CN111361570B

CN111361570B - Multi-target tracking reverse verification method and storage medium

Info

Publication number: CN111361570B
Application number: CN202010157046.XA
Authority: CN
Inventors: 李海鹏; 陈文强; 许立杰; 黄旭
Original assignee: Fujian Hante Cloud Intelligent Technology Co ltd
Current assignee: Fujian Hante Cloud Intelligent Technology Co ltd
Priority date: 2020-03-09
Filing date: 2020-03-09
Publication date: 2021-06-18
Anticipated expiration: 2040-03-09
Also published as: CN111361570A

Abstract

The invention relates to a multi-target tracking reverse verification method, which comprises the following steps: acquiring target data of a target vehicle through a sensor, wherein the target data comprises target position information and target speed information; performing first target matching according to the acquired target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment to obtain the current vehicle state and the vehicle state of the target vehicle at the previous moment; performing direction quick verification according to the obtained current vehicle state of the target vehicle and the vehicle state at the last moment, and judging whether the state transition of the target vehicle is reasonable or not; if the current time is reasonable, updating the target data of the target vehicle according to the acquired target data at the current time; and if the target data is not reasonable, performing secondary target matching according to the target data of the target vehicle at the previous moment. When the state transition of the target vehicle is suddenly changed, the target vehicle is checked again, so that the probability of error in tracking matching is reduced, and secondary matching of each target is avoided.

Description

Multi-target tracking reverse verification method and storage medium

Technical Field

The invention relates to the technical field of intelligent automobiles, in particular to a multi-target tracking reverse verification method and a storage medium.

Background

Today, ADAS (Advanced Driving assistance System) has become a major hot spot in the development of technology, and both host plants and suppliers devote a great deal of effort to them. And in terms of vehicles, the 'finished vehicle' with the ADAS installed is greatly improved in terms of both the operation stability and the safety, and even some regions have included the ADAS part of functions in the 'mandatory installation' regulation. The ADAS technology can be roughly divided into three major modules, namely "perception", "decision" and "control": the 'perception system' is mainly responsible for acquiring external environment information data, transmitting the external environment information data to the 'decision-making system' for analysis and processing, then forming a control signal to be input into the 'control system', and the controlled information is captured by the 'perception system' again to finally form a 'perception-decision-control-perception' closed-loop system. The accuracy and comprehensiveness of data output by the sensing system fundamentally influence decision making and control, and further determine the quality of products.

In the perception system, the multi-target tracking algorithm is a large research hotspot. The current sophisticated trace matching algorithm mainly includes GNN (nearest neighbor matching), JPDA (joint probability density association), MHT (multi-hypothesis tracing), and so on. The above mentioned algorithms are all "forward matching tracking", that is, the "most likely" matching object is calculated according to some established mechanism and the current "estimated data" and "observed data", so as to update the target state. If the algorithm is singly used, matching errors are likely to occur, and then phenomena such as false alarm, false detection, missing detection and the like occur.

Disclosure of Invention

Therefore, a multi-target tracking reverse verification method and a storage medium are needed to be provided, and the problems that matching errors may occur in the existing multi-target tracking algorithm, and further false alarms, false detections, missed detections and the like occur are solved.

In order to achieve the above object, the inventor provides a multi-target tracking reverse verification method, which comprises the following steps:

acquiring target data of a target vehicle through a sensor, wherein the target data comprises target position information and target speed information;

performing first target matching according to the acquired target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment to obtain the current vehicle state and the vehicle state of the target vehicle at the previous moment;

carrying out reverse quick verification according to the obtained current vehicle state of the target vehicle and the vehicle state at the last moment, and judging whether the state transition of the target vehicle is reasonable or not;

if the current time is reasonable, updating the target data of the target vehicle according to the acquired target data at the current time;

if the target data is not reasonable, performing second target matching according to the target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment; and different target matching algorithms are adopted for the first target matching and the second target matching.

Further preferably, the sensor is a laser radar, a millimeter wave radar, an ultrasonic radar or a visual camera.

