second parameter to a second preset value.
The calculation module 40 is further configured to obtain a radar detection probability; if the target to be detected is not shielded, acquiring the likelihood ratio of the target to be detected based on the radar detection probability; obtaining the second target confidence based on the likelihood ratio.
The calculation module 40 is further configured to, if the target to be detected is blocked, reduce the radar detection probability by a third preset value to obtain a corrected detection probability; and acquiring the likelihood ratio of the target to be detected based on the corrected detection probability.
The above modules may be functional modules or program modules, and may be implemented by software or hardware. For a module implemented by hardware, the modules may be located in the same processor; or the modules can be respectively positioned in different processors in any combination.
In this embodiment, a radar system is further provided, for implementing the target confidence obtaining method of the present application, where the radar system includes: the system comprises a detection radar and a control unit, wherein the detection radar is connected with the control unit; the detection radar is used for acquiring track information and track information of the target to be detected; the control unit is used for determining the shielded state of the target to be detected of the current frame based on the track data; if the target to be detected in the current frame is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the target to be detected is detected by a vehicle sensor at a corresponding moment, and the predicted nonexistence indicates that the target to be detected is not detected by the vehicle sensor at a corresponding moment; acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability; the control unit is further configured to implement the target confidence obtaining method of any of the above embodiments.
There is also provided in this embodiment an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.
Optionally, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:
and S1, acquiring the track data of the target to be detected.
S2, determining the shielded state of the target to be detected in the current frame based on the track data.
S3, if the object to be detected in the current frame is blocked, acquiring a first object confidence coefficient of the object to be detected in the previous frame and a change probability of a change of a predicted existence state of the object to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the object to be detected is detected by the vehicle sensor at the corresponding moment, and the predicted object nonexistence indicates that the object to be detected is not detected by the vehicle sensor at the corresponding moment.
And S4, acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability.
It should be noted that, for specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiments and optional implementations, and details are not described again in this embodiment.
In addition, in combination with the target confidence obtaining method provided in the foregoing embodiment, a storage medium may also be provided to implement in this embodiment. The storage medium having stored thereon a computer program; the computer program, when executed by a processor, implements any of the above-described embodiments of the target confidence obtaining method.
It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to be limiting. All other embodiments, which can be derived by a person skilled in the art from the examples provided herein without any inventive step, shall fall within the scope of protection of the present application.
It is obvious that the drawings are only examples or embodiments of the present application, and it is obvious to those skilled in the art that the present application can be applied to other similar cases according to the drawings without creative efforts. Moreover, it should be appreciated that such a development effort might be complex and lengthy, but would nevertheless be a routine undertaking of design, fabrication, and manufacture for those of ordinary skill having the benefit of this disclosure, and is not intended to limit the present disclosure to the particular forms disclosed herein.
The term "embodiment" is used herein to mean that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly or implicitly understood by one of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the patent protection. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims (10)