Further optimization, the step of performing first target matching according to the acquired target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment to obtain the current vehicle state and the vehicle state of the target vehicle at the previous moment specifically comprises the following steps:

determining the relative position of the target vehicle relative to the current vehicle according to the target position information of the target vehicle, wherein the relative position comprises the front, the rear, the left side, the right side, the left front side, the right front side, the left rear side and the right rear side of the current vehicle;

determining the running states of the target vehicle according to the relative position of the target vehicle relative to the current vehicle and the target speed information of the target vehicle, wherein the running states comprise static state, parallel state, approaching state, far away state, overtaking state and conceding state;

and obtaining the vehicle state of the target vehicle according to the obtained relative position and the running state of the target vehicle.

Further optimization, the "determining the relative position of the target vehicle with respect to the current vehicle according to the target position information of the target vehicle" specifically includes:

when the longitudinal distance of the target vehicle is greater than the preset forward distance of the current vehicle and the transverse distance of the target vehicle is between the preset right distance and the preset left distance of the current vehicle, determining that the relative position of the target vehicle is the front;

when the longitudinal distance of the target vehicle is smaller than the preset rear distance of the current vehicle and the transverse distance of the target vehicle is between the preset right distance and the preset left distance of the current vehicle, determining that the relative position of the target vehicle is the rear;

when the transverse distance of the target vehicle is smaller than the preset left distance of the current vehicle and the longitudinal distance of the target vehicle is between the preset forward distance and the preset rear distance of the current vehicle, determining that the relative position of the target vehicle is the left side;

when the transverse distance of the target vehicle is greater than the preset right distance of the current vehicle and the longitudinal distance of the target vehicle is between the preset forward distance and the preset rear distance of the current vehicle, determining that the relative position of the target vehicle is the right side;

when the longitudinal distance of the target vehicle is greater than the preset forward distance of the current vehicle, and the transverse distance of the target vehicle is less than the preset left distance of the current vehicle, determining that the relative position of the target vehicle is the left front side;

when the longitudinal distance of the target vehicle is greater than the preset forward distance of the current vehicle, and the transverse distance of the target vehicle is greater than the preset right distance of the current vehicle, determining that the relative position of the target vehicle is the right front side;

when the longitudinal distance of the target vehicle is smaller than the preset rear distance of the current vehicle and the transverse distance of the target vehicle is smaller than the preset left distance of the current vehicle, determining that the relative position of the target vehicle is the left rear side;

and when the longitudinal distance of the target vehicle is smaller than the preset rear distance of the current vehicle and the transverse distance of the target vehicle is larger than the preset right distance of the current vehicle, determining that the relative position of the target vehicle is the right rear side.

Further optimization, the "determining the operating state of the target vehicle according to the relative position of the target vehicle with respect to the current vehicle and the target speed information of the target vehicle" specifically includes:

when the absolute value of the vehicle longitudinal speed of the target vehicle is smaller than a preset static value, determining that the running state of the target vehicle is static;

when the relative speed of the longitudinal speed of the target vehicle and the current speed of the current vehicle is smaller than a preset parallel value, determining that the running state of the target vehicle is parallel;

when the relative position of the target vehicle is the front side, the left front side or the right front side, judging that the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is greater than a first preset distance value, or when the relative position of the target vehicle is the rear side, the left rear side and the right rear side, judging that the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is less than a second preset distance value, and judging that the running state of the target vehicle is away;

when the relative position of the target vehicle is the front, the left front side or the right front side, and the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is less than a first preset approach value, or when the relative position of the target vehicle is the rear, the left rear side and the right rear side, and the longitudinal speed of the target vehicle is greater than a second preset approach value, determining that the running state of the target vehicle is approaching;

when the relative position of the target vehicle is on the left side or the right side, and when the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is smaller than a preset back-off value, determining that the running state of the target vehicle is back-off;

and when the relative position of the target vehicle is on the left side or the right side, and when the relative speed of the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is greater than a preset overtaking value, determining that the running state of the target vehicle is overtaking.

The inventor also provides another technical scheme that: a storage medium storing a computer program which, when executed by a processor, performs the steps of:

Further optimization, the step of performing first target matching according to the obtained target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment to obtain the current vehicle state of the target vehicle and the vehicle state of the target vehicle at the previous moment specifically comprises the following steps:

Different from the prior art, according to the technical scheme, after the ADAS system of the vehicle carries out primary forward matching tracking on the target vehicle, secondary reverse verification is carried out again, and according to the continuity of the running state of the vehicle, when the state transition of the vehicle state of the target vehicle changes suddenly, the target vehicle is checked again, so that the probability of errors in tracking matching is reduced, computing resources are reduced, and secondary matching on each target is avoided.

Drawings

FIG. 1 is a schematic flow chart of a multi-target tracking reverse verification method according to an embodiment;

FIG. 2 is a schematic diagram of coordinates of a state partition of the subject vehicle according to an embodiment;

fig. 3 is a schematic structural diagram of a storage medium according to an embodiment.

Description of reference numerals:

310. a storage medium.

Detailed Description

To explain technical contents, structural features, and objects and effects of the technical solutions in detail, the following detailed description is given with reference to the accompanying drawings in conjunction with the embodiments.

Referring to fig. 1, the present embodiment provides a multi-target tracking reverse verification method, including the following steps:

step S110: acquiring target data of a target vehicle through a sensor, wherein the target data comprises target position information and target speed information;

step S120: performing first target matching according to the acquired target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment to obtain the current vehicle state and the vehicle state of the target vehicle at the previous moment;

step S130: carrying out reverse quick verification according to the obtained current vehicle state of the target vehicle and the vehicle state at the last moment, and judging whether the state transition of the target vehicle is reasonable or not;

if it is reasonable, execute step 141: updating target data of the target vehicle according to the acquired target data at the current moment;

if not, go to step 142: and performing secondary target matching according to the target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment.

Whether the vehicle is in a "stationary" or "running" state, the change in state of the vehicle is a "continuous" and "non-leaping" process. That is, a vehicle traveling forward at a high speed must go through a "deceleration state" and a "stop state" before entering a "reverse state", and none of the states can be passed through. The method comprises the steps that a current vehicle obtains target data of a target vehicle in real time through a sensor, after the target data of the target vehicle are obtained, the current target data of the target vehicle and the target data of the last moment are subjected to first target matching, the current vehicle state of the target vehicle and the vehicle state of the last moment are obtained, whether state transition of the current vehicle state of the target and the vehicle state of the last moment is reasonable or not is judged through reverse quick verification, namely whether the state mutation condition occurs or not is judged, if the state mutation does not occur, the state transition of the target vehicle is considered to be reasonable, if the state mutation occurs, the state transition of the target vehicle is considered to be unreasonable, and the target vehicle is subjected to second target matching, so that the probability of error occurrence of tracking matching is reduced, computing resources are reduced, and secondary matching of each target is. In order to cross-verify the correctness of the target matching, different target matching algorithms are adopted for the first target matching and the second target matching, and the adopted target matching algorithms can include JPDA (joint probabilistic data association), MHT (multiple hypothesis tracking), GNN (global nearest neighbor data association algorithm) and the like, so that the validity of the cross-verification is ensured.

In this embodiment, use vehicle barycenter or be close to the installation position of barycenter place for installing the sensor, be 0 ~ 360 in order to satisfy the sensor perception scope that the vehicle was installed, the sensor is lidar, millimeter wave radar, ultrasonic radar or vision camera. The type of the sensor is not mandatory, when the sensor is one of the target detection type sensors, the sensing range is 0-360 degrees, or the sensing range is 0-360 degrees through the combination of a plurality of sensors.