1. A target confidence obtaining method is characterized by comprising the following steps:
acquiring track data of a target to be detected;
determining the shielded state of the target to be detected of the current frame based on the track data;
if the target to be detected in the current frame is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the target to be detected is detected by a vehicle sensor at a corresponding moment, and the predicted nonexistence indicates that the target to be detected is not detected by the vehicle sensor at a corresponding moment;
and acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability.
2. The target confidence coefficient acquisition method according to claim 1, wherein the acquiring the track data of the target to be detected comprises:
detecting whether the point track exists in the current frame of the target to be detected based on the track data;
if the trace points exist, acquiring the correlation quality scores of the flight path data and the trace points of the target to be detected in the current frame;
and acquiring the second target confidence of the target to be detected in the current frame based on the associated quality score.
3. The method for obtaining the confidence of the target according to claim 2, wherein the obtaining the associated quality score of the track data and the point track of the target to be detected in the current frame comprises:
if the target to be detected has a plurality of point tracks in the current frame, acquiring association parameters of each point track and the track data, wherein the association parameters at least comprise the distance relationship, the speed relationship and the track service life between each point track and the track data;
determining the association quality score based on the association parameters.
4. The target confidence acquisition method according to claim 1, wherein the change probability includes a first parameter, a second parameter, a third parameter, and a fourth parameter;
the first parameter is the probability that the target to be detected exists in the predicted target of the previous frame and the predicted target of the current frame;
the second parameter is the probability that the target to be detected exists in the predicted target of the previous frame and does not exist in the predicted target of the current frame;
the third parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and exists in the predicted target of the current frame;
the fourth parameter is the probability that the target to be detected does not exist in the predicted target of the previous frame and does not exist in the predicted target of the current frame.
5. The method for obtaining the confidence of the target according to claim 4, wherein if the target to be detected in the current frame is occluded, the method further comprises the following steps of obtaining the confidence of the first target of the target to be detected in the previous frame, and obtaining the change probability of the predicted existence state of the target to be detected from the previous frame to the current frame, wherein the change probability is changed:
and reducing the first parameter to a first preset value and increasing the second parameter to a second preset value.
6. The target confidence obtaining method according to claim 1, wherein after determining the occluded state of the target to be detected in the current frame based on the track data, the method further comprises:
acquiring radar detection probability;
if the target to be detected is not shielded, acquiring the likelihood ratio of the target to be detected based on the radar detection probability;
obtaining the second target confidence based on the likelihood ratio.
7. The target confidence acquisition method of claim 6, further comprising, after the acquiring the radar detection probability:
if the target to be detected is blocked, reducing the radar detection probability by a third preset value to obtain a corrected detection probability;
and acquiring the likelihood ratio of the target to be detected based on the corrected detection probability.
8. An object confidence obtaining apparatus, comprising:
the acquisition module is used for acquiring track data of the target to be detected;
the occlusion detection module is used for determining the occluded state of the target to be detected of the current frame based on the track data;
the position detection module is used for acquiring a first target confidence coefficient of a target to be detected in a previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the vehicle sensor detects the target to be detected at a corresponding moment, and the predicted object nonexistence indicates that the vehicle sensor does not detect the target to be detected at the corresponding moment;
and the calculation module is used for acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability.
9. A radar system, comprising: the system comprises a detection radar and a control unit, wherein the detection radar is connected with the control unit;
the detection radar is used for acquiring track information and track information of the target to be detected;
the control unit is used for determining the shielded state of the target to be detected of the current frame based on the track data; if the target to be detected in the current frame is blocked, acquiring a first target confidence coefficient of the target to be detected in the previous frame and a change probability of a predicted existence state of the target to be detected from the previous frame to the current frame, wherein the predicted existence state comprises a predicted object existence and a predicted object nonexistence, the predicted existence indicates that the target to be detected is detected by a vehicle sensor at a corresponding moment, and the predicted nonexistence indicates that the target to be detected is not detected by the vehicle sensor at a corresponding moment; acquiring a second target confidence coefficient of the target to be detected in the current frame based on the first target confidence coefficient and the change probability;
the control unit is further configured to perform the target confidence acquisition method of any of claims 1 to 7.
10. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and the processor is configured to execute the computer program to perform the target confidence acquisition method of any of claims 1 to 7.
CN202111186684.5A 2021-10-12 2021-10-12 Target confidence coefficient acquisition method and device, radar system and electronic device Active CN114063079B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111186684.5A CN114063079B (en) 2021-10-12 2021-10-12 Target confidence coefficient acquisition method and device, radar system and electronic device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111186684.5A CN114063079B (en) 2021-10-12 2021-10-12 Target confidence coefficient acquisition method and device, radar system and electronic device

Publications (2)

Publication Number Publication Date
CN114063079A CN114063079A (en) 2022-02-18
CN114063079B true CN114063079B (en) 2022-06-21

Family

ID=80234293

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111186684.5A Active CN114063079B (en) 2021-10-12 2021-10-12 Target confidence coefficient acquisition method and device, radar system and electronic device

Country Status (1)

Country Link
CN (1) CN114063079B (en)

Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105809714A (en) * 2016-03-07 2016-07-27 广东顺德中山大学卡内基梅隆大学国际联合研究院 Track confidence coefficient based multi-object tracking method
CN106842142A (en) * 2015-12-03 2017-06-13 中国航空工业集团公司雷华电子技术研究所 Surface search radar high-volume point mark extraction algorithm based on sub-district treatment
CN108537822A (en) * 2017-12-29 2018-09-14 西安电子科技大学 Motion target tracking method based on weighting reliability estimating
CN109285180A (en) * 2018-08-31 2019-01-29 电子科技大学 A kind of road vehicle tracking of 3D
CN110018489A (en) * 2019-04-25 2019-07-16 上海蔚来汽车有限公司 Target tracking method, device and controller and storage medium based on laser radar
CN110222764A (en) * 2019-06-10 2019-09-10 中南民族大学 Shelter target detection method, system, equipment and storage medium
CN110363790A (en) * 2018-04-11 2019-10-22 北京京东尚科信息技术有限公司 Target tracking method, device and computer readable storage medium
CN110626355A (en) * 2019-09-17 2019-12-31 浙江吉利汽车研究院有限公司 Radar blocking diagnosis method and device and terminal
CN111127513A (en) * 2019-12-02 2020-05-08 北京交通大学 Multi-target tracking method
CN112198503A (en) * 2020-10-16 2021-01-08 无锡威孚高科技集团股份有限公司 Target track prediction optimization method and device and radar system
US10928820B1 (en) * 2019-11-12 2021-02-23 Baidu Usa Llc Confidence levels along the same predicted trajectory of an obstacle
CN112578345A (en) * 2020-11-20 2021-03-30 福瑞泰克智能系统有限公司 Radar blocking detection method, device, equipment and storage medium
WO2021072696A1 (en) * 2019-10-17 2021-04-22 深圳市大疆创新科技有限公司 Target detection and tracking method and system, and movable platform, camera and medium
CN112896157A (en) * 2021-02-08 2021-06-04 福瑞泰克智能系统有限公司 Defensive driving control method, device and system, vehicle-mounted terminal and storage medium
CN113256686A (en) * 2021-06-28 2021-08-13 上海齐感电子信息科技有限公司 System and method for tracking accurate visual target
CN113420682A (en) * 2021-06-28 2021-09-21 阿波罗智联(北京)科技有限公司 Target detection method and device in vehicle-road cooperation and road side equipment

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2974031A1 (en) * 2016-07-22 2018-01-22 Appropolis Inc. Method of tracking one or more mobile objects in a site and a system employing same
US11501572B2 (en) * 2018-03-26 2022-11-15 Nvidia Corporation Object behavior anomaly detection using neural networks