In this embodiment, the "performing the first target matching according to the obtained target data of the target vehicle at the current time and the target data of the target vehicle at the previous time to obtain the current vehicle state and the vehicle state of the target vehicle at the previous time" specifically includes the following steps:

The target data acquired by the sensor comprises target position information and target speed information, wherein the target position information comprises longitudinal position information and transverse position information, and the target speed information comprises longitudinal speed information and transverse speed information. Dividing the relative position of the target vehicle into a front side, a rear side, a left side, a right side, a left front side, a right front side, a left rear side and a right rear side according to the acquired target position information of the target vehicle; and determining the running state of the target vehicle according to the relative position and the target speed information of the target vehicle, further determining the vehicle state of the target vehicle, and further performing reverse quick verification according to the current vehicle state and the last vehicle state of the target vehicle after the target speed of the target vehicle is obtained, so as to judge whether the state transition of the target vehicle is reasonable.

Specifically, the "determining the relative position of the target vehicle with respect to the current vehicle according to the target position information of the target vehicle" specifically includes:

In a coordinate diagram of the division of the target vehicle state shown in fig. 2, a coordinate system is established with a geometric center of a plane of the current vehicle as a circular point, an X axis represents a forward direction of the current vehicle, a Y axis represents a transverse direction of the current vehicle, Dist _ L/Dist _ Right represents two straight lines parallel to the forward direction of the vehicle, which are respectively a preset left distance and a preset Right distance of the current vehicle, wherein Dist _ L is located on a left side of the current vehicle, and Dist _ Right is located on a Right side of the current vehicle (Dist _ L, Dist _ Right can also be replaced by a lane line). And Dist _ Front/Dist _ Rear are two straight lines perpendicular to the advancing direction of the current vehicle, and are respectively a preset forward distance and a preset Rear distance of the current vehicle, wherein Dist _ Front is located in Front of the current vehicle, and Dist _ Rear is located behind the current vehicle.

When the longitudinal distance (Det _ X) of the target vehicle is greater than the preset forward distance Dist _ Front of the current vehicle, and the transverse distance (Det _ Y) of the target vehicle is between the preset Right distance Dist _ Right and the preset left distance Dist _ L of the current vehicle, the target vehicle is confirmed to be located in Front, and the specific formula is as follows:

when the longitudinal distance of the target vehicle is smaller than the preset Rear distance Dist _ Rear of the current vehicle, and the transverse distance of the target vehicle is between the preset Right distance Dist _ Right and the preset left distance Dist _ L of the current vehicle, the target vehicle is confirmed to be positioned behind, and the specific formula is as follows:

when the transverse distance of the target vehicle is smaller than the preset left distance Dist _ L of the current vehicle, and the longitudinal distance of the target vehicle is between the preset forward distance Dist _ Front and the preset Rear distance Dist _ Rear of the current vehicle, the target vehicle is confirmed to be located on the left side, and the specific formula is as follows:

when the transverse distance of the target vehicle is greater than the preset Right distance Dist _ Right of the current vehicle, and the longitudinal distance of the target vehicle is between the preset forward distance Dist _ Front and the preset Rear distance Dist _ Rear of the current vehicle, the target vehicle is confirmed to be positioned on the Right side, and the specific formula is as follows:

when the longitudinal distance of the target vehicle is greater than the preset forward distance Dist _ Front of the current vehicle and the transverse distance is less than the preset Left distance Dist _ Left of the current vehicle, determining that the target vehicle is positioned on the Front Left side, wherein the specific formula is as follows:

when the longitudinal distance of the target vehicle is greater than the preset forward distance Dist _ Front of the current vehicle and the transverse distance is greater than the preset Right distance Dist _ Right of the current vehicle, the target vehicle is confirmed to be positioned on the Right Front side, and the specific formula is as follows:

when the longitudinal distance of the target vehicle is smaller than the preset Rear distance Dist _ Rear of the current vehicle and the transverse distance is smaller than the preset Left distance Dist _ Left of the current vehicle, the target vehicle is defined to be positioned on the Left Rear side, and the specific formula is as follows:

when the longitudinal distance of the target vehicle is smaller than the preset Rear distance Dist _ Rear of the current vehicle and the transverse distance is larger than the preset Right distance Dist _ Right of the current vehicle, the target vehicle is confirmed to be positioned at the Right Rear side, and the specific formula is as follows:

specifically, the "determining the operating state of the target vehicle according to the relative position of the target vehicle with respect to the current vehicle and the target speed information of the target vehicle" specifically includes:

The vehicle can be divided into five states of 'still', 'far away', 'close', 'overtaking' and 'parallel' according to the running state of the target vehicle. The symbol Det _ Vx represents the longitudinal vehicle speed of the target vehicle, and Det _ Vy represents the lateral vehicle speed of the target vehicle. Meanwhile, the symbol Ego _ V represents the current vehicle's own vehicle speed, the following relationship exists:

1. when the target vehicle is in any area and the absolute value of the longitudinal speed of the target vehicle is smaller than a preset Static value, namely the absolute value of the absolute speed of the target vehicle is smaller than a certain Static value (V _ Static), the target vehicle is considered to be in a Static state, and the specific formula is as follows:

|Det_Vx|≤V_Static

2. when the target vehicle is in any one area, when the relative speed of the vehicle longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is smaller than a preset parallel value, namely when the absolute value of the relative vehicle speed of the target vehicle is smaller than a certain value (V _ Relate), the target vehicle is considered to be in a parallel state, and the specific formula is as follows:

|Det_Vx-Ego_V|≤V_Relate

3. when the target vehicle is in a front area, a left front area and a right front area, and the relative speed of the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is judged to be greater than a first preset distance value, namely the vehicle speed of the target vehicle is greater than the vehicle speed of the current vehicle by more than a certain range (V _ leave _ f), the target vehicle is considered to be in a distance state; meanwhile, when the target vehicle is in a rear area, a left rear area and a right rear area, and when the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is judged to be smaller than a second preset departing value, namely the vehicle speed of the target vehicle is smaller than the vehicle speed of the current vehicle by a certain range (V _ leave _ r) below, the target vehicle is also judged to be in a departing state, and the specific formula is as follows:

4. when the target vehicle is in a front area, a left front area and a right front area and the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is smaller than a first preset approach value, namely the vehicle speed of the target vehicle is smaller than the vehicle speed of the current vehicle within a certain range (V _ close _ f), the target vehicle is considered to be in an approach state; meanwhile, when the target vehicle is in a rear area, a left rear area and a right rear area, the longitudinal speed of the target vehicle is greater than a second preset approach value, and the vehicle speed of the target vehicle is greater than the vehicle speed of the current vehicle by a certain range (V _ close _ r), the target vehicle is also judged to be in an approach state, and the specific formula is as follows:

5. when the target vehicle is in a left area and a right area, and when the relative speed of the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is smaller than a preset yielding value, namely the vehicle speed of the target vehicle is smaller than the vehicle speed of the current vehicle within a certain range (V _ under), determining that the target vehicle is in a yielding state, wherein a specific formula is as follows;

Det_Vx-Ego_V＜V_under

6. when the target vehicle is in the left area and the right area, and when the relative speed between the longitudinal speed of the target vehicle and the current vehicle speed of the current vehicle is greater than a preset overtaking value, namely the vehicle speed of the target vehicle is greater than the current vehicle speed of the current vehicle by a certain range (V _ over), the target vehicle is judged to be in an overtaking state, and the specific formula is as follows:

Det_Vx-Ego_V＞V_over

according to the relative positions of the target vehicle including "front", "rear", "left side", "right side", "left front side", "right front side", "left rear side", "right rear side", and the running states of the target vehicle including "stationary", "approaching", "departing", "parallel", "passing", "yielding", the vehicle states of the target vehicle can be obtained including:

according to the vehicle state transition rule, when the vehicle state of the target vehicle is transitioned, the corresponding steps are necessarily obeyed, and the transition rule is continuously changed, and the specific transition rule is as follows:

when the original target vehicle state of the target vehicle is forward/stationary, the transition state is: front/stationary, front/near, left front/stationary, right front/stationary;

when the original target vehicle state of the target vehicle is rear/stationary, the transition state is as follows: back/still, back/far, left back/far, right back/far;

when the original target vehicle state of the target vehicle is left side/stationary, the transfer state is as follows: left/still, left/give-back, left back/still;

when the original target vehicle state of the target vehicle is right side/stationary, the transition state is: right side/still, right side/give-off, right rear side/still;

when the original target vehicle state of the target vehicle is left front side/stationary, the transition state is as follows: left anterior/stationary, left anterior/near, anterior/stationary;

when the original target vehicle state of the target vehicle is right front side/stationary, the transition state is: right anterior/stationary, right anterior/near, anterior/stationary;

when the original target vehicle state of the target vehicle is left rear side/stationary, the transition state is as follows: left posterior/static, left posterior/distant, posterior/static;

when the original target vehicle state of the target vehicle is right rear side/stationary, the transition state is as follows: right rear/still, right rear/far away, rear/still;

when the original target vehicle state of the target vehicle is forward/parallel, the transfer state is as follows: front/parallel, front/close, front/far, left front/parallel, right front/parallel;

when the original target vehicle state of the target vehicle is rear/parallel, the transfer state is as follows: rear/parallel, rear/close, rear/far, left rear/parallel, right rear/parallel;

when the original target vehicle state of the target vehicle is left side/parallel, the transfer state is as follows: left/parallel, left/overtaking, left/yielding;

when the original target vehicle state of the target vehicle is right side/parallel, the transfer state is as follows: right side/parallel, right side/overtaking, right side/yielding;

when the original target vehicle state of the target vehicle is left front side/parallel, the transfer state is as follows: left anterior/parallel, left anterior/near, left anterior/far, front/parallel;

when the original target vehicle state of the target vehicle is right front side/parallel, the transition state is as follows: right anterior/parallel, right anterior/proximal, right anterior/distal, anterior/parallel;

when the original target vehicle state of the target vehicle is left rear side/parallel, the transfer state is as follows: left rear/parallel, left rear/close, left rear/far, rear/parallel;

when the original target vehicle state of the target vehicle is the right rear side/parallel, the transition state is as follows: right rear side/parallel, right rear side/close, right rear side/far away, rear/parallel;

when the original target vehicle state of the target vehicle is forward/close, the transition state is as follows: front/near, front/parallel, left front/near, right front/near;

when the original target vehicle state of the target vehicle is rear/approaching, the transition state is as follows: rear/near, rear/parallel, left rear/near, right rear/near;

when the original target vehicle state of the target vehicle is left front side/close, the transition state is as follows: left anterior/near, left anterior/parallel, left/give-off, front/near;

when the original target vehicle state of the target vehicle is right front side/approaching, the transition state is: right anterior/proximal, right anterior/parallel, right side/giving-off, anterior/proximal;

when the original target vehicle state of the target vehicle is left rear side/close, the transition state is as follows: left rear/approach, left rear/parallel, left side/overtaking, rear/approach;

when the original target vehicle state of the target vehicle is right rear side/approaching, the transition state is as follows: right rear side/approach, right side/overtaking, rear/approach;

when the original target vehicle state of the target vehicle is forward/far, the transition state is as follows: forward/far, forward/parallel, left front/far, right front/far;

when the original target vehicle state of the target vehicle is rear/far, the transition state is as follows: rear/far away, rear/parallel, left rear/far away, right rear/far away;

when the original target vehicle state of the target vehicle is left front side/far away, the transfer state is as follows: left anterior/distal, left anterior/parallel, anterior/distal;

when the original target vehicle state of the target vehicle is right front side/far away, the transition state is as follows: right anterior/distal, right anterior/parallel, anterior/distal;

when the original target vehicle state of the target vehicle is left rear side/far away, the transfer state is as follows: left rear/far, left rear/parallel, rear/far;