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106842142A (en) * 2015-12-03 2017-06-13 中国航空工业集团公司雷华电子技术研究所 Surface search radar high-volume point mark extraction algorithm based on sub-district treatment
CN105809714A (en) * 2016-03-07 2016-07-27 广东顺德中山大学卡内基梅隆大学国际联合研究院 Track confidence coefficient based multi-object tracking method
CN108537822A (en) * 2017-12-29 2018-09-14 西安电子科技大学 Motion target tracking method based on weighting reliability estimating
CN110363790A (en) * 2018-04-11 2019-10-22 北京京东尚科信息技术有限公司 Target tracking method, device and computer readable storage medium
CN109285180A (en) * 2018-08-31 2019-01-29 电子科技大学 A kind of road vehicle tracking of 3D
CN110018489A (en) * 2019-04-25 2019-07-16 上海蔚来汽车有限公司 Target tracking method, device and controller and storage medium based on laser radar
CN110222764A (en) * 2019-06-10 2019-09-10 中南民族大学 Shelter target detection method, system, equipment and storage medium
WO2020248957A1 (en) * 2019-06-10 2020-12-17 中南民族大学 Method, system and apparatus for detecting occluded target object, and storage medium
CN110626355A (en) * 2019-09-17 2019-12-31 浙江吉利汽车研究院有限公司 Radar blocking diagnosis method and device and terminal
WO2021072696A1 (en) * 2019-10-17 2021-04-22 深圳市大疆创新科技有限公司 Target detection and tracking method and system, and movable platform, camera and medium
US10928820B1 (en) * 2019-11-12 2021-02-23 Baidu Usa Llc Confidence levels along the same predicted trajectory of an obstacle
CN111127513A (en) * 2019-12-02 2020-05-08 北京交通大学 Multi-target tracking method
CN112198503A (en) * 2020-10-16 2021-01-08 无锡威孚高科技集团股份有限公司 Target track prediction optimization method and device and radar system
CN112578345A (en) * 2020-11-20 2021-03-30 福瑞泰克智能系统有限公司 Radar blocking detection method, device, equipment and storage medium
CN112896157A (en) * 2021-02-08 2021-06-04 福瑞泰克智能系统有限公司 Defensive driving control method, device and system, vehicle-mounted terminal and storage medium
CN113256686A (en) * 2021-06-28 2021-08-13 上海齐感电子信息科技有限公司 System and method for tracking accurate visual target
CN113420682A (en) * 2021-06-28 2021-09-21 阿波罗智联(北京)科技有限公司 Target detection method and device in vehicle-road cooperation and road side equipment

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
基于多传感器融合的智能车道路障碍物检测;娄新雨;《中国优秀硕士学位论文全文数据库工程科技II辑》;20210215(第2期);全文 *
基于深度特征与抗遮挡策略的运动目标跟踪;火元莲;《西北师范大学学报(自然科学版)》;20200715(第4期);全文 *
复杂环境下基于目标存在概率的多传感器目标跟踪算法;洪帅鑫;《中国优秀硕士学位论文全文数据库信息科技辑》;20200115(第1期);全文 *

Also Published As

Publication number Publication date
CN114063079A (en) 2022-02-18

Similar Documents

Publication Publication Date Title
US9500748B2 (en) Target recognition apparatus
EP3671272A1 (en) Vehicle sensor fusion based on fuzzy sets
CN111247525A (en) Lane detection method and device, lane detection equipment and mobile platform
WO2010134428A1 (en) Vehicular environment estimation device
CN109738884B (en) Object detection method and device and computer equipment
CN112580571A (en) Vehicle running control method and device and electronic equipment
CN109684944B (en) Obstacle detection method, obstacle detection device, computer device, and storage medium
CN111580109A (en) Vehicle-mounted millimeter wave radar shielding detection method based on amplitude characteristic statistics
CN115131748B (en) Method and system for improving target tracking and identifying accuracy rate of radar and vision all-in-one machine
JP2021033510A (en) Driving assistance device
CN112598715A (en) Multi-sensor-based multi-target tracking method, system and computer readable medium
CN114475593B (en) Travel track prediction method, vehicle, and computer-readable storage medium
CN109263649B (en) Vehicle, object recognition method and object recognition system thereof in automatic driving mode
US20220108552A1 (en) Method and Apparatus for Determining Drivable Region Information
CN114063079B (en) Target confidence coefficient acquisition method and device, radar system and electronic device
CN114296095A (en) Method, device, vehicle and medium for extracting effective target of automatic driving vehicle
US11841419B2 (en) Stationary and moving object recognition apparatus
CN116774228A (en) Determination method and device for drivable area, electronic device and storage medium
CN115494494A (en) Multi-sensor target fusion method and device, electronic equipment and storage medium
JP4644590B2 (en) Peripheral vehicle position detection device and peripheral vehicle position detection method
CN115951344A (en) Data fusion method and device for radar and camera, electronic equipment and storage medium
CN115294172A (en) Target detection method and device, electronic equipment and storage medium
CN114120699B (en) Vehicle detection method, device and medium based on sliding pane
EP4131946A1 (en) Object tracking device, object tracking method, and recording medium
CN117741614A (en) Radar signal processing method, processing device, terminal and storage medium

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