when the original target vehicle state of the target vehicle is right rear side/far away, the transfer state is as follows: right rear side/away, right rear side/parallel, rear/away;

when the original target vehicle state of the target vehicle is left side/overtaking, the transfer state is as follows: left side/overtaking, left side/parallel, left front side/far away;

when the original target vehicle state of the target vehicle is right side/overtaking, the transfer state is as follows: right side/overtaking, right side/parallel, right front side/far away;

when the original target vehicle state of the target vehicle is left side/yielding, the transfer state is as follows: left/yielding, left/parallel, left rear/far away;

when the original target vehicle state of the target vehicle is right side/yielding, the transfer state is as follows: right/give, right/parallel, right back/far.

The vehicle state of the target vehicle at the last moment can only be transferred to a corresponding state, and when the target vehicle has unreasonable state change, the target vehicle is matched with two adjacent frames at a high probability to generate errors, so that whether the matching and tracking are correct or not is verified quickly in a reverse mode, computing resources are saved, and secondary matching of each target is avoided.

Referring to fig. 3, in another embodiment, a storage medium 310 stores a computer program, which when executed by a processor performs the steps of:

and if the target data is not reasonable, performing secondary target matching according to the target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment.

In this embodiment, the sensor is a laser radar, a millimeter wave radar, an ultrasonic radar, or a vision camera.

In this embodiment, the step of "performing the first target matching according to the obtained target data of the target vehicle at the current time and the target data of the target vehicle at the previous time to obtain the current vehicle state and the vehicle state of the target vehicle at the previous time" specifically includes the following steps:

In this embodiment, the "determining the relative position of the target vehicle with respect to the current vehicle according to the target position information of the target vehicle" specifically includes:

In this embodiment, the "determining the operating state of the target vehicle according to the relative position of the target vehicle with respect to the current vehicle and the target speed information of the target vehicle" specifically includes:

It should be noted that, although the above embodiments have been described herein, the invention is not limited thereto. Therefore, based on the innovative concepts of the present invention, the technical solutions of the present invention can be directly or indirectly applied to other related technical fields by making changes and modifications to the embodiments described herein, or by using equivalent structures or equivalent processes performed in the content of the present specification and the attached drawings, which are included in the scope of the present invention.

Claims

1. A multi-target tracking reverse verification method is characterized by comprising the following steps:

2. The multi-target tracking reverse validation method according to claim 1, wherein the sensor is a laser radar, a millimeter wave radar, an ultrasonic radar, or a vision camera.

3. The multi-target tracking reverse verification method according to claim 1, wherein the step of performing the first target matching according to the obtained target data of the target vehicle at the current moment and the target data of the target vehicle at the previous moment to obtain the current vehicle state and the vehicle state of the target vehicle at the previous moment specifically comprises the following steps:

4. The multi-target tracking reverse verification method according to claim 3, wherein the "determining the relative position of the target vehicle with respect to the current vehicle according to the target position information of the target vehicle" specifically comprises:

5. The multi-target tracking reverse verification method according to claim 3, wherein the determining the operating state of the target vehicle according to the relative position of the target vehicle with respect to the current vehicle and the target speed information of the target vehicle specifically comprises:

6. A storage medium storing a computer program, the computer program when executed by a processor performing the steps of:

7. The storage medium of claim 6, wherein the sensor is a lidar, a millimeter wave radar, an ultrasonic radar, or a visual camera.

8. The storage medium according to claim 6, wherein the step of performing the first target matching according to the acquired target data of the target vehicle at the current time and the target data of the target vehicle at the previous time to obtain the current vehicle state and the vehicle state of the target vehicle at the previous time specifically comprises the steps of:

9. The storage medium according to claim 8, wherein the "determining the relative position of the target vehicle with respect to the current vehicle based on the target position information of the target vehicle" specifically includes:

10. The storage medium according to claim 8, wherein the "determining the operating state of the target vehicle according to the relative position of the target vehicle with respect to the current vehicle and the target speed information of the target vehicle" specifically comprises: